> Technically, when someone does a Google search for “Williams Sonoma Cast Iron Skillet,” they probably would have clicked on one of the first 10 organic results, EVERY ONE OF WHICH leads to their website. But, y’know what ol’ Billy Ma’s performance marketers couldn’t then do: prove their value to their bosses.
> [picture of that search term and williams sonoma ads with shopping links]
The main problem here is that if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements.
The spend is necessary in a defensive way because Google creates a bidding war even for the hyper relevant.
edit: I just checked and if you search "williams sonoma skillet", if WS was not paying for [green] then the very first "result" (ad) would be Food52 [red] https://imgur.com/a/9Nnxs6h
I just tried "airbnb paris" and the first result is, somewhat predictably, an ad that is not airbnb. But the second one is also an ad, this time from airbnb. So they clearly didn't keep their spend dialed down to zero, and are aware of the need to advertise on their own keyword.
Forget about the corporate POV for a second, as a user, if I'm searching for X, I probably want X in my results, no?
It's the fundamental contradiction in Google's search ads model. If Google delivers users the thing they want, ads by definition have to be things users don't want.
Advertisements are fundamentally not what search service users want regardless of searching for a specific brand or not. (Its what advertisers want others to want)
If a search engine encounters the query "Williams Sonoma 12 inch skillet", retail outlets selling it should organically appear in results without the need for ads.
If it’s $39.99 on W-S.com, I definitely want to see the place that’s willing to sell the same pan for $34.99, the EBay listing for $32.50, and the one selling a competing similar one for $29.99 even if those sites don’t have the organic search SEO juice to land in the top 10.
As a Google search user, why wouldn’t I value these? If Google doesn’t serve them and another search engine does, I’d be inclined to switch to the other one.
> If it’s $39.99 on W-S.com, I definitely want to see the place that’s willing to sell the same pan for $34.99, the EBay listing for $32.50, and the one selling a competing similar one for $29.99 even if those sites don’t have the organic search SEO juice to land in the top 10.
But why would those sites not have the "organic search SEO juice"? If those sites are actually good places to buy that kind of thing, a good search engine should direct a user searching for that thing there.
I think it's fairly clear that Amazon and Ebay would show up as the top organic results.
What wouldn't show up is lmm-cookware.com, launched early October 2021, who is trying to establish themselves as a new destination for cooking enthusiasts. That domain is new, the business is new, no Google reviews, but they've got the pan in stock, are willing to sell it for $30 shipped, and have an advertising budget to reach customers.
As a consumer, I want to be able to learn about that offer, make my own decision where to buy the pan, and I'm happy to use a search engine who will show me that offer. lmm-cookware, I, and the search engine all win from this outcome. I don't care whether lmm-cookware has been a good place to buy pans such as these for the last 180 days; I care whether they're a good place to buy it right now.
> I think it's fairly clear that Amazon and Ebay would show up as the top organic results.
Why? How websites are ranked "organically" is somethine that is up to the search engine. A user-focused search engine would have no problem including factors like price (for shopping sites) that the user cares about. An even better one might provide the user with options to refine the ranking criteria for each search.
Meanwhile you keep assuming that somehow the site willing to give you the best deal will also be the one winning the advertising bid. That makes no sense as they are also the site spending the most on advertisement which they have to recoup somehow.
> As a consumer, I want to be able to learn about that offer, make my own decision where to buy the pan.
You seem to be under the impression that you would be missing out on an offer without ads. But if you only look at a finite set of links you are always missing some offers - ads only change which offers you see. And they do that not based on any judgment of wheter it would be a good offer for you but only by how much those sites paid.
In such a world, I’m assuming their listings would be near the top as lots of people click on Amazon’s and EBay’s organic results (due to the wide selection on their marketplaces and generally very competitive pricing and good delivery track record). What people click on seems a good proxy for a search engine of “what are people looking for? (also phrased as “what will bring people back to my search engine next time?”)
In terms of winning the SERP paid ad bid, I don’t care if the search engine shows me the ad that’s best for them, because that will still give the newer business the chance to win the bid. If they choose not to, well, in that case I don’t see their ad. If the search engine doesn’t have ads, then in all cases I don’t see their ad.
If they go completely insanely rogue (like showing me a paid mesothelioma ad regardless of my search term), they either lose my future search traffic (“foosearch never has what I’m looking for”) or they lose the advertiser (“mesothelioma ads convert well for us on barsearch but not on foosearch; I need to lower or pull my bids on foosearch”)
> I think it's fairly clear that Amazon and Ebay would show up as the top organic results.
> What wouldn't show up is lmm-cookware.com, launched early October 2021, who is trying to establish themselves as a new destination for cooking enthusiasts.
If a new site is what users are probably looking for, then it should be top of the organic results. Google and other search engines already use signals like when the site was updated, whether it has a social media "buzz", or what price it's selling something at. It's highly unlikely that spending advertising money is a good signal; yes, promising new companies do spend advertising money, but so do established companies and out-and-out scammers.
Helping the user discover new sites is maybe part of what a search engine should be doing. But mixing ads into the search results is too intrusive and anti-user a way to achieve that (if it even does). Maybe there's a time and a place for advertising, but it should be clearly separated from organic results; Google used to be good at that (indeed they were famous for having advertising that was less intrusive than their competitors), but they've been getting steadily worse.
Yes, i was thinking similarly. The GP wants links, not ads (specifically), but if the trustworthy domains are saying this particular skillet is low quality then you want that first, rather than an ad.
I like this pedanticism, but I still think you can even argue the "by definition" piece of it. I suppose it ends up being somewhat tautological in the end.
If you really like Skillet X, and I ask you which skillet I should buy, and you tell me, "Skillet X," then it is not an ad.
If you really like Skillet X, then somebody gives you $5 to recommend it if anybody asks you what skillet to buy, and then when I ask, you tell me, "Skillet X...I mean you should know that somebody gave me $5 to say that to you but honestly I was going to say that anyway" then is it an ad?
It kinda intuitively feels to me like if it doesn't alter the result, it's not an ad, it's just somebody taking advantage of another person's willingness to hand them money and doing nothing in return.
The unstated implicit assumption in the question being asked in both situations is this: you’re being asked for your unbiased opinion. All ads are inherently biased, but by not disclosing your compensation, you’re not answering the question as asked. This bias, even if disclosed, renders your recommendation, and our hypothetical innocent recommendation, suspect.
I guess the argument is that if google could make the perfect search result, ad spending would, by definition either be exploitative of the advertiser by providing no change, or providing value by changing the results from perfection.
In reality, Google is not perfect and you can argue that ads do provide value by promoting relevant content, even if its gameable by our capitalist system.
Yes, it's by definition because the definition is tendentious.
The definition: "If Google [meaning Google's organic results] delivers users the thing they want"
The consequence: "ads have to be things users don't want" because advertisers whose product is wanted will be included whether they pay or not.
Now, you could widen it to say, look Google is still delivering the thing users want, they just sometimes do it by organic results and sometimes do it by ads, but that's a very convoluted reading of the definition, which clearly is using Google as a stand in for "Google's organic results".
The problem with my argument isn't that it failed to be a tautology. It's that the definition is questionable, and that's the profitable angle of attack: "users don't know what they want" "there can be multiple equally good options" etc. But saying it's not by definition is just silly. The definition is the whole thing up for debate!
>>> If Google delivers users the thing they want, ads by definition have to be things users don't want.
Ben Evans has a tweet something like "half of facebook devs are working out how to code the algorithm to serve you just what you want to see, the other half are working out how to get the algorithm to serve you what advertisers want you to see. "
By any definition, a podcast is part of "internet advertising" but there's less conflict there. Sure, I want to hear the hosts and not the ads, but if the ad is relevant, I don't mind the interruption, and hey sometimes it's an interesting product for me. There's a tension between the content and the ads, but not a fundamental contradiction.
No, this is the outcome of all interest targeting based models. You're interested in X, and they either don't show you X (to get X to pay) or they show you Y (which you aren't interested in) because Y paid.
Whereas with demographic targeting, hey I'm in demographic X and wasn't looking for a Y now, but sure, I'll remember it for when I'm in the market for one later.
This is the fundamental issue with all advertising-based models. Eventually, they all run into the problem of having to continue to grow. The only way to continue to grow is to display more ads, thus compromising the user experience, which starts the downfall. AOL was a great example of this. Google seems to be heading this way.
Google's original values were the ability to provide better search (common answer) and be fast (less common answer) - both of which were a complete contrast to the Alta Vista and other search engines. I could easily see Google facing disruption from a new player - but I don't think it will be another search engine. Probably a paradigm/systemic shift.
If you search for Hello Fresh, you'll get a Hello Fresh ad offering 14 free meals. If you search for Blue Apron, you'll get a Hello Fresh ad for 16 free meals. Firms are willing to spend more to win incremental business from a competitor, and consumers benefit.
Now you could argue that the search engine should show organic results to help the consumer get the best deal, and I would agree with you.
I don't think it is that clear, actually. If I sell a car that is as good as the Ford one you searched for, but 80% of the cost, wouldn't you want that ad to appear when you searched for ford?
Respectfully, I’d want the car you mention included if I search for “full size sedan” or “alternatives to ford”, and not to appear if I search for “ford”. This is a pipe dream of course with the way AdWords works, but that is what I prefer :)
If the discount car is relevant, then I want the search engine to show it to me whether or not you happened to pay them for an ad. And vice versa if it's not relevant. As a user, I don't care who is getting paid by whom; I care about seeing relevant content.
Exactly. And it's bad for customers too. My mum (70) just fell for it, and she's pretty clued about computers and technology. She searched for a flight with Ryanair, so entered Ryanair on Google. First result was some scummy reseller, which sold her the same ticket for a higher price with some "extras".
Bidding on a competitor's brand name keyword should be banned. But Google can't resist double-dipping.
I mean.. RTFM? Look at the address bar? The branding of the site? Pay attention and be careful? This stuff only works on 70 year olds. It stands to reason it won't be viable in a few decades.
This should be allowed, because this means smaller competitors have a shot at the customer base of a more established competitor. If we ban it, it's just shoring up established players.
Worst part is that the competitor has nothing to lose, so they basically will spend the whole margin of a sale on the ad. Which means that the original company has to do the same.
Where I worked before, Google could get $4 for ever $1 we made. And we actually delivered the service, and people googled our name. Pretty crazy...
> if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would
If I as a customer am using such specific search terms, then I would assume that my intention is to find and possibly buy this specific product. The results of other brands might be annoying, but why should I click on them? The relevant results are still displayed on the first page.
I think there is an under-appreciated average search engine user in the comment:
People will typically write their intent on the search engine even when they could simply directly to the website.
Case in point: The top 10 bing searches are for websites, including FB, Google, Youtube [1]. This traffic is highly competitive and should (as in all competitive markets) be bid among competitors.
The address bar is the search bar. My wife never types "facebook.com". She types "facebook", gets the google search page, and then clicks on facebook from there. It pisses me off that if I start typing facebook, Chrome doesn't autocomplete to facebook.com. In contrast, if I'm in Safari, and type "n" I get "news.ycombinator.com" autocompleted.
Small business people in my area of UK have always done this, type in the box in the middle (usually Google, occasionally Bing or some other service). But my pretty tech-literate kids do it too, even when I show them how they 'should' do it and that is faster, and they don't need the extra click to get where they're going ... mad!
On Chrome on Win10 as I use at work though, typing in the address bar, with my settings, I get auto-fill of addresses (the history search is noticeably missing vs Firefox) including the option to use 'tab to search' on a domain.
For example if the customer was previously not familiar with Williams Sonoma but had seen an ad in the paper or on TV or on the subway etc, that caused them to search for Williams Sonoma cast iron skillet.
And beside that a lot of people routinely misclick or click on ads not understanding that they are ads even when they are marked as such.
So the theory is that if I search for "Williams Sonoma cast iron skillet", but misclick on some competitor, I'm buying that instead? I don't know, I don't usually assume the rest of humanity to be a lot dumber than me.
People don’t need to be dumb to do that. Being inexperienced with computers or confused about computers is enough to cause people to do a lot of things like that. There are a lot of people out there who are inexperienced with or confused by computers.
But my main point was that the brand may not be all that important in the first place even though it was included as part of the search term. The person making the search could’ve seen an ad and been intrigued by the product, but upon landing on a competitor site they may choose to buy a similar product from them instead.
If you are deeply into the kind of thing you are buying, you will make a lot of research to find the best one. But there are a lot of things we buy that we don’t care as deeply about, and where we may choose the first one that fits the bill sufficiently well.
So much this on phones. I think I missclicked adds like 10 times in a year, before realizing there where adblockers on Firefox mobile. Zero legitimate clicks. Never happens me on desktop, unless there is a screen jump scam.
The competing ads will often say something like "Introducing [Product Y] which costs 25% less and is 10% less smellier than [Product X you searched for]"
If you've never heard of Product Y, you might be intrigued and click. Maybe you want something less smelly!
Your logic about the ads is sound, but your experience as a customer does not mean all customers exhibit this behavior. The best course of action is to test this conclusion which can be reliably done with a Google Ads Experiment.
Ultimately I think it decreases google's value as a search engine (to the user searching) by a very small amount, but nets them a high immediate return. I bet it's hard to quantify the net effect over time, and it would be a really hard sell internally to not allow it.
I've noticed the quality of google search decrease drastically over the last 15+ years or so. I don't think that's directly tied to ad buys though.
I have searched for "Digikey something" and wound up with clicking on a "Mouser ad".
At this point, Google's first page is so bad that I can almost build an anti search engine. Search for a term, and then exclude all the sites on the first page from ever showing up permanently ever again.
runnaroo showed things could be done better. The problem is that doing better doesn't seem to convert to profit.
Freakonomics had an episode with an economist that worked with Ebay on ad buys. This sort of "buy ads on your own keywords" was shown to have zero impact on sales to the point that they cut completely stopped advertising when the search included "ebay".
This will work for someone like eBay (people searching eBay want eBay) but for other "brand-name" terms it may NOT work - people searching for Travelocity or whoever is the hotness there may be perfectly happy with the first "similar enough" link.
There's no way to know until they do an A/B test like eBay did I guess. EBay was certain that they needed to do the on brand ads before this economist showed up too.
It's when the phrase is at or near generalization that it comes into play - someone searching for Kleenex likely doesn't care what tissue they find, and many people who Google Google would be happy to click Bing if it popped up first.
"Dominos" might be one where people wouldn't mind ordering from Pizza Hut as they're using it as a generic term for pizza.
eBay is a strong brand where people searching for eBay are going to click on eBay, almost no matter what. For other less recognizable brands or crowded categories, this is often not the case.
Branded search terms are almost always less incremental than non-branded (ie: "lodge logic" vs. "cast iron skillet"), but the actual incrementality of the terms is something every advertiser should be testing continuously.
> if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements
And if they were not above organically they would simply buy the advertising space that William Sonoma purchased. It's one of the slimiest things Google does - allows competitors to purchase advertising space on a query specifically crafted to find a particular source. It's nothing more than a shakedown.
Yeah, I think the only on-brand search ads that should be allowed are ones for totally unrelated products OR those ads are placed below what on-brand search results would provide. Google has no incentive to fix this though because it's an extra tax they charge the entire online advertising space (+ all the other search providers do it). Carefully crafted legislation could put an end to this tax.
> Yeah, I think the only on-brand search ads that should be allowed are ones for totally unrelated products...
The problem is that this is subjective and would need to be automated somehow. I think Google's original sin here is making the ads look so much like organic search results. Someone placing an ad against a competitors brand name would not be a huge issue if organic results were still front and centre like in the "good old days" of early 2000s Google.
This is actually trivially automatable. First of all, brand keywords are something Google supports explicitly for this kind of targetting so they know from that direction.
More generally though, if you paid for an advertisement and the natural search result has you first anyway, then you should not be charged for any clicks to this advertisement.
Well, on a good search engine, IMO, you'd have a predicted store based on the manufacturer (eg their preferred seller) then perhaps a list of top 10 competitors and something like "most recorded purchases after this search are from seller X" with "the most popular similar store is seller Y".
Of course Google wouldn't give that data out as then many companies wouldn't need to advertise at all to get top billing.
That's secondary & the reason Google doesn't have this is likely because the sellers don't want to provide this. E.g. Amazon doesn't want potential customers being easily redirected to Walmart purchases. One of the many reasons Froogle died.
The simple solution is that your paid advertisements are free if it gets clicked when you're the top result anyway. That way it doesn't cost the brand any money to bid on advertisements for their own brand.
Why is that a shakedown? Shouldn't competitors be able to advertise their products? Any users who are actually looking for Williams Sonoma will find it. Any who are open to having their minds changed, or are interested in competing products, will be interested in competitors ads.
should your competitors be able to buy 100% of the screen space on mobile with images where your organic result has none? 50% sure maybe but literally google doesn't even put the organic results above the fold on mobile. Often on google on mobile the organic results are 2 screens down.
> The main problem here is that if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements.
This is what is known as "on brand" Search ads. I like to call these effectively the "Google Tax" because publishers/retailers are forced to pay Google for the traffic they would have already received had the ad not been there.
I've seen way too many companies look at their analytics and say "see we get 20x ROAS on on brand! why would we turn it off?!?". Because silly, people are already going to go to your site without you paying for the traffic. I wouldn't be surprised if 25% of Google's ad search revenues come from this.
I purposely click the non-ad link when searching in Google (because I'm too lazy to type the full url). Does that help at all, i.e., prevents unnecessary ad spend from bidding?
I do the same, but 99% of people do not, so it's really a drop in the bucket.
I do wish there were some regulation such that if a user is searching for a trademarked or copyrighted term, that the best organic search result for that term should be required to show up first. I'm fine with showing competitor ads, but I don't think they should be able to show up above the trademark owner's ads.
It feels like it is close (but not really) to a trademark violation. If I search for "A" and get a screen full of results for "B", if I need to scroll to even see mentions of "A", then that feels like it is really close to "passing off". I could see an argument for saying that this "passing off" causes customer confusion that requires trademark enforcement.
I realise that would probably be an impossible argument to win under current law (IANAL), but it _feels_ so very close.
US case law is extremely clear that bidding on and running ads on trademarked keywords is not a trademark violation, so long as you aren't misrepresenting that the replacement is the trademarked thing; see e.g. https://www.americanbar.org/groups/litigation/committees/bus....
Way more than a dollar. If you really want to cost a company a pretty penny, click on the ad after searching for an "intent" to do something rather than the exact brand. Some of those Adwords cost many tens of dollars.
I haven't seen anyone mention the fact that a big factor that contributes to how the Google Ads "black box" determines how to rank ads is via an Ad Quality measurement.
Each ad is given a quality score, and since your own website will be guaranteed to be the most authoritative source for a keyword search with your own branded keyword(s) in it, by default the quality of your own ads will be much higher than your competition.
This means in practice that the cost for a brand to be shown first on their own branded keywords is much lower than their average CPC, let alone the cost for their competitors to be there.
The very fact that you have to bid on clicks for your own damn website by paying Google, because Google has prioritized ads to the point of displaying them higher and making them almost indistinguishable from the "organic" search result proves they have intentionally created a tax on the internet.
It will get worse, to the point where to get auto-completed in the omnibox you'll have to pay. TBH I am not even sure if that isn't already the case!
I'd rather ignore some ads which are less intrusive than places like Amazon than pay a subscription fee to use search, gmail and docs which are funded by ads.
There are other options. For example, funding a publicly-owned search engine with $1 billion a year in infrastructure and development costs would be 50x cheaper than what the US spends on foodstamps. It'd be 1/10th the NSF's annual budget.
Not that that's the best option, but we could consider some new ideas.
A publicly owned search engine, that will absolutely not be used for propaganda purposes by, say, the Department of State, or by a megalomaniac elected representative.
Need to build consent for a war with <Country>? Give me access to the search index, and a few weeks, and it'll be done.
I didn't say it was the best option. But wow, you just trust a billion lines of closed source written to satisfy ad optimization criteria with zero transparency? I mean, we can subject things to public oversight.
> But wow, you just trust a billion lines of closed source written to satisfy ad optimization criteria with zero transparency?
I trust that a multi-national serving multiple markets is going to give me less nationally biased information than the government of a superpower (with political ambitions that span the globe.)
It's why I generally don't turn to the Voice of America for news.
You think Google search doesn't adjust its search results based on your geography? It absolutely does, and that's just one of hundreds of inputs into its completely inscrutable machine-learning driven tangle of a ranking algorithm. Your trust is very badly misplaced if you think that Google can't be/isn't already corrupted by the corrupt governments it must deal with to do daily business.
Lol if you think that isn't happening right now. Google made a change to its algorithms sometime in the past 5 years to surface "authoritative" sources, meaning sources that toe the state department line. For example, the World Socialist Web Site, which has a consistent anti-war line reported being demoted.
"Google blocked every one of the WSWS’s 45 top search terms" (8/2017)
"In a statement before a Senate hearing on October 28, Sundar Pichai, the CEO of Google’s parent company Alphabet, admitted that the dominant online search company has censored the World Socialist Web Site.
At the Senate Commerce Committee hearing, when asked by Republican Senator Mike Lee of Utah to provide the name of one left-wing “high profile person or entity” that has been censored by Google, Pichai named the WSWS."
"Google refines its search results to curb fake news" (4/2017)
(the comments below this article are funny because it's full of liberal-aligned comments on the disproven Trump-Russia conspiracy theory that is still given life in mainstream outlets)
Sure, but there are no shortage of top level generals and intelligence people that identify with Democrat agenda, now that democrats finally gave up on things like freedom of speech, privacy, individual liberty and such.
A government owned search engine is 100% bad idea, it will cost a ton, it will cost more than equivalent business developed one, it will be abused by governments and law enforcement, your traffic will be combined with government data to make a last century dictatorship police state weep for joy
be a money maker for people you very much don’t admire
most of all there is no public value for it when the government can legislate operators
It's called Quality score (soon to be discontinued) and has 3 components. Only one of which is landing page related. And nothing prevents competitors from building "more relevant" landing pages than the original brand.
Google keyword planner tool shows cpc (cost per click) for keywords. For an online gambling keyword it shows me 14.97€ (for one click). There are other industries with cpc's of 80€ and more.
No, it's got to be a real person intentionally clicking the link, and very large teams of skilled people work exclusively on figuring out which clicks fall in this category. Cursor movement patterns, delay time, whether the link was actually rendering on the display when it was clicked, IP address, cookies, signed-in identities, previous click patterns, location, all of these things and more factor in. When people talk about Google hoovering up data, it is in large part for this purpose.
Sounds like a good use of AI, if not one of the best! Let’s train an AI using real humans and have it click these links. Run the client in an P2P network so multiple can contribute to this.
If ads are no longer profitable, then this should sink both the online advertising market as well as Google.
Clickfraud has been around forever and teams detecting clickfraud have also been around forever. Automation that impersonates human users is not a new idea.
Yes I’m well aware, yet AI hasn’t existed for as long as clickfraud and we both know at some point a program with enough sophistication will look like a human.
Before Google banned much of the advertising on the category, some addiction treatment keywords cost north of $300 per click. “Drug rehab Las Vegas” was my favorite, because if they didn’t use a tight keyword match, they would get charged for people trying to find the Rehab pool party.
Yes. As an experiment, I tried running a Google Ads campaign for my https://chezmoi.io open source project bidding on "dotfile manager". Twenty clicks cost me $20. I terminated the experiment quickly.
That could be the boycott of the 21st century. Don't like how a company is behaving? Get a social media campaign to have people make intentful searched in different search engines, click through, go to a signup form/add stuff to a cart, and then just leave.
If I can do 10 of these, which would take a few minutes, I could cost a company $100. Which means, with the right tools for spoofing my identity, I could cost them $1000 an hour. If I spend a good work day on this, I could probably cost them $10,000.
If I convince 10 people to do this with me for a day, we could cost them $100,000. We might be able to cost them $1M in a week.
Said systems can probably cope with most of the regular stuff, but what about an army of Twitter followers? Most are going to be signed in, they are going to be spread out geographically, they are going to be real people with real browsers with cookies, etc.
Yes, but those aren’t spoofed identities, they’re actual humans intentionally clicking the link. This is harder for the defender to defend against, but it’s also vastly harder for the attacker to scale. Defenses against this sort of “organic” traffic do exist, because this is something that people committing ad fraud attempt to do using compromised browsers of real people.
Edit: I mostly want to stress that there’s a huge body of (largely proprietary) work on this topic. Thousands of skilled people have been full-time employed for decades thinking about exclusively this topic, on both the offensive and defensive side, in a continuously escalating arms race. Anyone just getting started on this topic has a couple of decades of literature review to catch up on before they should imagine that their proposals are novel. I don’t mean to discourage you if it’s a topic you find interesting, but you should bear in mind that the answer to “But has anyone thought of XYZ?” is “Yes” for basically all values of XYZ. There’s a lot of money at stake if the answer is no.
It sounds like a really fascinating topic! If someone is interested, where he/she would need to start? any papers that review most/all methods that are up to date?
I've done that once in a blue moon, but it doesn't really accomplish anything except transferring that money to Google. Since I used to work for Google, thanks, I guess :-)
I mostly click on ads as chaff (mostly I block them but sometimes unblock them just do do some random clicking in the hope of confounding some profiles being compiled on me)
> I like to call these effectively the "Google Tax" because publishers/retailers are forced to pay Google for the traffic they would have already received had the ad not been there.
Isn't this just the usual problem with advertising? You have to do it because the other players are doing it. If nobody did it, it would still be the same cake to be shared.
This case is more on the nose, but only because of some fairness assumption that "Williams Sonoma Cast Iron Skillet" _ought_ to be traffic for Williams Sonoma.
> Isn't this just the usual problem with advertising?
Not quite: OP's point is about searches that specifically include your brand name. The "usual problem" with advertising in the zero-sum
sense you propose is for eyeballs in general (billboards etc). The google tax here is more invidious, being specifically about searches that include your brand.
I think it's the same problem even if it looks more suspect.
I think the important thing here is that companies don't decide to increase total advertising budget just because of the existence of a new channel. They have a budget of X. They have to decide how to break it up. Google, with brand targeting in search, is making a play for money that would've otherwise gone to Fox or ClearChannel or whoever. The brands redirect some budget to this new channel, as it shows its effectiveness.
Many of them still spend some on traditional channels, to get that first appeal and be the "Williams Sonoma" in "Williams Sonoma Cast Iron Skillet." Many of those people aren't just going to click through to Lodge ads instead. But less than they did when traditional channels were the only game in town.
Same dilemma, though:
If nobody advertised in Google Search or non-Google channels, people would just buy whatever skillets they found in whatever stores they found, or what their friends told them about.
If one brand advertised in offline media, more of those people would buy that brand.
So all brands advertise in offline media. And then it comes down to effectiveness of the campaigns + the same criteria that it would've otherwise - availability/placement, word of mouth, etc.
Then, if only one brand advertised in search targeting other brand names too, more of those people would buy that brand.
So all brands advertise on their brand keywords in search.
But ultimately it's the same game in both places. With the same net, Google just captures some spend that what would've gone to non-Google places previously.
When people search Pepsi Cola they expect to get a link to Pepsi or information about it. They don’t expect to see Coca Cola or RC Cola, etc. now in this case Pepsi makes sure search isn’t polluted by paying Google good money to always be the top results for that term.
Now, if you searched on the generic ‘cola’ or ‘soda pop’ then yes you expect to see those who bid higher to be at the top and at the bottom those who bid nothing unless organically somehow they ended elsewhere.
Advertisers are the customers, or am I missing something? To be fair, I don't consume any adblockable ads or TV ads (maybe once a quarter for TV) so I don't know if people click them, how much does purchasing ads affect the normal search score?
>pay Google for the traffic they would have already received had the ad not been there.
But this line of reasoning begs the question. If this system wasn't in place, then Google search would not exist as it does and the search traffic would not necessarily exist in the same way, no?
I don't see why the search engine would work differently or be any less popular if Google used its algorithms to determine intent (that's Google's whole thing) and then automatically (or via human moderation) prevent companies from showing ads when there is clear intent to find another company's brand. Google would be somewhat less profitable, perhaps, but the search product would still work fine (arguably even better) for the people searching.
> Google search would not exist as it does and the search traffic would not necessarily exist in the same way, no
yes and no. Google search didn't have as intrusive and competitive ads as it did before and the search function was effectively the same. Nowadays, if you search "Williams Sonoma Cast Iron Skillet" you get 5~6 ads and 2-3 scrolls to get to the real organic results. 6~8 years it was like 1 ad and you could see the first result.
At this point I'm actually wondering how much more discussion will be required about "Williams Sonoma Cast Iron Skillet" to get HN among the top search results.
Another reason why ads for "Williams Sonoma Cast Iron Skillet" exist are not just to combat Williams Sonoma competitors but also scummy SEO search engine spam, I mean what if you don't pay to have your stuff in the first results and some hustle comes along that makes fake product review sites off of scraped results and engineers getting their stuff all among the first results, since UX research shows people not finding a relevant hit in the first few results do not go to the next page in results it implies that ads are required to give the correct research for something obvious like this because otherwise you have to take a chance with the present day subpar google results.
I agree with the rant how Google introduced a tax that wouldn't have existed in a world without Google. And, technically speaking, it would have been possible for Google to not charge for that tax (but would you have done that in their shoes? It's not like their market share is suffering).
But if you are implying that brands should not invest in on-brand search campaigns, then this is a really bad advice. It's a known fact that targeting your competitors' branded terms is ROI positive, which definitionally means that the affected brand is unable to capture all the customers who were initially searching for it.
> But if you are implying that brands should not invest in on-brand search campaigns, then this is a really bad advice.
I'll be more explicit - you should test the difference. I usually recommend companies do a blackout month where they turn off all programmatic ads and then do a like-for-like comparison.
To further emphasis my point before, I've seen ad ops agencies say "hey you've got an overall 15x ROAS" and then find out that 80% of their ad spend is on-brand (give them 25x or whatever) and the off-brand (which is 20% of the spend) is giving them 5x. So their ROAS is inappropriately distributed to try to reach an overall ROAS goal.
There are bad actors in every field. I would just be careful in using such anecdotes to distract from the overall message, which as you spelled it out in detail seems sound (with perhaps just a note that a 5X ROAS is not bad for most unbranded campaigns, but I get your point).
It’s less the Google tax than the “advertising is now efficient” tax.
Any other medium would theoretically have the same problem: if Ovaltine doesn’t sponsor kids’ radio shows in the 1950s and someone thinks they can deploy capital to grow a competitor, that someone will buy that slot. People couldn’t do this because there were human processes and relationships slowing this marketplace down. The thing that Google did was make it possible to test this at small scale.
It’s pretty shocking. Had similar discussions w/ co’s dominating organic and paying ~~fucking con artists~~ affiliate marketers. IMO those are faaaaar worse than poorly allocated search spend..
An ad directing shoppers to your competitors who pay them is not a protection racket, no more than a travel agency directing travelers to businesses that they have a relationship with.
It's the entire business model of aggregators and intermediaries and middle men, which compose roughly half the world's working population.
You tell a cab driver to take you to Morton's Meats and instead they take you to Shelly's Steaks (who pays them a kickback each time they do this), and then charge you for that detour to earn their kickback when you demand they take you to Morton's like you asked in the first place.
Is this appropriate behavior by the taxi driver?
Would a regular person consider this deserving of punishment?
Is Google's search results advertising bidding system ethically equivalent to this?
It's a poor analogy. Google isn't taking you to the wrong website, they are showing you the alternatives at the same time as the search result you wanted.
Wait. Is thst true though? I'm a technical person, with a lot of very specific searches. Around 70% of the time, I don't actually get the results that answer what I'm looking for, no matter how many quotation marks I surround my strings and no matter if I use site:x.com, and various other Google hacks I've had to come up with to try and find my relevant results.
The reason this happens is because the Google result shows my words in the excerpt, but when I click in, there's zero percent of the content to be found. That feels like a bait-and-switch akin to the taxi analogy.
My relevant results now either don't exist, or is in an absolute sea of mind-bogglingly bad results that it wears me out such that I would rather have had no results to begin with.
In this sense, I feel like Google is absolutely purposefully showing me results that I didn't want to go to.
It is. If you want a more accurate analogy, it's like getting into a cab and asking to go to Morton's Meats and the cab driver saying "are you sure? I know this really great place called Shelley's Steaks that's better and cheaper! What do you say?" and then you get to make a call as to whether to trust the cab driver (google) has your best interests at heart or not, and whether to go where you originally wanted or to where you were recommended.
This happens in real life all the time. Cab drivers get kick backs from specific places they have relationships with if they deliver people there, so you have to wonder whether they're actually recommending a better place for you or a better place for them, just like you do with Google.
And yes, I've been referred to a place by a cab driver when asking for a recommendation and found it didn't actually serve my needs at all, and in retrospect it was obvious from the conversation I was steered there without care for what I really wanted.
And then Billy's and Shelly's actually make terrible steaks with all kinds of hidden costs and cheap side dishes because they cut corners in order to make up for the loss on their annual advertising spend.... Google helping business... Yeah. :L
Don't forget the real-time bidding market. If you're wearing an expensive watch, Shelly's will give the driver a 10% bonus, or if you have a nice pocket square, then Morton's will offer you a window table.
> The reason this happens is because the Google result shows my words in the excerpt, but when I click in, there's zero percent of the content to be found.
I’ve seen this too, but I’ve always assumed it has to do with meta tags or something.
No, it's due to illegal SEO practices where they serve Googlebot a plate of GPT3 meets keyword spam, and then serve you a page full of advertising and phishing links.
I've been wondering for a long time what's been going on. This makes a lot of sense!
So if the User Agent is Google, the website produces relevant-sounding content with no JavaScript spam? I never realized GPT-3 could be used for this. The next time I see a bait-and-switch website in my search, I'll try changing my User Agent to Google.
Supposedly the Chrome safe browsing service crawls the web looking for malware. Since that's focused on finding malware they likely make it look like a normal user in terms of user agent and IP address, so it can probably detect these pages.
I believe the signals from the safe browsing service are used to affect search result ranking, but I have no idea if they look for different content being served to the Google crawler.
GPT3-alike, not GPT3 specifically. Google’s bot can’t detect when it’s being fed plausible gibberish, or worse, stolen content with random keywords injected. I’m very curious if your Googlebot surfing comes up with any interesting outcomes!
Reminds me of the tuk tuk drivers in Thailand who will show you alternative destinations like a tailor who gives them kickbacks before dropping you off at your destination.
You pay the taxi to take you to an address. You don't pay anything to google, you decided to use their ad funded "search engine", which they don't charge you anything for, and you complain that you get... Ads?
Down that road lies the argument that Google should be treated like a public utility. If the only mode of transport available were sponsored taxis, then people would be rightly demanding for the ability to pay for a car or use a government run public transit system.
Okay, assume the taxi rides are free. The taxi driver is now wasting your time taking you all across town just for a selfish kickback. Meanwhile, you value your time, presumably, so you're getting more and more annoyed as they do this. I don't see how removing the cost of taxi somehow makes this scenario acceptable. What am I missing?
Eh... that's not a great analogy. The equivalent of your story would be if you clicked a link for Morton's on the Google SERP and it actually redirected you to Shelly's.
What Google is doing is more like taking your destination, then asking "hey, you might like this other steakhouse more, want to give that a try instead?" before driving. Potentially annoying if the answer is always no, but... the answer isn't always no.
I think the analogy is pretty spot on. The whole point is that your query expresses your intent to Google. Assuming their algorithm can figure out what your intent is (it certainly can in this case) then they are intentionally acting counter to your expressed intent.
A point to add though that highlights the issue is that for many many years prior, Google had gotten users used to the ideal that search results were fair and not influenced by payments under the table though. The injustice is turning a free and fair situation into one of deception and payola that will likely get much worse because rising consumer cost is the end result.
Search isn't influenced by payments, ads are. It's not that hard to understand. If the idea of ads offends you that much, the organic results are a couple inches down.
The abuse is worse on the merchant's side, but on the consumers, it's more like getting a cab to go to Morton's Meats, instead they take you to Shelly's Steaks (because they got a kickback), then say they'll take you to Morton's for no extra charge, but tell you how great Shelly's is, and you're already at Shelly's, so is it worth the extra effort?
So they're allowed to waste my time and earn a kickback, as long as they give me a free cab ride from their kickback to where I wanted to go in the first place?
That sounds like it only works if my time that they wasted for selfish gain is valued at $0, which is obviously true as far as they are concerned, but certainly isn't true to me.
I don't defend dark patterns, it obviously is one there, but your time is worth zero when you pay zero. The analogy would be free taxi drivers taking your to their kicking back spots first.
If your time is worth X, spend X. It may prove a bit difficult since the paid search engine marketplace isn't thriving.
How much would you be willing to pay for a search engine that only displays organic results? There are actually free solutions easy to set up to get that from Google results, but I'm wondering since you mention value of time.
They tell you they’re going to make a short detour which will hardly inconvenience you or cost you more than a couple of bucks for extra via Shelly’s Steaks en route to Morton’s meats. They neglect to mention that SS is giving them a kickback.
That’s what the ad does. It adds a bit of cognitive load where you have to view and then ignore the non-William Sonoma ad but ultimately, and without much fuss, let’s you go off to WS.
> ethically equivalent
No. Google ads are fine when seen this way. The can driver’s behavior is more egregious if their decision to take a detour isn’t communicated to you AND runs up your fare by enough to materially affect you, whatever that level may be.
Disclosure - I’m actively trying to un-Google my life so definitely not a fanboy.
You ask the waiter for a Coke, they ask you if you want Pepsi instead (because they get kickbacks to sell Pepsi). Is that appropriate behavior for a restaurant?
How much is Google charging you for your searches? I get them for free. Sure, sometimes it's a pain to scroll down past the ad results when they're not relevant but it only costs me time and attention.
Most restaurants don’t carry both Coke and Pepsi, which in this analogy would be the taxi driver truthfully saying “Morton’s is closed”. That’s a fine time to offer an alternative, but that’s not what’s being discussed here.
"All too often the middleman is milking both sides."
By definition a middleman is milking both sides. Were a middleman not there, it would definitionally be better for both parties. The only case where this is not true is the one where the middleman makes no money
No, that assumes perfect market information. Middlemen can provide benefit by allowing people to offload research and reputation tracking to a third party that they trust, allowing them to invest time and effort in that one entity instead of a bunch of smaller ones.
For example, do you buy off Amazon or NewEgg, Best Buy, etc or do you just use Craigslist, Ebay, the local paper, or even the local flea/market swap meet for electronics? You can probably find stuff for half as much or even less at some of those locations, but it might cost you a lot in research time or lost money in the end...
My apologies, my phrase "milking" was meant to imply "overly burdensome resource taking" when compared with their value add. IE cost too much
To rephrase, all too often middlemen are there because of legislative/financial props from government or other incumbent advantages and - as you say - both sides would be better without them.
In this case - as in many middlemen cases - Google provides value - their search services. This "value" is the hook for their place in the transaction.
If Google provided a useless list of random results for every search term then it is obvious they would lose their place.
But where is the crossover point? Where Google doesn't provide enough perceived value compared to their perceived cost such that the market abandons them?
Peoples growing dislike of ads and - from this article - advertisers changing perception of ad value will change the cost vs value equation.
Hidden kickbacks are a racket though. Google has very deliberately been making their ads less and less distinct from their top search results, in a frog-boiling way.
Remember when they made a point about how they had a blue background and were on the right rail? Then they got moved above the organic results, then the blue changed to yellow, and every PM working on SERPs shrunk the size of "sponsored" and made the yellow a shade lighter for the next 5 years and got a nice bonus for it.
Put some money on the altar, who knows what might happen if you don't. The Algorithm is only impartial until it isn't. You surely don't want to go to irrelevancy?
I don't think Google has a profit rate massively higher than other similar companies (say, Apple or Facebook). It gets a lot of profit because it is large.
They also have a responsibility towards the person searching. I know more than a few people that have been conned into thinking an Ad provider is the official site for a service.
Especially with government websites. Lots of shady companies use ads above the actual search results to drive people to their site. Charge people for a service that is free.
Would you prefer a monopoly where only Williams Sonoma can advertise on WS keywords? I’m sure if you wanted to start a little premium homewares brand you’d feel differently. Bidding on competitor keywords allows small and big brands to compete for the same transactions, and allows brands to introduce alternatives to customers who might only have one giant brand top of mind.
Defensive branded search is generally only a fraction of budget anyway. I don’t find this practice the least bit onerous or extractive from Google — they’re just allowing for competition, as they should.
I want a WS skillet, and the first X results are for lodge... I don't need a lodge, I don't want a lodge, i want a WS, and google is not showing me what I want.
Is the premise here I search for “ford electric car” get 3 ad for other makers, and ford in the top 1-2 search results
So I click on the Toyota ad cause it’s on top, and buy a Toyota? Which is taking business from Ford ?
What is the answer, no ads hmm that’s not happening on a free search engine.
Mark the ad as an ad hmm done
Don’t show other results for any brand related keyword search, when all keywords are blocked probably nothing left to show
What should the ideal free search engine show when I search for “intel vs AMD”, should intel complain the first result is for a tech site.
I do genuinely wonder if the conclusion that the brand company lost money is valid, I do wonder if those searches are like ppl putting cnn in a search bar or generic searches
imagine you are driving to a specific restaurant see a billboard ad for another restaurant so change directions and go there, would you dislike the bus/building with the sign
And wouldn't any users who end up buying a Lodge pan have been legitimately converted by effective lodge advertising? I don't think anyone who's only interested in Williams Sonoma will just go ahead and buy a Lodge pan. Unless you're suggesting that simply because the user entered "Williams Sonoma" in the search bar that page somehow "belongs" to them, which seems a bit absurd.
> The main problem here is that if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements.
When you put it that way this sounds like racketeering.
If you take as a foundational assumption that the brand name in the search string means the traffic is in some sense owned by the brand, then you're absolutely right. It's essentially racketeering. Certainly brands often view traffic that way.
Personally I resist the idea that a brand owns my attention because I used a keyword, but that's one of my many personal quirks.
It's a little dicer in the cases where we can pretty much infer that they're trying to get to a specific thing and would have happily used a direct URL to the page if they had the wherewithal to do so. In this case the ad placement is basically a brand trying to hijack my attention while I'm in the process of seeking a thing out.
I struggle to think of a meat world analogy. It would be something like if I dialed my girlfriend up on the phone and, instead of routing me straight to her, I had to navigate through a switchboard asking me "How about talking to these sexy singles in your area instead?" And in order to prevent this, my girlfriend would then have to pay the company to route my call straight to her.
Of course this isn't a perfect metaphor because there's a lot of different ways people use a search bar, especially now that search bars are merged into URL bars. But that sort of gets at what it is about this that feels sleazy.
I think it also exposes the core problem. The metaphor rests on knowing intent with certainty. It's perhaps possible that certain clarity and pretty much inferring might not always be the same, especially with how search and URL bars have merged.
But I understand completely. If you genuinely feel like you know that person's intent with certainty, someone else having a crack at their attention along the way feels like a violation of your relationship.
Personally I resist the idea that a brand owns my attention because I used a keyword
What kind of keyword are you referring to here? Because brands do own certain keywords, they're called trademarks. If you don't want a brand to feel entitled to your attention, don't include registered trademarks in your search query.
(Yes, I know, trademarks are limited to a specific market segment. Doesn't invalidate the basic premise of searching on trademarks though).
If I search for "Toyota RAV4", the first (ad) result is "Hyundai Tuscon". If I search for "AWS Cert", my first (ad) result is "Microsoft Learn". Et cetera et cetera :|
There is absolutely nothing independently governing and monitoring whether performance is correct on ads. It's all done in private, and you're forced to compete against SEO and many other things to succeed on a daily and even minute-to-minute basis.
This is the real price of a constant threat to Net Neutrality, and allowing one monopolistic company to dominate mobile devices, web browsers, search results, and the largest video service on the entire Internet.
Their plan to corner and manipulate what everyone's freedom of choice and to secure their funnel of permanent revenue is considered cute to investors, but no one realizes how bad this will get in 5 more years.
Of course the numbers are fudged when you consider how they've turned analytics on their once very useful platform into a confusing mess, and when they announced that they were going to retire the system after it has killed off competition, because they can simply gather any analytical report they want privately from their web browser.
Public front-end statics are no longer trust worthy because they can be manipulated to drive platform revenue and engagement. The best and most accurate stats are provided only internally, to executive leadership that owns platforms.
Because we now use them for email, video views, browsing, phones, etc, they have key insight that can even be used for corporate espionage, your ideas can literally be beaten to market because your virtual assistant caught you mentioning keywords then reported you applying for your patent and corporate loan.
Most people have no idea about how bad this all can get. We'll find out soon enough though.
When ad revenue drops on platforms, the platforms simply reduce organic visibility which drives the need for regular ad spending for companies in order to remain visible on social platforms... AirBNB is riding a wave of prior popularity and name recognition, I guarantee they will go back to a certain point of obscurity at some point because they reduced their ad spend, and then be forced to promote heavily as they did once before.
It's all creates a new cycle of financial deception and manipulation on platforms. For very profitable companies, advertising is usually manageable, but for startups, for small business, and for independent creators, this practice is devastating financially, and fruitless on top of the financial loss of paying for promotion. These platforms also made promises to woo users based on free organic growth, which somehow conveniently disappeared due to covert and convenient EULA updates over time.
our brand name "Paymo" is considered by google a synonym for things like project management. we get competitors that are advertising broadly on our brand name... it's truly mind blowing.
So in a way, big advertisers that are leaders in their markets would be favorable to adblockers, because they free them of the need to do defensive advertising.
One could argue that no advertising is hard for the challengers, but in today's situation they are outspent anyway, so what do they have to lose?
The conclusion is that both incumbents and challengers would be better off in a world where no advertising exists.
Yeah, it's mostly a prisoner's dilemma where if someone spends on ads then their competitors also have to spend. The only one really profiting in the end are the ad companies.
> The main problem here is that if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements.
Except they won't click those competitor links, because they are already specifically looking for Williams Sonoma
I always skip the ad links and click the organic one, even if it's for the correct company/product. Because when I click on the ad link 9/10 is some weird ad funnel page radther than the actual product page I wanted to go to.
- review sites like Yelp (interestingly one of Google's loudest antitrust critics)
- domain names (arguably the entire ICANN generic top level domains sell-off)^1; US trademark grantees who fail to take action against confusingly similar domain names risk losing trademark protection
In the case of Google and Yelp, the tech company can manipulate the alleged "algorithm" (read: no humans involve so you cant sue us, haha) behind the scenes and determine the "visibility" of the ad/review.
Ads are the perfect "business model" for tech companies because there is so little scrutiny of the ad services delivered. It is like philanthropy. When we make a donation, we generally do not track what happens to the money afterwards. We get a warm fuzzy feeling from making the donation as it is "doing the right thing". Then we leave the recipient to do as they please. How many companies buy ads on Google because they feel it is "the right thing to do". How many feel their donation was unwarranted after the ads fail to produce results.
Defensive ad purchases, defensive domain name registrations and even defensive gTLD purchases are one side of the coin.
Another is the "winner take all" line-of-thinking (80/20, network effects, etc.) that tech companies worship that they in turn project onto customers. Competing for visibility on a fully searchable web of enormous capacity presupposes (artifically) that visible space is scarce. Ad auctions for the purpose of placing an ad on page one of hundreds of thousands of pages of results. The truth is that many people used to read newspapers from beginning to end. The whole paper, not just page one.
In a physical newspaper, there are ads on many pages. Often there are in fact no ads on page one. How many Google customers are encouraged to purchase ads that will appear on SERP #2 (do they even have ads on SERP #2). Imagine if newspapers tried to create a bidding war for page one of the physical newspaper. With the online versions, it seems that is exactly what happens. Do not blame the newspapers, do not blame Williams-Sonoma, blame the "tech" companies. This only reflects the pathetic "tech" company psychology, not the thinking of the businesses who give them money for online ad services.
AirBnB has such good brand awareness that it’s not surprising they don’t need to advertise very much anymore”. On the other hand, advertising might be a pretty good investment for their competitors, like VRBO.
It's weird that Rand Fishkin has been in the marketing space for so long, but somehow hasn't encountered a single effective performance marketing team that measures their spend well and complements the efforts of SEO and social media marketers (like Rand).
Hard not to see this as a clumsy sales pitch for his company, especially when it starts with such a disingenuous example/quote (of course AirBnb didn't have to spend on performance marketing during a global pandemic where everybody was suddenly looking for a getaway...)
> It's weird that Rand Fishkin has been in the marketing space for so long, but somehow hasn't encountered a single effective performance marketing team that measures their spend well and complements the efforts of SEO and social media marketers (like Rand).
I mean I guess he would know from personal experience? He hadn't been the CEO for a year or two at the time, but he was still very much the public face of Moz when they "asked 28% of Mozzers to leave"[1], who worked on products complementing SEO. They used various euphemisms, but ultimately it's because they made no money[2].
It's well known that the self reported performance numbers from Google Ads and Facebook Ads are inflated, however those ad channels still do drive real value. I run an ecommerce marketing attribution company (ThoughtMetric) so I have first hand knowledge and data about this subject.
The most common source of inflation in Google/FB self reported performance numbers is multiple ad channels taking credit for the same order. If a customer clicks a Google ad then clicks a Facebook Ad then makes a purchase, each ad channel will claim credit for that purchase. In reality each ad platform only has claim to ~50% of the purchase (depending on what attribution algorithm you want to use).
In terms of knowing that real value is produced from these ad channels, I see it every day in many of our customers data. Clients will increase or decrease ad spend and there will be a correlated increase/decrease in sales.
tldr; Google/Facebook ads over report their numbers but do ultimately drive sales according to the data I have first hand access to.
One of the most vexing problems for me is how to calculate how much to spend. We had a year of double ad budget, but we didn’t see a corresponding rise in sells. Only 10%. We assumed we reached market saturation, but it still a lot of guesswork, even when meticulously diving into GA data.
I always used to just capture the old UTMZ cookie data (or rebuilt it when the actual UTMZ cookie went away) and stored that with the order data. It's not the most sophisticated attribution method but it at least gets around these sorts of double counting issues with orders.
No, advertising is not a scam.
It's only a scam if you pay for the visitors that would have come anyway (buying your own keywords, and also retargeting/remarketing in most situation).
Exactly. It's not about the total sales digital marketing generates, but the incremental ones it brings. It can be quite efficiently evaluated, so definitely it's NOT a scam.
Ok, read it. The article falsely assumes that performance marketing relies purely on the vendor conversion data. I don't know any business out there, which would do that.
The example given in the article of the planner ads isn't a good one. The author is trying to make the argument that performance marketing mostly unfairly takes credit for sales that would have happened away, but this particular sale, even if it had happened, would have been unlikely to have gone to the specific brand that was advertising. If they hadn't placed that ad, they wouldn't have made that revenue. Some other planner company would have.
(If you have a near-monopoly on planners, of course, such an ad would indeed have been a waste. The author would have come to you by default once he decided to get a planner. )
(So I do think that performance advertising is partly a racket, but it can work really well if you know what you're doing).
One additional thing people don't call out is that a lot of the budgets spent on these platforms are "learning" budgets. Agencies play this card really well. They'll tell you, "oh, you need to increase your budget, and test all these different combinations of ads/targeting/landing pages/etc so that you can learn what works (or the AI behind the platform can learn what works)". And obviously, in "learning" mode, you're ignoring the ROI.
I've seen people spend substantial amounts of money in "learning mode", and the platforms are kind of incentivized to make the learning less efficient so it takes longer and more spend for you to get to ROI positive (or to learn that you will never get there).
> the platforms are kind of incentivized to make the learning less efficient
Any successful platform is not, because by definition they have enough ads, and so want to focus on driving conversions for advertisers so that demand, and hence revenue, increases.
Agencies are often super shady though, on that we agree.
The example given in the article seems very off-base: the author is vaguely planning to purchase some type of planner, and he gets an ad for a specific planner from a specific retailer. There has to be a pretty tiny chance that he would have ended up at that exact site without the ad, so if the ad does lead to a conversion, it's incremental revenue.
It's interesting the OP appears to accept they're in the uncanny valley of "well FB knew somehow I was going to buy that one and showed me the ad".
For me it just made me think, well I'm doing something right because Facebook mostly offers me shitty white label stuff that's being sold using fraud (I've made complaints to Advertising Standards [UK] a few times about Facebook ads) and that I would never buy.
Not to mention, "you’ve just visited some articles recommending fancy paper notebook products" makes it sound like he viewed some spammy top 10 list full of paid affiliate links, so he's probably succumbing to performance advertising regardless of Facebook's later involvement.
Advertising as we know it emerged as a melding of government propaganda research and behavioral economics during world wars, so of course it’s often scammy.
It sounds “deep state” but it’s actually plainly documented in government files and written about by reliable sources.
Remember we’re still emerging from an era of whispering the same old story of morality and obligation to each other.
I am not at all interested in helping someone build a fertilizer empire or pillow brand. Politically my hands are tied to doing so if I want a life.
Poor people effectively live a life of quota and state sanctioned limits on their access to material support by cutting social programs with public support.
Advertising America as anything but a sanctimonious police state is a scam.
The "Century of the Self" documentary series about Bernays and the rise of propaganda/marketing is really great. Recommend it for anyone who hasn't seen it yet.
> Reliably someone comes along every few months to question [performance marketing]. I always come back to analyses of incrementality as the real proof.
> Take an audience of X people. Divide them in two. Show ads to your test group, don't show to control. Watch your business grow and gauge the lift between the two audiences.
The companies that know how to advertise at scale do this constantly and can gauge the real effect of their ad dollars. Facebook, Google and others make these tests possible in their platforms, while other software suites such as Impact Altitude and VisualIQ allow you to do this kind of analysis and testing as well.
> In the end, most of it proves out to be incremental. There are notable exceptions of course, but when are there not?
I take issue with the conflation of Performance Advertising with Awareness Advertising.
The whole point of _performance_ advertising is that it's effect is _measurable_. If AirBnB spent $500m+ on performance advertising, they should be able to trace that back to an exact amount of revenue. If you are a brand in this scenario, you can conduct split tests by sampling the conversion rate of users from advertising vs non-advertising. Again, it should be simple to see if the conversion rate for users targeted via advertising has increased or is unchanged.
In the branded search examples, again, as an advertiser, you can see what searches are associated with your leads. While you do have to compete for attention on your own branded search (to compete against competitors taking the slot), you should also be able to recognize unbranded terms which drive conversions for you. Again, assuming this is actually _performance_ marketing, you would be able to look at the cost of these placements and the ROI, and the impact would be measurable.
The rest of the article is largely composed of straw-man arguments that imply the results are not measurable, when in fact they are (if done right).
disclaimer: I'm the CTO of a performance marketing platform. The vast majority of conversions on our platform happen same-session. There's a very easy way to measure this effect -> pause your campaigns and immediately see conversions fall.
Reminder that Uber turned off $100M of their ad spend and saw basically no change in conversions: https://twitter.com/nandoodles/status/1345774768746852353. I’m pretty sure they thought they were working initially (through conversion tracking).
As you mentioned: The most important test (that you should be doing every now and then) is to actually pause the ad and see if revenue falls. Anything else is just an approximation and should be treated as such.
I think performance advertising is a great idea if you don't rank very high in search results for your category.
"Rental stays in XYZ city" will bring up an airbnb result all day, probably in the top five results. Therefore, paying $3 a click to be placed above your own search result is probably silly.
I happen to have a small side project and advertise it with a very low budget on facebook, google, and bing. It works. I don't rank very high because my SEO skills are poor, but google ads absolutely drive real and interested people to my site.
My side project isn't for sale or behind a paywall, so I'm not sure if they have their wallets out. But I know for certain that they're real and seem to be interested in my offering enough to click around and sometimes drop a message.
What if this article is missing the forest for the trees a bit? In my experience performance advertising is almost always paired with awareness advertising. The latter makes you aware of the brand/product/whatever then the former nudges you to act/buy/whatever.
So if you’re buying or even just evaluating performance ads without considering the bigger picture you might come to erroneous conclusions.
Take the Lego Movie example from the article. The $65 million movie is no doubt an awareness play. Could you make the case that you should also increase your performance budget to help capture more of the demand you just generated with the movie? Or should you just hope that people go from the movie theater to buy Lego unprompted? Is it worth it for Lego to advertise to people who walk out of the theater and search for “Lego Batman set” or whatever? I think so, even though evaluating such branded search campaigns individually might make them seem inefficient.
It seems very easy to dismiss the performance advertising as a scam when you evaluate it in a vacuum. As noted in the article it’s important (and very difficult) to understand the incremental outcome of any channel or campaign. That incrementality includes awareness campaigns.
After more than a decade in advertising and marketing I am now more than ever unwilling to accept simple or definitive answers to highly complicated questions. At best I hope that we can unwind some of the overall complexity so we can have a chance to trust some of those definitive answers.
The problem is it's not what's sold on the tin - the promise of pay per click advertisement is often that you can track the results of your spend more easily, and also that it can get buyers at the time they are searching for a sale. The article seems to contradict those two points at least.
see also "Consumer Heterogeneity and Paid Search Effectiveness: A Large Scale Field Experiment"
(Tom Blake, Chris Nosko & Steven Tadelis) in which EBay did the same experiment (they stopped advertising), and also "The Unfavorable Economics of Measuring the Returns on Advertising (Lewis and Rao) which talks about how hard it is to measure impact.
There are so many holes in these arguments that you could drive a truck through it. Which is so infuriating, because a big part of my professional career consisted of watching all of Rand's SEO videos and really appreciating them. I really thought he was a genius. But then over the past few years, he started sharing more general views on entrepreneurship, and those takeaways just didn't really make much sense. Basically, his own VC-funded company turned into a shit show and suddenly he started advocating against VC in general (as opposed to taking an honest look at the mistakes that his company made). So in the past few years, I tried to reconcile in my head: how can a genius make such imperfect conclusions? My initial takeaway was that he's blinded by his own mistakes and shifting the blame, which seems perfectly reasonable and understandable. Frankly, I would probably feel and act the same way.
But after reading this article, it finally dawned on me. He makes imperfect conclusions in everything he touches, it's just that in some fields those conclusions can be more easily proved to be wrong than in others. SEO is the perfect field where a polished presenter can get away with imperfect conclusions for years - trust me, I know, I made a living for years in this field, and I am very familiar with the nature of this work. Most of the time, you have no idea what the black box really does, and instead you're just trying to guess what might have happened. Most importantly, there are many ways to skin a cat in SEO, and just because your approach is net positive doesn't mean that you truly are delivering the global maximum (or that the net positive gain was ROI positive). In short, it's impossible to know who's right and who's wrong, and Rand's videos convinced me that he's right, but I am no longer sure. I just rewatched one of them, and can easily see how his conclusions are just... opinions.
While we may or may never find out if his SEO opinions were the global maximum, we can quantifiably demonstrate that his opinions on content marketing are not solid. This whole essay he wrote can be replaced with "hey performance marketers, don't trust the platform numbers and instead do your incrementality studies." Platforms like Facebook will give you those for free if you reach a certain spend level, and you can also get them from 3rd party providers like measured.com. In other words, if you're a performance marketer and you're not conducting incrementality studies, then you're very early in your career and are not following the best practices. Simple as that - no need to extrapolate from there and reach all sorts of additional conclusions (which is obviously a pattern in Rand's behavior) - calling into question a perfectly investable marketing channel, conflating the needs of a public company with everyone else's needs, using words like scam, etc.
I am really disappointed to have to write this, but you would have been better off not reading this article. If Rand is really advocating that the majority of entrepreneurs should follow his advice and focus on PR instead of performance marketing, then perhaps an honest thing to ask would be - how is that working out for his own company? AFAIK, SparkToro is nowhere close to replicating the growth of his previous company, which is honestly disappointing for someone with such a huge reach and name recognition.
Can I politely suggest that you share with us the holes you see in the arguments and how you see them? Whether or not Rand is any good at what he does or says will be clear by your expert disassembly of his arguments.
Rereading the post, I agree it came across as more personal than it should have been. I wish I was able to frame it more as a response to all the people advocating for alternative marketing ideas which frankly never work out. Instead, I brought in his own company into the narrative, which was below the belt. I know first hand how hard it is to get something to work, and how unlikely it is to happen time after time. I wish I could edit the comment, but HN won't let me at this point.
People on this board love the idea that anyone who isn't an engineer is an idiot and everything that everyone else does that isn't writing code is somehow worthless. In broad strokes, they think that all people who do people-facing work are charlatans that are constantly trying to hide how worthless they are because talking to people, etc. can't possibly provide value. Then, they (like so), ask you to break down your arguments so they can pin the tail on the donkey that is the logical fallacy that they will use to discount your (almost certainly correct and informed) opinion.
Unsurprisingly, those people are wrong and marketing works.
It's also fair to say that if you let the marketers drive business decisions it's a recipe for disaster. Pushing low-quality product out the door just to hit some quarterly sales goal, that's what marketers will push. Epic failures with disastrous long-term consequences are the typical result (Boeing MAX for example).
Actually, the 737 MAX disaster was driven from the business side. Somebody promised Southwest $1 million per airframe if differences training was needed.
The organization was too brittle to accept criticism from their eng. organization, as most are, until a crisis.
My understanding was that Boeing wanted to prevent Airbus from winning contracts and so pushed the schedule hard. That's the sales and marketing division talking isn't it?
- Brian's comments about Airbnb's approach to performance marketing are used to imply that Airbnb's lessons apply to other businesses. Very few businesses - especially those that start from small numbers and need to grow them - have 90% of their traffic mix come from repeat customers. Using Brian's comments helps fuel the narrative that brand marketing is a better investment than performance marketing, which is correct in some cases (Nike, Airbnb), and not in others (most startups). Also, as another member pointed out, Airbnb's performance marketing budget is still well over $200M/year, which no responsible/public company would spend if it wasn't returning a great ROAS. Finally, Airbnb is known for all sorts of marketing shenanigans in their early days, and they certainly can't take the credit for a pure brand play.
- Calling performance marketing platforms a scam (repeatedly, both in the title and in the narrative) doesn't explain how those same performance marketing platforms are carrying the majority of traffic acquisition in most of the B2C companies that went public this year (and practically all of the DTC ones). Calling into question the accuracy of measurement is one thing. Calling it a scam is wrong and designed to rank on HN rather than to be reflective of the true value of those platforms.
- As I pointed out in my original post, all you have to do is use incrementality studies and 98% of the criticism instantly goes away. Rand implies that you have to do your own studies (by eg, following Avinash Kaushik's methodology) which is 100% wrong - Facebook will do them for you if you reach a certain spend limit, or 3rd parties will as well with no spend limits. Also, from experience, this really becomes an issue once you spend meaningful amounts on two platforms at the same time. His rant on this subject has an iota of truth and a whole lot of sensationalism mixed together, and overall leads to wrong conslusions.
- He conflates "paid search" with "all performance marketing platforms", including "paid social." It would have been helpful to point out that the challenges with branded terms are entirely isolated to paid search and have nothing to do with paid social.
- My favorite sensationalist tactic: frame a strong accusation as a question. This way you get the clicks, but you can still cover your ass by linking to resources that with enough research would allow the reader to answer the question with a "No." But in lieu of that research, the implication is that the answer is a "Yes." You'll see this tactic used by less reputative media sources, and I was disappointed to see Rand do the same.
Your first point is very hopeful, but not what I've experienced. The bigger the company, the more slack for bad decisions.
Your other points I think relate to scale. No advice can be universal, and if you read the article as absolutist, your take makes sense. If you read it as "hey, your mix is likely wrong", a lot of the criticism fades.
I think we've lost a bit of creativity in marketing. The Lego movie example is a really good one. I think it is probably good this happened, as a lot of creativity was performative (how do I win an ad award/impress my peers) and not about increasing sales, but we've perhaps shifted the balance too far, and there is likely some areas with good ROS that are now better bets.
Scale brings both, headwinds and tailwinds. The outcome often hinges on how the balance of those two forces plays out.
Ecommerce example: every year the CPMs go up and your paid margin goes down. But every year you have a larger email list, so the balance of paid to unpaid shifts.
General example: every year you get more of the late-stage employees who care less and less about your company. But every year you can afford to pay for more layers of management, which will keep an eye on the underperformers.
This list goes on and on... The headwinds are driven by external forces, whereas the tailwinds end up working out based on your specific execution of the opportunities that present themselves to you. This is where an experienced operational team can make a huge difference.
> I think we've lost a bit of creativity in marketing.
You absolutely cannot rely on performance marketing forever. It's a shot in the arm until you have reached enough [fill in whatever you wnat] so that you can leverage that momentum to reach the escape velocity. So it's not good forever, but it's a great catalyst.
> Rand implies that you have to do your own studies [...] which is 100% wrong - Facebook will do them for you
So FB, who earns on my spending, offers me to measure for me whether my spending makes sense for me. Why would FB ever tell me to spend less? Don't they like money?
It's fair to question credibility, but you have to realize that their results can be compared against 3rd parties, and it would lead to a lot of reputational damage (and also legal risk) to systematically defraud your advertisers. All it takes is one smart team to out the entire enterprise.
The causal incrementality of ad spend is rarely negative. You expect to see decreasing returns with additional ad spend. For each dollar in ad spend, one advertiser might want to see $3 in revenue, another $5. Incrementality studies then allow advertisers to tune spend to their operating points.
Facebook does like money. Do you think lying to the biggest advertisers in the world is the best long-term strategy, or do you think instead it might be better to report out the most accurate results possible?
Facebook has a history of screwing advertisers ("want your page to show up in people's feeds? Get more follows, here's how to buy ads targeting that", then one year later "we've decided to downgrade pages in people's feeds"...), bankers (their IPO pricing), and essentially everyone they have a business relationship with. I'd expect them to stretch the truth as far as they can get away with.
There's literally a team at Facebook (Marketing Science), part of who's job is to tell advertisers what actually works and often, this can lead to them telling advertisers to spend less.
And this is an incredibly smart business strategy in that FB know (from experiments et al) that their platform works, and if they can show incrementality, then advertisers will invest more in the platform.
In game theory terms, it's a good strategy for a repeated game, which advertising definitely is.
To your point that AirBnB's and other global brands don't extend to most businesses, Facebook's recent outage provided such an experiment and indeed a lot of small businesses saw massive effects https://mashable.com/article/facebook-outage-small-business-...
Thank you for sharing this. Facebook received a lot of criticism for how they lobbied publicly against iOS14. Their reputation is ruined, and anything they say will always be taking with a cynical afterthought. But as much as it pains me to say it, they are 100% right in saying that iOS14 is indirectly going to create a lot of damage for SMBs and companies trying to disrupt the status quo. There's more nuance to it, but think of it this way: if your marketing budget plans on reaching 100m Americans each year, then the negative impact on micro targeting is not quite as bad since you're going for the mainstream customer anyway. But if you're trying to reach only 1m people, then losing micro targeting is a matter of life or death.
"Also, as another member pointed out, Airbnb's performance marketing budget is still well over $200M/year, which no responsible/public company would spend if it wasn't returning a great ROAS. Finally, Airbnb is known for all sorts of marketing shenanigans in their early days, and they certainly can't take the credit for a pure brand play."
My experience is exactly the opposite. The larger the budget, the less real hard analysis is done. This is especially true with the rise of attribution modeling which allows marketers to essentially motion blur the data.
The larger the budget, the more sophisticated the teams and the methodology. Also, more players eager to accuse marketing of poor performance (eg: sales and revenue teams). Not sure where you made that experience, but that marketing team would get fired within a quarter at any large scale company I am familiar with. Large companies are inefficient in many ways, but not in what pays the bills.
Many of the large companies will say "we have a complex sales cycle" (and they often do) and leverage multi-touch attribution models which removes the direct ROI calculation.
>Also, as another member pointed out, Airbnb's performance marketing budget is still well over $200M/year, which no responsible/public company would spend if it wasn't returning a great ROAS
That's not really an argument. You wish that a responsible/public company wouldn't do that but we have seen much dumber behavior.
> When P&G turned off $200 million of their digital ad spending, they saw NO CHANGE in business outcomes. When Chase reduced their programmatic reach from 400,000 sites showing its ads to 5,000 sites (a 99% decrease), they saw NO CHANGE in business outcomes. When Uber turned off $120 million of their digital ad spending meant to drive more app installs, they saw NO CHANGE in the rate of app installs.
Like, this is probably true for advertisers who have a well known brand. One would expect incrementality to be lower, because everyone's already heard of the brand.
However, if you are a new entrant/small business this is not the case, because nobody has ever heard of you.
As an example, look at TikTok. They spent insane amounts of money on FB ads to get as many installs as possible, but I'd suspect that they don't do this anymore because they've got enough brand equity that it doesn't make as much financial sense.
That was an incredibly misleading article. If you check the links, you find the following:
1. P&G didn't turn off that spend, it shifted it to other marketing platforms.
2. Chase didn't change its marketing spend, it just concentrated it from 400k sites to 5k sites.
3. Uber found out that their agency was committing fraud and AFAIK the case is still being litigated.
I hate being that guy, but check the links. Those reporters really must have an agenda or something (or are just struggling to get the clicks, so they need to make a story out of nothing).
To 1.
"P&G’s $200 million digital cut were reinvested into areas with “media reach” including television, audio and ecommerce"
So they awitched to other platforms with mostly no back channel and no user tracking. So the whole tracking seems to be pointless
My point is not exclusively advertising but that it is wishful thinking that companies always make sensible decisions. Often enough, useless or even harmful things continue to be pushed through, simply so that one doesn't have to admit to oneself that one has made a mistake.
They would rather try to achieve a positive result with larger investments after all, which usually fails and leads to even larger losses, than to exit early and get away with smaller losses.
Therefore, the statement that a reasonable company would not act this way is not covered by reality. Companies should not act this way, but they do.
Hmmm. Marketing works!? Well having been a sucessfull .com era ad co. founder, I suceeded in raking in the cash of the uniniated (automobile, etc, so no tears) only to see companies literally obliterated by marketing. TV was then still 20 times the spend. Still, marketing had 0 clue. Fun to drink with at the theatre, sure. I was CTO, so, shrug? No, I was incensed and tried to talk other managers client side out of giving ME money. I know some clever people in marketing, but when they smell budget, it's a movie shoot in the sahara time.
I do think things are better in some sense today. But it is a hungry beast, marketing. Oh, and lies and statistics.
You linked to an article that offers no guidance on how to actually do it, which is more indicative of the article itself than of the methodology. One of the vendors in this space has a good amount of content on this topic, which you can find at the bottom of this page: https://www.measured.com/. Dislaimer: I have nothing to do with this vendor or the industry in general (apart from successfully spending money on performance channels).
My apologies I should have prefaced it with "for others confused as to what incrementality is all I could find was ..."
It's a surprise to see a vendor have a clear explanation on their site however - and it's worth quoting i think
>>> Incrementality testing creates an experiment that systematically withholds media channel exposure to a representative subset of users (the control group) while maintaining normal media channel exposure to the broader user set (the test group). If the control group is both sizable enough to be statistically significant and selected at random such that they are broadly representative of the user base, then the media channel’s incremental contribution can be determined by the difference in business outcome (conversion, revenue, profitability, etc.) between the test and control groups.
I worked on several web search engines in the 2000s. During that time I ended up next to someone who was in the SEO business a handful of times at dinners or meetings. When they explained what they were doing to me I kept thinking "this kind of explains how belief systems and religions are formed".
You would have these very complete theories of how search engine ranking worked - completely disconnected from reality and unburdened by any actual knowledge. If there was ever a cargo cult, SEO was it.
I still cringe when people say that they have "done SEO" on their site while I'm in the room. It is kind of like telling their mechanic they've sacrificed a goat so their car will get better mileage.
What's wrong with saying someone's "done SEO" if everyone at the table has the context for what that entails?
Content strategy, keyword research, Search Console integration, site speed improvements, local search business profiles -- all of these contribute to SEO. It's unwieldy to talk about them all, so sometimes people can use a shorthand.
I cringe because usually they will be talking about things that come out of what can only be called superstition and they tend to think that's it. Meanwhile, I would be sitting there and knowing a great deal about exactly what goes on when whatever web search engine i was working for at the time actually did, and not being allowed to discuss any of it.
I'd always give the same advice. Don't try to outsmart the search engine, focus on quality and consistency over time. That will always reward you in the long run. Try to game the search engine and you will find yourself in a weapons race you can't sustain over time. If you find exploitable tactics to get better ranking than you deserve, sooner or later you'll probably be pushed down the ranking. And this is really important if you're successful enough to have something to lose.
Some of what people label SEO is just "make good web sites", and that indeed works. But some of it is just pure rubbish. And making a good website isn't just "doing SEO" - it is making a good website. And a lot of people don't.
Which is why I can't take people seriously when they present it as some singular task they have completed.
My take on this is that its sort of like the prisoner's dilemma in that if everybody said "We'll not spend anything on ads[1], let the organic results speak for themselves" then you might get one result, but anyone who has been de-indexed from Google knows that having no results on the front page means a huge drop in traffic for you. And if your business depends on web conversions well that is a pretty measurable loss of revenue.
Take one product. Make a thousand different businesses who sell them. In one ad, Britney Spears is holding that frying pan. In the other, it's the Undertaker.
Instead of product and customer, think value added. You buy a server for $0.065 an hour. You buy traffic at $0.01 a click. That's your raw commodities. How do you make a profit off of that? Supposedly if you're good enough you can get elected president. I'm sure there's room in there for you to pay your bills.
I learned of a bro through a targeted advertisement and I am looking at it for my next rental. I don’t believe AirBnB’s conclusion can be interpreted as “advertisement is irrelevant.”
I was at one company several years ago and made two big changes to digital advertising:
#1 - Eliminated all paid search other than some limited branded search terms and shifted all the money to affiliates who were way better at making profits on the keywords we were competing on
#2 - Eliminated all display advertising after running numerous experiments showing it provided almost not incremental conversions, even though the platforms happily took credit for them.
Those two things drove our blended CAC down substantially and by building better affiliate relationships, sales actually increased.
The lesson here is that you need to try a lot of things out and you should be continuously questioning what you're doing and running specific experiments to gut check effectiveness of any ad platform that is slapping cookies on wide groups of users and claiming conversions.
My suspicion is that this is near impossible at any large organization, even one as new as Airbnb. I can just imagine someone walking into a team of 20+ performance marketers and suggesting they need to experiment to determine if any of it is remotely effective. COVID forced them into this but it's something that they should have already been doing.
I also suspect that the top line focus/obsession of most VC-backed companies make this type of exercise seem almost counterintuitive. Don't mess with or question the momentum.
Maybe your situation or industry is an exception, but my view is that affiliate marketing is one of the most corrosive influences on today's internet. It drives a truly mind boggling amount of spam of all types, from affiliates who add literally no value other than inserting their affiliate code into the transaction. The proportion of content on the web that provides no true new information, but exists solely to drive affiliate traffic, is surely massive.
So, it's great for your business that this worked, but I personally don't see any strategy that leans on affiliates to be worth celebrating.
A lot of things can fall under the "affiliate" umbrella but what we did was doing 1-on-1 deals with content providers who were acquiring a lot of relevant traffic. It was not promo code spamming.
#2 is basically an industry standard for ad firms (at least good ones), you'll usually only get people buying those ads to fill budgets or to do branded advertising that's only purpose is to raise brand awareness. I don't really agree with your #1 point, but that really depends on your vertical. You couldn't really do that well in service industries or financial/legal services, etc.
In my example, I saw a lot of competitors try and fail at acquiring customers profitably. You could see the ebbs and flows of them trying as well as the well-financed players dominating spend. When the industry numbers came out, things never lined up.
That said, I’ve worked at brands where paid search and paid social are incredibly valuable channels.
Performance marketing is essentially the only way most startup marketers can go. The reason is that they generally have 6 to 12 months to prove themselves as performance marketers. Generally the CEO has sold some insane growth model to the VCs. As a marketer, you do the math: $ growth required/ASP = number of deals. You get to your lead targets by reversing the funnel (Closed Won -> Opportunity -> Lead). And you see you need a sh*t-ton of leads to ever get close to hitting that number. And you need them quick because the number just bigger every quarter. So you can't even think about long-term approaches - you have to grab at tactics that you know can maybe generate the number of leads - even if you know in your heart-of-hearts that they won't generate anywhere near the number of opportunities.
This is all just my opinion and I'm sure there are better marketers than me that make this work, but this is what I've seen happen in too many VC-funded/backed companies. And it gets even harder when we're at the super levels of funding that we're now seeing. So if I'm a relatively small company (funded to $50m) that is spending $5m a year on marketing, and I'm competing against a company with $500m in funding and, say $100m on marketing, then it's even worse.
Yup. Scaling marketing without accurate LTV's is normally a terrible idea. I've seen so many startups go bust because of mistakes here (especially if you have a long sales cycle/top heavy revenue (like most f2p games)).
But, as you said, the incentives point towards terrible outcomes, so this keeps happening.
The article does have a point about attributed sales vs incremental sales - as a CFO (whether I qualify as the "hard-nosed" type in the article, I don't know), I'm bugged every time a marketing guy starts talking about how this campaign has been/will be "ROI-positive", and have had a few heated discussions on why this is mostly not demonstrable until you're willing to pull the plug (which you can do with more or less intelligence in order to minimize your risks).
On the other hand, while I do indeed believe that the "ROI" from Performance Advertising is something between just false and deliberately misleading, the bigger picture that I'm interested in is marketshare. Because when looking at market share, it's not a question of incrementality anymore, but whether you're growing slower/faster than your competitors, and your cost of doing that, and at what point you're OK to 'buy' marketshare, in the sense of losing money in the pursuit of growth, and how much. And then, OK, let's talk about ROI on that basis - most of the time, achieving this will indeed require tools from the Performance Adversiting toolbox, which allow you to conveniently track the amount of marketshare (i.e. sales) you bought.
368 comments
[ 5.3 ms ] story [ 301 ms ] thread> [picture of that search term and williams sonoma ads with shopping links]
The main problem here is that if Williams Sonoma was not advertising on that search term, Lodge and Food52 etc etc would, and then those companies would be above the Williams Sonoma organic placements.
The spend is necessary in a defensive way because Google creates a bidding war even for the hyper relevant.
edit: I just checked and if you search "williams sonoma skillet", if WS was not paying for [green] then the very first "result" (ad) would be Food52 [red] https://imgur.com/a/9Nnxs6h
I just tried "airbnb paris" and the first result is, somewhat predictably, an ad that is not airbnb. But the second one is also an ad, this time from airbnb. So they clearly didn't keep their spend dialed down to zero, and are aware of the need to advertise on their own keyword.
It's the fundamental contradiction in Google's search ads model. If Google delivers users the thing they want, ads by definition have to be things users don't want.
As a Google search user, why wouldn’t I value these? If Google doesn’t serve them and another search engine does, I’d be inclined to switch to the other one.
But why would those sites not have the "organic search SEO juice"? If those sites are actually good places to buy that kind of thing, a good search engine should direct a user searching for that thing there.
What wouldn't show up is lmm-cookware.com, launched early October 2021, who is trying to establish themselves as a new destination for cooking enthusiasts. That domain is new, the business is new, no Google reviews, but they've got the pan in stock, are willing to sell it for $30 shipped, and have an advertising budget to reach customers.
As a consumer, I want to be able to learn about that offer, make my own decision where to buy the pan, and I'm happy to use a search engine who will show me that offer. lmm-cookware, I, and the search engine all win from this outcome. I don't care whether lmm-cookware has been a good place to buy pans such as these for the last 180 days; I care whether they're a good place to buy it right now.
Why? How websites are ranked "organically" is somethine that is up to the search engine. A user-focused search engine would have no problem including factors like price (for shopping sites) that the user cares about. An even better one might provide the user with options to refine the ranking criteria for each search.
Meanwhile you keep assuming that somehow the site willing to give you the best deal will also be the one winning the advertising bid. That makes no sense as they are also the site spending the most on advertisement which they have to recoup somehow.
> As a consumer, I want to be able to learn about that offer, make my own decision where to buy the pan.
You seem to be under the impression that you would be missing out on an offer without ads. But if you only look at a finite set of links you are always missing some offers - ads only change which offers you see. And they do that not based on any judgment of wheter it would be a good offer for you but only by how much those sites paid.
In terms of winning the SERP paid ad bid, I don’t care if the search engine shows me the ad that’s best for them, because that will still give the newer business the chance to win the bid. If they choose not to, well, in that case I don’t see their ad. If the search engine doesn’t have ads, then in all cases I don’t see their ad.
If they go completely insanely rogue (like showing me a paid mesothelioma ad regardless of my search term), they either lose my future search traffic (“foosearch never has what I’m looking for”) or they lose the advertiser (“mesothelioma ads convert well for us on barsearch but not on foosearch; I need to lower or pull my bids on foosearch”)
> What wouldn't show up is lmm-cookware.com, launched early October 2021, who is trying to establish themselves as a new destination for cooking enthusiasts.
If a new site is what users are probably looking for, then it should be top of the organic results. Google and other search engines already use signals like when the site was updated, whether it has a social media "buzz", or what price it's selling something at. It's highly unlikely that spending advertising money is a good signal; yes, promising new companies do spend advertising money, but so do established companies and out-and-out scammers.
Helping the user discover new sites is maybe part of what a search engine should be doing. But mixing ads into the search results is too intrusive and anti-user a way to achieve that (if it even does). Maybe there's a time and a place for advertising, but it should be clearly separated from organic results; Google used to be good at that (indeed they were famous for having advertising that was less intrusive than their competitors), but they've been getting steadily worse.
I do actually think you have a good point.
If you really like Skillet X, and I ask you which skillet I should buy, and you tell me, "Skillet X," then it is not an ad.
If you really like Skillet X, then somebody gives you $5 to recommend it if anybody asks you what skillet to buy, and then when I ask, you tell me, "Skillet X...I mean you should know that somebody gave me $5 to say that to you but honestly I was going to say that anyway" then is it an ad?
It kinda intuitively feels to me like if it doesn't alter the result, it's not an ad, it's just somebody taking advantage of another person's willingness to hand them money and doing nothing in return.
In reality, Google is not perfect and you can argue that ads do provide value by promoting relevant content, even if its gameable by our capitalist system.
The definition: "If Google [meaning Google's organic results] delivers users the thing they want"
The consequence: "ads have to be things users don't want" because advertisers whose product is wanted will be included whether they pay or not.
Now, you could widen it to say, look Google is still delivering the thing users want, they just sometimes do it by organic results and sometimes do it by ads, but that's a very convoluted reading of the definition, which clearly is using Google as a stand in for "Google's organic results".
The problem with my argument isn't that it failed to be a tautology. It's that the definition is questionable, and that's the profitable angle of attack: "users don't know what they want" "there can be multiple equally good options" etc. But saying it's not by definition is just silly. The definition is the whole thing up for debate!
Ben Evans has a tweet something like "half of facebook devs are working out how to code the algorithm to serve you just what you want to see, the other half are working out how to get the algorithm to serve you what advertisers want you to see. "
We are doing something wrong
No, this is the outcome of all interest targeting based models. You're interested in X, and they either don't show you X (to get X to pay) or they show you Y (which you aren't interested in) because Y paid.
Whereas with demographic targeting, hey I'm in demographic X and wasn't looking for a Y now, but sure, I'll remember it for when I'm in the market for one later.
Google's original values were the ability to provide better search (common answer) and be fast (less common answer) - both of which were a complete contrast to the Alta Vista and other search engines. I could easily see Google facing disruption from a new player - but I don't think it will be another search engine. Probably a paradigm/systemic shift.
Now you could argue that the search engine should show organic results to help the consumer get the best deal, and I would agree with you.
Bidding on a competitor's brand name keyword should be banned. But Google can't resist double-dipping.
Where I worked before, Google could get $4 for ever $1 we made. And we actually delivered the service, and people googled our name. Pretty crazy...
If I as a customer am using such specific search terms, then I would assume that my intention is to find and possibly buy this specific product. The results of other brands might be annoying, but why should I click on them? The relevant results are still displayed on the first page.
People will typically write their intent on the search engine even when they could simply directly to the website.
Case in point: The top 10 bing searches are for websites, including FB, Google, Youtube [1]. This traffic is highly competitive and should (as in all competitive markets) be bid among competitors.
https://ahrefs.com/blog/top-bing-searches/
Small business people in my area of UK have always done this, type in the box in the middle (usually Google, occasionally Bing or some other service). But my pretty tech-literate kids do it too, even when I show them how they 'should' do it and that is faster, and they don't need the extra click to get where they're going ... mad!
On Chrome on Win10 as I use at work though, typing in the address bar, with my settings, I get auto-fill of addresses (the history search is noticeably missing vs Firefox) including the option to use 'tab to search' on a domain.
And beside that a lot of people routinely misclick or click on ads not understanding that they are ads even when they are marked as such.
But my main point was that the brand may not be all that important in the first place even though it was included as part of the search term. The person making the search could’ve seen an ad and been intrigued by the product, but upon landing on a competitor site they may choose to buy a similar product from them instead.
If you are deeply into the kind of thing you are buying, you will make a lot of research to find the best one. But there are a lot of things we buy that we don’t care as deeply about, and where we may choose the first one that fits the bill sufficiently well.
Doesn't that just make for effective advertising? I'm failing to see the harm here.
So much this on phones. I think I missclicked adds like 10 times in a year, before realizing there where adblockers on Firefox mobile. Zero legitimate clicks. Never happens me on desktop, unless there is a screen jump scam.
If you've never heard of Product Y, you might be intrigued and click. Maybe you want something less smelly!
I've noticed the quality of google search decrease drastically over the last 15+ years or so. I don't think that's directly tied to ad buys though.
At this point, Google's first page is so bad that I can almost build an anti search engine. Search for a term, and then exclude all the sites on the first page from ever showing up permanently ever again.
runnaroo showed things could be done better. The problem is that doing better doesn't seem to convert to profit.
If I search for a Dell computer, no way will I buy Apple.
Likewise if I use Bing to search for Chrome, no way will I download edge except for user error.
"Dominos" might be one where people wouldn't mind ordering from Pizza Hut as they're using it as a generic term for pizza.
Branded search terms are almost always less incremental than non-branded (ie: "lodge logic" vs. "cast iron skillet"), but the actual incrementality of the terms is something every advertiser should be testing continuously.
And if they were not above organically they would simply buy the advertising space that William Sonoma purchased. It's one of the slimiest things Google does - allows competitors to purchase advertising space on a query specifically crafted to find a particular source. It's nothing more than a shakedown.
https://twitter.com/jasonfried/status/1168986962704982016
The problem is that this is subjective and would need to be automated somehow. I think Google's original sin here is making the ads look so much like organic search results. Someone placing an ad against a competitors brand name would not be a huge issue if organic results were still front and centre like in the "good old days" of early 2000s Google.
Or Google, one of the wealthiest companies on the face of the planet, could hire some staff to handle this.
More generally though, if you paid for an advertisement and the natural search result has you first anyway, then you should not be charged for any clicks to this advertisement.
Of course Google wouldn't give that data out as then many companies wouldn't need to advertise at all to get top billing.
The simple solution is that your paid advertisements are free if it gets clicked when you're the top result anyway. That way it doesn't cost the brand any money to bid on advertisements for their own brand.
This is what is known as "on brand" Search ads. I like to call these effectively the "Google Tax" because publishers/retailers are forced to pay Google for the traffic they would have already received had the ad not been there.
I've seen way too many companies look at their analytics and say "see we get 20x ROAS on on brand! why would we turn it off?!?". Because silly, people are already going to go to your site without you paying for the traffic. I wouldn't be surprised if 25% of Google's ad search revenues come from this.
I do wish there were some regulation such that if a user is searching for a trademarked or copyrighted term, that the best organic search result for that term should be required to show up first. I'm fine with showing competitor ads, but I don't think they should be able to show up above the trademark owner's ads.
I realise that would probably be an impossible argument to win under current law (IANAL), but it _feels_ so very close.
https://addons.mozilla.org/en-US/firefox/addon/ublock-origin...
https://chrome.google.com/webstore/detail/ublock-origin/cjpa...
Each ad is given a quality score, and since your own website will be guaranteed to be the most authoritative source for a keyword search with your own branded keyword(s) in it, by default the quality of your own ads will be much higher than your competition.
This means in practice that the cost for a brand to be shown first on their own branded keywords is much lower than their average CPC, let alone the cost for their competitors to be there.
It will get worse, to the point where to get auto-completed in the omnibox you'll have to pay. TBH I am not even sure if that isn't already the case!
Not that that's the best option, but we could consider some new ideas.
Need to build consent for a war with <Country>? Give me access to the search index, and a few weeks, and it'll be done.
I trust that a multi-national serving multiple markets is going to give me less nationally biased information than the government of a superpower (with political ambitions that span the globe.)
It's why I generally don't turn to the Voice of America for news.
You think Google search doesn't adjust its search results based on your geography? It absolutely does, and that's just one of hundreds of inputs into its completely inscrutable machine-learning driven tangle of a ranking algorithm. Your trust is very badly misplaced if you think that Google can't be/isn't already corrupted by the corrupt governments it must deal with to do daily business.
"Google blocked every one of the WSWS’s 45 top search terms" (8/2017)
https://www.wsws.org/en/articles/2017/08/04/goog-a04.html
"Google admits to censoring the World Socialist Web Site" (11/2020)
https://www.wsws.org/en/articles/2020/11/04/goog-n04.html
"In a statement before a Senate hearing on October 28, Sundar Pichai, the CEO of Google’s parent company Alphabet, admitted that the dominant online search company has censored the World Socialist Web Site.
At the Senate Commerce Committee hearing, when asked by Republican Senator Mike Lee of Utah to provide the name of one left-wing “high profile person or entity” that has been censored by Google, Pichai named the WSWS."
"Google refines its search results to curb fake news" (4/2017)
https://www.engadget.com/2017-04-25-google-tweaks-search-to-...
(the comments below this article are funny because it's full of liberal-aligned comments on the disproven Trump-Russia conspiracy theory that is still given life in mainstream outlets)
most of all there is no public value for it when the government can legislate operators
It's called Quality score (soon to be discontinued) and has 3 components. Only one of which is landing page related. And nothing prevents competitors from building "more relevant" landing pages than the original brand.
"I need a lawyer in ___ right now". That lead can be worth a lot.
"Looking to buy a new car today", etc.
There surely are browser extensions that click on every link, just to make the companies pay, and make advertisers look even shitter.
If ads are no longer profitable, then this should sink both the online advertising market as well as Google.
Lemme guess, you’re in ads?
If I convince 10 people to do this with me for a day, we could cost them $100,000. We might be able to cost them $1M in a week.
Very, very sophisticated systems exist to prevent spoofed identities from counting as ad clicks.
Edit: I mostly want to stress that there’s a huge body of (largely proprietary) work on this topic. Thousands of skilled people have been full-time employed for decades thinking about exclusively this topic, on both the offensive and defensive side, in a continuously escalating arms race. Anyone just getting started on this topic has a couple of decades of literature review to catch up on before they should imagine that their proposals are novel. I don’t mean to discourage you if it’s a topic you find interesting, but you should bear in mind that the answer to “But has anyone thought of XYZ?” is “Yes” for basically all values of XYZ. There’s a lot of money at stake if the answer is no.
I mostly click on ads as chaff (mostly I block them but sometimes unblock them just do do some random clicking in the hope of confounding some profiles being compiled on me)
Isn't this just the usual problem with advertising? You have to do it because the other players are doing it. If nobody did it, it would still be the same cake to be shared.
This case is more on the nose, but only because of some fairness assumption that "Williams Sonoma Cast Iron Skillet" _ought_ to be traffic for Williams Sonoma.
Not quite: OP's point is about searches that specifically include your brand name. The "usual problem" with advertising in the zero-sum sense you propose is for eyeballs in general (billboards etc). The google tax here is more invidious, being specifically about searches that include your brand.
"Williams Sonoma Cast Iron Skillet"
and
"Cast Iron Skillet"
I think the important thing here is that companies don't decide to increase total advertising budget just because of the existence of a new channel. They have a budget of X. They have to decide how to break it up. Google, with brand targeting in search, is making a play for money that would've otherwise gone to Fox or ClearChannel or whoever. The brands redirect some budget to this new channel, as it shows its effectiveness.
Many of them still spend some on traditional channels, to get that first appeal and be the "Williams Sonoma" in "Williams Sonoma Cast Iron Skillet." Many of those people aren't just going to click through to Lodge ads instead. But less than they did when traditional channels were the only game in town.
Same dilemma, though:
If nobody advertised in Google Search or non-Google channels, people would just buy whatever skillets they found in whatever stores they found, or what their friends told them about.
If one brand advertised in offline media, more of those people would buy that brand.
So all brands advertise in offline media. And then it comes down to effectiveness of the campaigns + the same criteria that it would've otherwise - availability/placement, word of mouth, etc.
Then, if only one brand advertised in search targeting other brand names too, more of those people would buy that brand.
So all brands advertise on their brand keywords in search.
But ultimately it's the same game in both places. With the same net, Google just captures some spend that what would've gone to non-Google places previously.
Now, if you searched on the generic ‘cola’ or ‘soda pop’ then yes you expect to see those who bid higher to be at the top and at the bottom those who bid nothing unless organically somehow they ended elsewhere.
But this line of reasoning begs the question. If this system wasn't in place, then Google search would not exist as it does and the search traffic would not necessarily exist in the same way, no?
yes and no. Google search didn't have as intrusive and competitive ads as it did before and the search function was effectively the same. Nowadays, if you search "Williams Sonoma Cast Iron Skillet" you get 5~6 ads and 2-3 scrolls to get to the real organic results. 6~8 years it was like 1 ad and you could see the first result.
Another reason why ads for "Williams Sonoma Cast Iron Skillet" exist are not just to combat Williams Sonoma competitors but also scummy SEO search engine spam, I mean what if you don't pay to have your stuff in the first results and some hustle comes along that makes fake product review sites off of scraped results and engineers getting their stuff all among the first results, since UX research shows people not finding a relevant hit in the first few results do not go to the next page in results it implies that ads are required to give the correct research for something obvious like this because otherwise you have to take a chance with the present day subpar google results.
But if you are implying that brands should not invest in on-brand search campaigns, then this is a really bad advice. It's a known fact that targeting your competitors' branded terms is ROI positive, which definitionally means that the affected brand is unable to capture all the customers who were initially searching for it.
I'll be more explicit - you should test the difference. I usually recommend companies do a blackout month where they turn off all programmatic ads and then do a like-for-like comparison.
To further emphasis my point before, I've seen ad ops agencies say "hey you've got an overall 15x ROAS" and then find out that 80% of their ad spend is on-brand (give them 25x or whatever) and the off-brand (which is 20% of the spend) is giving them 5x. So their ROAS is inappropriately distributed to try to reach an overall ROAS goal.
Any other medium would theoretically have the same problem: if Ovaltine doesn’t sponsor kids’ radio shows in the 1950s and someone thinks they can deploy capital to grow a competitor, that someone will buy that slot. People couldn’t do this because there were human processes and relationships slowing this marketplace down. The thing that Google did was make it possible to test this at small scale.
That's an entirely different advertising model and market.
It's the entire business model of aggregators and intermediaries and middle men, which compose roughly half the world's working population.
Is this appropriate behavior by the taxi driver?
Would a regular person consider this deserving of punishment?
Is Google's search results advertising bidding system ethically equivalent to this?
The reason this happens is because the Google result shows my words in the excerpt, but when I click in, there's zero percent of the content to be found. That feels like a bait-and-switch akin to the taxi analogy.
My relevant results now either don't exist, or is in an absolute sea of mind-bogglingly bad results that it wears me out such that I would rather have had no results to begin with.
In this sense, I feel like Google is absolutely purposefully showing me results that I didn't want to go to.
This happens in real life all the time. Cab drivers get kick backs from specific places they have relationships with if they deliver people there, so you have to wonder whether they're actually recommending a better place for you or a better place for them, just like you do with Google.
And yes, I've been referred to a place by a cab driver when asking for a recommendation and found it didn't actually serve my needs at all, and in retrospect it was obvious from the conversation I was steered there without care for what I really wanted.
Google is more like the following.
"Hi, I would like to go to Morton's meats"
"Sure, I can take you to Shelly's"
"No, I said Morton's"
"Oh, the steak house downtown. Sure, I can take you to Billy's"
"What? No! Morton, you hear me"
"Ah, yeah, Morton, here you go.... by the way, the other places paid for ads"
Much less transparent, much more obnoxious.
I’ve seen this too, but I’ve always assumed it has to do with meta tags or something.
So if the User Agent is Google, the website produces relevant-sounding content with no JavaScript spam? I never realized GPT-3 could be used for this. The next time I see a bait-and-switch website in my search, I'll try changing my User Agent to Google.
I believe the signals from the safe browsing service are used to affect search result ranking, but I have no idea if they look for different content being served to the Google crawler.
Nobody forced you in that free taxi, you got in on your own. Use another service if you don't like the cost of "free" (ads).
What Google is doing is more like taking your destination, then asking "hey, you might like this other steakhouse more, want to give that a try instead?" before driving. Potentially annoying if the answer is always no, but... the answer isn't always no.
That sounds like it only works if my time that they wasted for selfish gain is valued at $0, which is obviously true as far as they are concerned, but certainly isn't true to me.
(Also, I did not intend to imply you were supporting or defending dark patterns — my apologies.)
That’s what the ad does. It adds a bit of cognitive load where you have to view and then ignore the non-William Sonoma ad but ultimately, and without much fuss, let’s you go off to WS.
> ethically equivalent
No. Google ads are fine when seen this way. The can driver’s behavior is more egregious if their decision to take a detour isn’t communicated to you AND runs up your fare by enough to materially affect you, whatever that level may be.
Disclosure - I’m actively trying to un-Google my life so definitely not a fanboy.
How much is Google charging you for your searches? I get them for free. Sure, sometimes it's a pain to scroll down past the ad results when they're not relevant but it only costs me time and attention.
If the cab ride was free sure, but you are explicitly paying for a cab to take you to your intended route.
HUGE difference.
All too often the middleman is milking both sides.
Google has value due to the search engine capabilities. But people realising they are being milked changes the value proposition.
I think Google is safe for now, though I don't think that is a good thing.
By definition a middleman is milking both sides. Were a middleman not there, it would definitionally be better for both parties. The only case where this is not true is the one where the middleman makes no money
For example, do you buy off Amazon or NewEgg, Best Buy, etc or do you just use Craigslist, Ebay, the local paper, or even the local flea/market swap meet for electronics? You can probably find stuff for half as much or even less at some of those locations, but it might cost you a lot in research time or lost money in the end...
To rephrase, all too often middlemen are there because of legislative/financial props from government or other incumbent advantages and - as you say - both sides would be better without them.
In this case - as in many middlemen cases - Google provides value - their search services. This "value" is the hook for their place in the transaction.
If Google provided a useless list of random results for every search term then it is obvious they would lose their place.
But where is the crossover point? Where Google doesn't provide enough perceived value compared to their perceived cost such that the market abandons them?
Peoples growing dislike of ads and - from this article - advertisers changing perception of ad value will change the cost vs value equation.
But I still think Google is safe for now.
Compare: https://ycharts.com/companies/GOOG/profit_margin https://ycharts.com/companies/AAPL/profit_margin https://ycharts.com/companies/FB/profit_margin
Of course, manufacturing companies have a lot less: https://ycharts.com/companies/GM/profit_margin
Especially with government websites. Lots of shady companies use ads above the actual search results to drive people to their site. Charge people for a service that is free.
Defensive branded search is generally only a fraction of budget anyway. I don’t find this practice the least bit onerous or extractive from Google — they’re just allowing for competition, as they should.
But in marketing, it's somehow fair game?
I want a WS skillet, and the first X results are for lodge... I don't need a lodge, I don't want a lodge, i want a WS, and google is not showing me what I want.
If I search for my bank's app, the top results on Apple's app store are apps for competing banks, which are pretty explicitly not what I want.
So I click on the Toyota ad cause it’s on top, and buy a Toyota? Which is taking business from Ford ?
What is the answer, no ads hmm that’s not happening on a free search engine.
Mark the ad as an ad hmm done
Don’t show other results for any brand related keyword search, when all keywords are blocked probably nothing left to show
What should the ideal free search engine show when I search for “intel vs AMD”, should intel complain the first result is for a tech site.
I do genuinely wonder if the conclusion that the brand company lost money is valid, I do wonder if those searches are like ppl putting cnn in a search bar or generic searches
imagine you are driving to a specific restaurant see a billboard ad for another restaurant so change directions and go there, would you dislike the bus/building with the sign
And wouldn't any users who end up buying a Lodge pan have been legitimately converted by effective lodge advertising? I don't think anyone who's only interested in Williams Sonoma will just go ahead and buy a Lodge pan. Unless you're suggesting that simply because the user entered "Williams Sonoma" in the search bar that page somehow "belongs" to them, which seems a bit absurd.
When you put it that way this sounds like racketeering.
Personally I resist the idea that a brand owns my attention because I used a keyword, but that's one of my many personal quirks.
I struggle to think of a meat world analogy. It would be something like if I dialed my girlfriend up on the phone and, instead of routing me straight to her, I had to navigate through a switchboard asking me "How about talking to these sexy singles in your area instead?" And in order to prevent this, my girlfriend would then have to pay the company to route my call straight to her.
Of course this isn't a perfect metaphor because there's a lot of different ways people use a search bar, especially now that search bars are merged into URL bars. But that sort of gets at what it is about this that feels sleazy.
I think it also exposes the core problem. The metaphor rests on knowing intent with certainty. It's perhaps possible that certain clarity and pretty much inferring might not always be the same, especially with how search and URL bars have merged.
But I understand completely. If you genuinely feel like you know that person's intent with certainty, someone else having a crack at their attention along the way feels like a violation of your relationship.
What kind of keyword are you referring to here? Because brands do own certain keywords, they're called trademarks. If you don't want a brand to feel entitled to your attention, don't include registered trademarks in your search query.
(Yes, I know, trademarks are limited to a specific market segment. Doesn't invalidate the basic premise of searching on trademarks though).
Opening google in incognito mode:
If I search for "Toyota RAV4", the first (ad) result is "Hyundai Tuscon". If I search for "AWS Cert", my first (ad) result is "Microsoft Learn". Et cetera et cetera :|
This is the real price of a constant threat to Net Neutrality, and allowing one monopolistic company to dominate mobile devices, web browsers, search results, and the largest video service on the entire Internet.
Their plan to corner and manipulate what everyone's freedom of choice and to secure their funnel of permanent revenue is considered cute to investors, but no one realizes how bad this will get in 5 more years.
Of course the numbers are fudged when you consider how they've turned analytics on their once very useful platform into a confusing mess, and when they announced that they were going to retire the system after it has killed off competition, because they can simply gather any analytical report they want privately from their web browser.
Public front-end statics are no longer trust worthy because they can be manipulated to drive platform revenue and engagement. The best and most accurate stats are provided only internally, to executive leadership that owns platforms.
Because we now use them for email, video views, browsing, phones, etc, they have key insight that can even be used for corporate espionage, your ideas can literally be beaten to market because your virtual assistant caught you mentioning keywords then reported you applying for your patent and corporate loan.
Most people have no idea about how bad this all can get. We'll find out soon enough though.
When ad revenue drops on platforms, the platforms simply reduce organic visibility which drives the need for regular ad spending for companies in order to remain visible on social platforms... AirBNB is riding a wave of prior popularity and name recognition, I guarantee they will go back to a certain point of obscurity at some point because they reduced their ad spend, and then be forced to promote heavily as they did once before.
It's all creates a new cycle of financial deception and manipulation on platforms. For very profitable companies, advertising is usually manageable, but for startups, for small business, and for independent creators, this practice is devastating financially, and fruitless on top of the financial loss of paying for promotion. These platforms also made promises to woo users based on free organic growth, which somehow conveniently disappeared due to covert and convenient EULA updates over time.
One could argue that no advertising is hard for the challengers, but in today's situation they are outspent anyway, so what do they have to lose?
The conclusion is that both incumbents and challengers would be better off in a world where no advertising exists.
Except they won't click those competitor links, because they are already specifically looking for Williams Sonoma
- review sites like Yelp (interestingly one of Google's loudest antitrust critics)
- domain names (arguably the entire ICANN generic top level domains sell-off)^1; US trademark grantees who fail to take action against confusingly similar domain names risk losing trademark protection
1. The gTLD application alone as $185,000 (https://en.wikipedia.org/wiki/Gtld)
In the case of Google and Yelp, the tech company can manipulate the alleged "algorithm" (read: no humans involve so you cant sue us, haha) behind the scenes and determine the "visibility" of the ad/review.
Ads are the perfect "business model" for tech companies because there is so little scrutiny of the ad services delivered. It is like philanthropy. When we make a donation, we generally do not track what happens to the money afterwards. We get a warm fuzzy feeling from making the donation as it is "doing the right thing". Then we leave the recipient to do as they please. How many companies buy ads on Google because they feel it is "the right thing to do". How many feel their donation was unwarranted after the ads fail to produce results.
Defensive ad purchases, defensive domain name registrations and even defensive gTLD purchases are one side of the coin. Another is the "winner take all" line-of-thinking (80/20, network effects, etc.) that tech companies worship that they in turn project onto customers. Competing for visibility on a fully searchable web of enormous capacity presupposes (artifically) that visible space is scarce. Ad auctions for the purpose of placing an ad on page one of hundreds of thousands of pages of results. The truth is that many people used to read newspapers from beginning to end. The whole paper, not just page one.
In a physical newspaper, there are ads on many pages. Often there are in fact no ads on page one. How many Google customers are encouraged to purchase ads that will appear on SERP #2 (do they even have ads on SERP #2). Imagine if newspapers tried to create a bidding war for page one of the physical newspaper. With the online versions, it seems that is exactly what happens. Do not blame the newspapers, do not blame Williams-Sonoma, blame the "tech" companies. This only reflects the pathetic "tech" company psychology, not the thinking of the businesses who give them money for online ad services.
Hard not to see this as a clumsy sales pitch for his company, especially when it starts with such a disingenuous example/quote (of course AirBnb didn't have to spend on performance marketing during a global pandemic where everybody was suddenly looking for a getaway...)
I mean I guess he would know from personal experience? He hadn't been the CEO for a year or two at the time, but he was still very much the public face of Moz when they "asked 28% of Mozzers to leave"[1], who worked on products complementing SEO. They used various euphemisms, but ultimately it's because they made no money[2].
[1] https://moz.com/blog/moz-is-doubling-down-on-search [2] https://twitter.com/randfish/status/765973082611781633
The most common source of inflation in Google/FB self reported performance numbers is multiple ad channels taking credit for the same order. If a customer clicks a Google ad then clicks a Facebook Ad then makes a purchase, each ad channel will claim credit for that purchase. In reality each ad platform only has claim to ~50% of the purchase (depending on what attribution algorithm you want to use).
In terms of knowing that real value is produced from these ad channels, I see it every day in many of our customers data. Clients will increase or decrease ad spend and there will be a correlated increase/decrease in sales.
tldr; Google/Facebook ads over report their numbers but do ultimately drive sales according to the data I have first hand access to.
(If you have a near-monopoly on planners, of course, such an ad would indeed have been a waste. The author would have come to you by default once he decided to get a planner. )
One additional thing people don't call out is that a lot of the budgets spent on these platforms are "learning" budgets. Agencies play this card really well. They'll tell you, "oh, you need to increase your budget, and test all these different combinations of ads/targeting/landing pages/etc so that you can learn what works (or the AI behind the platform can learn what works)". And obviously, in "learning" mode, you're ignoring the ROI.
I've seen people spend substantial amounts of money in "learning mode", and the platforms are kind of incentivized to make the learning less efficient so it takes longer and more spend for you to get to ROI positive (or to learn that you will never get there).
Any successful platform is not, because by definition they have enough ads, and so want to focus on driving conversions for advertisers so that demand, and hence revenue, increases.
Agencies are often super shady though, on that we agree.
For me it just made me think, well I'm doing something right because Facebook mostly offers me shitty white label stuff that's being sold using fraud (I've made complaints to Advertising Standards [UK] a few times about Facebook ads) and that I would never buy.
It sounds “deep state” but it’s actually plainly documented in government files and written about by reliable sources.
Remember we’re still emerging from an era of whispering the same old story of morality and obligation to each other.
I am not at all interested in helping someone build a fertilizer empire or pillow brand. Politically my hands are tied to doing so if I want a life.
Poor people effectively live a life of quota and state sanctioned limits on their access to material support by cutting social programs with public support.
Advertising America as anything but a sanctimonious police state is a scam.
> Reliably someone comes along every few months to question [performance marketing]. I always come back to analyses of incrementality as the real proof.
> Take an audience of X people. Divide them in two. Show ads to your test group, don't show to control. Watch your business grow and gauge the lift between the two audiences. The companies that know how to advertise at scale do this constantly and can gauge the real effect of their ad dollars. Facebook, Google and others make these tests possible in their platforms, while other software suites such as Impact Altitude and VisualIQ allow you to do this kind of analysis and testing as well.
> In the end, most of it proves out to be incremental. There are notable exceptions of course, but when are there not?
Any specific tips/links on how to do this with Google/Youtube retargeting?
The whole point of _performance_ advertising is that it's effect is _measurable_. If AirBnB spent $500m+ on performance advertising, they should be able to trace that back to an exact amount of revenue. If you are a brand in this scenario, you can conduct split tests by sampling the conversion rate of users from advertising vs non-advertising. Again, it should be simple to see if the conversion rate for users targeted via advertising has increased or is unchanged.
In the branded search examples, again, as an advertiser, you can see what searches are associated with your leads. While you do have to compete for attention on your own branded search (to compete against competitors taking the slot), you should also be able to recognize unbranded terms which drive conversions for you. Again, assuming this is actually _performance_ marketing, you would be able to look at the cost of these placements and the ROI, and the impact would be measurable.
The rest of the article is largely composed of straw-man arguments that imply the results are not measurable, when in fact they are (if done right).
disclaimer: I'm the CTO of a performance marketing platform. The vast majority of conversions on our platform happen same-session. There's a very easy way to measure this effect -> pause your campaigns and immediately see conversions fall.
As you mentioned: The most important test (that you should be doing every now and then) is to actually pause the ad and see if revenue falls. Anything else is just an approximation and should be treated as such.
"Rental stays in XYZ city" will bring up an airbnb result all day, probably in the top five results. Therefore, paying $3 a click to be placed above your own search result is probably silly.
I happen to have a small side project and advertise it with a very low budget on facebook, google, and bing. It works. I don't rank very high because my SEO skills are poor, but google ads absolutely drive real and interested people to my site.
So if you’re buying or even just evaluating performance ads without considering the bigger picture you might come to erroneous conclusions.
Take the Lego Movie example from the article. The $65 million movie is no doubt an awareness play. Could you make the case that you should also increase your performance budget to help capture more of the demand you just generated with the movie? Or should you just hope that people go from the movie theater to buy Lego unprompted? Is it worth it for Lego to advertise to people who walk out of the theater and search for “Lego Batman set” or whatever? I think so, even though evaluating such branded search campaigns individually might make them seem inefficient.
It seems very easy to dismiss the performance advertising as a scam when you evaluate it in a vacuum. As noted in the article it’s important (and very difficult) to understand the incremental outcome of any channel or campaign. That incrementality includes awareness campaigns.
After more than a decade in advertising and marketing I am now more than ever unwilling to accept simple or definitive answers to highly complicated questions. At best I hope that we can unwind some of the overall complexity so we can have a chance to trust some of those definitive answers.
But after reading this article, it finally dawned on me. He makes imperfect conclusions in everything he touches, it's just that in some fields those conclusions can be more easily proved to be wrong than in others. SEO is the perfect field where a polished presenter can get away with imperfect conclusions for years - trust me, I know, I made a living for years in this field, and I am very familiar with the nature of this work. Most of the time, you have no idea what the black box really does, and instead you're just trying to guess what might have happened. Most importantly, there are many ways to skin a cat in SEO, and just because your approach is net positive doesn't mean that you truly are delivering the global maximum (or that the net positive gain was ROI positive). In short, it's impossible to know who's right and who's wrong, and Rand's videos convinced me that he's right, but I am no longer sure. I just rewatched one of them, and can easily see how his conclusions are just... opinions.
While we may or may never find out if his SEO opinions were the global maximum, we can quantifiably demonstrate that his opinions on content marketing are not solid. This whole essay he wrote can be replaced with "hey performance marketers, don't trust the platform numbers and instead do your incrementality studies." Platforms like Facebook will give you those for free if you reach a certain spend level, and you can also get them from 3rd party providers like measured.com. In other words, if you're a performance marketer and you're not conducting incrementality studies, then you're very early in your career and are not following the best practices. Simple as that - no need to extrapolate from there and reach all sorts of additional conclusions (which is obviously a pattern in Rand's behavior) - calling into question a perfectly investable marketing channel, conflating the needs of a public company with everyone else's needs, using words like scam, etc.
I am really disappointed to have to write this, but you would have been better off not reading this article. If Rand is really advocating that the majority of entrepreneurs should follow his advice and focus on PR instead of performance marketing, then perhaps an honest thing to ask would be - how is that working out for his own company? AFAIK, SparkToro is nowhere close to replicating the growth of his previous company, which is honestly disappointing for someone with such a huge reach and name recognition.
Unsurprisingly, those people are wrong and marketing works.
The organization was too brittle to accept criticism from their eng. organization, as most are, until a crisis.
- Calling performance marketing platforms a scam (repeatedly, both in the title and in the narrative) doesn't explain how those same performance marketing platforms are carrying the majority of traffic acquisition in most of the B2C companies that went public this year (and practically all of the DTC ones). Calling into question the accuracy of measurement is one thing. Calling it a scam is wrong and designed to rank on HN rather than to be reflective of the true value of those platforms.
- As I pointed out in my original post, all you have to do is use incrementality studies and 98% of the criticism instantly goes away. Rand implies that you have to do your own studies (by eg, following Avinash Kaushik's methodology) which is 100% wrong - Facebook will do them for you if you reach a certain spend limit, or 3rd parties will as well with no spend limits. Also, from experience, this really becomes an issue once you spend meaningful amounts on two platforms at the same time. His rant on this subject has an iota of truth and a whole lot of sensationalism mixed together, and overall leads to wrong conslusions.
- He conflates "paid search" with "all performance marketing platforms", including "paid social." It would have been helpful to point out that the challenges with branded terms are entirely isolated to paid search and have nothing to do with paid social.
- My favorite sensationalist tactic: frame a strong accusation as a question. This way you get the clicks, but you can still cover your ass by linking to resources that with enough research would allow the reader to answer the question with a "No." But in lieu of that research, the implication is that the answer is a "Yes." You'll see this tactic used by less reputative media sources, and I was disappointed to see Rand do the same.
I could go on but hopefully this will suffice.
Your other points I think relate to scale. No advice can be universal, and if you read the article as absolutist, your take makes sense. If you read it as "hey, your mix is likely wrong", a lot of the criticism fades.
I think we've lost a bit of creativity in marketing. The Lego movie example is a really good one. I think it is probably good this happened, as a lot of creativity was performative (how do I win an ad award/impress my peers) and not about increasing sales, but we've perhaps shifted the balance too far, and there is likely some areas with good ROS that are now better bets.
Ecommerce example: every year the CPMs go up and your paid margin goes down. But every year you have a larger email list, so the balance of paid to unpaid shifts.
General example: every year you get more of the late-stage employees who care less and less about your company. But every year you can afford to pay for more layers of management, which will keep an eye on the underperformers.
This list goes on and on... The headwinds are driven by external forces, whereas the tailwinds end up working out based on your specific execution of the opportunities that present themselves to you. This is where an experienced operational team can make a huge difference.
> I think we've lost a bit of creativity in marketing.
You absolutely cannot rely on performance marketing forever. It's a shot in the arm until you have reached enough [fill in whatever you wnat] so that you can leverage that momentum to reach the escape velocity. So it's not good forever, but it's a great catalyst.
So FB, who earns on my spending, offers me to measure for me whether my spending makes sense for me. Why would FB ever tell me to spend less? Don't they like money?
Facebook does like money. Do you think lying to the biggest advertisers in the world is the best long-term strategy, or do you think instead it might be better to report out the most accurate results possible?
Long-term strategy? https://www.businessinsider.com/facebook-allowed-fraud-hacke...
There's literally a team at Facebook (Marketing Science), part of who's job is to tell advertisers what actually works and often, this can lead to them telling advertisers to spend less.
And this is an incredibly smart business strategy in that FB know (from experiments et al) that their platform works, and if they can show incrementality, then advertisers will invest more in the platform.
In game theory terms, it's a good strategy for a repeated game, which advertising definitely is.
My experience is exactly the opposite. The larger the budget, the less real hard analysis is done. This is especially true with the rise of attribution modeling which allows marketers to essentially motion blur the data.
That's not really an argument. You wish that a responsible/public company wouldn't do that but we have seen much dumber behavior.
https://www.forbes.com/sites/augustinefou/2021/01/02/when-bi...
However, if you are a new entrant/small business this is not the case, because nobody has ever heard of you.
As an example, look at TikTok. They spent insane amounts of money on FB ads to get as many installs as possible, but I'd suspect that they don't do this anymore because they've got enough brand equity that it doesn't make as much financial sense.
1. P&G didn't turn off that spend, it shifted it to other marketing platforms.
2. Chase didn't change its marketing spend, it just concentrated it from 400k sites to 5k sites.
3. Uber found out that their agency was committing fraud and AFAIK the case is still being litigated.
I hate being that guy, but check the links. Those reporters really must have an agenda or something (or are just struggling to get the clicks, so they need to make a story out of nothing).
https://en.m.wikipedia.org/wiki/Escalation_of_commitment
I do think things are better in some sense today. But it is a hungry beast, marketing. Oh, and lies and statistics.
Interesting article
1. "industry" is moving away from last touch attribution (pretty much what Rand complains about)
2. "Incrementality strives to identify the causal event of a conversion,"
The way to identify the causal action seems to be :
errr ... that's difficult
It's a surprise to see a vendor have a clear explanation on their site however - and it's worth quoting i think
>>> Incrementality testing creates an experiment that systematically withholds media channel exposure to a representative subset of users (the control group) while maintaining normal media channel exposure to the broader user set (the test group). If the control group is both sizable enough to be statistically significant and selected at random such that they are broadly representative of the user base, then the media channel’s incremental contribution can be determined by the difference in business outcome (conversion, revenue, profitability, etc.) between the test and control groups.
You would have these very complete theories of how search engine ranking worked - completely disconnected from reality and unburdened by any actual knowledge. If there was ever a cargo cult, SEO was it.
I still cringe when people say that they have "done SEO" on their site while I'm in the room. It is kind of like telling their mechanic they've sacrificed a goat so their car will get better mileage.
Content strategy, keyword research, Search Console integration, site speed improvements, local search business profiles -- all of these contribute to SEO. It's unwieldy to talk about them all, so sometimes people can use a shorthand.
I'd always give the same advice. Don't try to outsmart the search engine, focus on quality and consistency over time. That will always reward you in the long run. Try to game the search engine and you will find yourself in a weapons race you can't sustain over time. If you find exploitable tactics to get better ranking than you deserve, sooner or later you'll probably be pushed down the ranking. And this is really important if you're successful enough to have something to lose.
Some of what people label SEO is just "make good web sites", and that indeed works. But some of it is just pure rubbish. And making a good website isn't just "doing SEO" - it is making a good website. And a lot of people don't.
Which is why I can't take people seriously when they present it as some singular task they have completed.
[1] or black hat SEO for that matter.
Instead of product and customer, think value added. You buy a server for $0.065 an hour. You buy traffic at $0.01 a click. That's your raw commodities. How do you make a profit off of that? Supposedly if you're good enough you can get elected president. I'm sure there's room in there for you to pay your bills.
I think this is overall a waste of resources and I'd like to see a more virtuous system, but I fail to imagine one.
#1 - Eliminated all paid search other than some limited branded search terms and shifted all the money to affiliates who were way better at making profits on the keywords we were competing on
#2 - Eliminated all display advertising after running numerous experiments showing it provided almost not incremental conversions, even though the platforms happily took credit for them.
Those two things drove our blended CAC down substantially and by building better affiliate relationships, sales actually increased.
The lesson here is that you need to try a lot of things out and you should be continuously questioning what you're doing and running specific experiments to gut check effectiveness of any ad platform that is slapping cookies on wide groups of users and claiming conversions.
My suspicion is that this is near impossible at any large organization, even one as new as Airbnb. I can just imagine someone walking into a team of 20+ performance marketers and suggesting they need to experiment to determine if any of it is remotely effective. COVID forced them into this but it's something that they should have already been doing.
I also suspect that the top line focus/obsession of most VC-backed companies make this type of exercise seem almost counterintuitive. Don't mess with or question the momentum.
So, it's great for your business that this worked, but I personally don't see any strategy that leans on affiliates to be worth celebrating.
That said, I’ve worked at brands where paid search and paid social are incredibly valuable channels.
This is all just my opinion and I'm sure there are better marketers than me that make this work, but this is what I've seen happen in too many VC-funded/backed companies. And it gets even harder when we're at the super levels of funding that we're now seeing. So if I'm a relatively small company (funded to $50m) that is spending $5m a year on marketing, and I'm competing against a company with $500m in funding and, say $100m on marketing, then it's even worse.
But, as you said, the incentives point towards terrible outcomes, so this keeps happening.
On the other hand, while I do indeed believe that the "ROI" from Performance Advertising is something between just false and deliberately misleading, the bigger picture that I'm interested in is marketshare. Because when looking at market share, it's not a question of incrementality anymore, but whether you're growing slower/faster than your competitors, and your cost of doing that, and at what point you're OK to 'buy' marketshare, in the sense of losing money in the pursuit of growth, and how much. And then, OK, let's talk about ROI on that basis - most of the time, achieving this will indeed require tools from the Performance Adversiting toolbox, which allow you to conveniently track the amount of marketshare (i.e. sales) you bought.