198 comments

[ 4.4 ms ] story [ 272 ms ] thread
To be fair you should factor whitespace out when you're calculating the area.

In the recent screen capture all the whitespace is added to the non-search results.

In the not-so-recent results there is a fair amount of whitespace added to the search results (as well as non-search).

I honestly thought about cutting all the white space completely, but then after having a chat with our designer... He said that every page is visually divided into blocks (even if there are no explicit visual borders of these blocks), subjectively, of course. This, the "F-pattern" and "left-to-right" orientation for the most of us brought us to an agreement that the last screen was ok... The subjective "blocks" are somewhat like this.

But I agree, I shouldve been more precise

That makes some sense; sort of defining what the whitespace is "devoted" to.
I noticed this as well. The older search results have the same width as the new results.

It's also interesting to note that the old results page is displaying multiple results for single domains using result nesting, while the new results page displays only one listing for each domain. The end result is that the same number of 'effective search results' are displayed on the page for both versions.

The gradual transformation of white space into content is a natural evolution in the life of any web product. While the feature set is simple, the design can be sparse. Mature products benefit from rich feature sets, with the drawback of having to sacrifice the sparse visual design that may have initially distinguished them from the competition.

(comment deleted)
This seems to me to be a similar kind of evaluation that Google would do on its own work. Nice angle.
(comment deleted)
I've been using Adblock+ for the last 6+ years. Granted the 13.8m users [1] are a drop in the bucket compared to googles non Adblock+ viewers. I was a little shocked to see this post as it is much different than what I see.

For the search "saas help desk" (https://www.google.ca/#hl=en&q=saas+help+desk)

I see: http://imgur.com/xMB6t

They see: http://1.bp.blogspot.com/-5Vq-qWuUPZE/UER9ChRkKQI/AAAAAAAAAd...

[1] https://addons.mozilla.org/en-US/firefox/addon/adblock-plus/

wow -- just installed Adblock+. the experience was a lot like finally progressing far enough in my career to not have to wear a tie; i feel freed from bondage...
They are being good guys about it; they could easily block all "Ad-Blockers" and ask you to disable the adblocker to continue (as some sites already do).
You underestimate the ingenuity of geeks. If Google decided to prevent people from using Ad-blockers, 24 hours later a new Ad-blocker would be released to get around it, or the existing ones would be updated.
Except such an arms race is highly asymmetric in Google's favour.

Google can have a team of 10 working full time racing Ad-Blocker's volunteers.

Google can ask for ever more aggressive permissions from web masters, Ad-Blockers must play nice with browsers' plugin api.

Google can hire away the most productive Ad-Blocker developers.

Google can ban ad-blockers from chrome's plugin repository and halve the ad-blocker userbase overnight.

Google has the warchest for a multi year engagement. Most plugin authors cannot last a single year fighting full time.

If Google went total war then ad-blockers would disappear within a year. It would be a massive PR hit, maybe even a legal issue. The risks make it all a numbers game, but there does exist a point where Google can no longer ignore ad-blockers. Thus ad-block users must be careful and not become too numerous.

All those arguments sounds similar to arguments about why the RIAA would have easily stamped out file sharing.

I would say that the arms race is highly asymmetrical against Google. Building ad-blocking technology isn't that hard, so the labor force force is significantly larger than Google could ever afford to combat.

Google is smart to not take on this battle. Only a small percentage of user's install ad-blocking software, and those users would not likely click on ads anyway. Now, if only Google could teach the RIAA of its ways.

If listening to music necessarily required that you download a new 'app' for every play, then the RIAA could shut down piracy pretty easily.
Bluray security is pretty close to that. Somehow I still keep seeing bluray rips on the internet.
That's because the content doesn't necessarily require you to download a new app every play - the content is the same every play, and any software/license download is just a nuisance.

Google search results are different every time, and the freshness of a result is essential to its utility. If you always needed "today's version of The Movie" then piracy would be much easier to thwart.

I don't buy that for a second. Where is the asymmetry on Google's end? Google has to stop adblockers everywhere, meanwhile, adblockers only need to have a lucky strategy once, and social distribution spreads it everywhere. It's exactly the opposite of what you assert.

And at the end of the day, the web page is rendered on hardware (a display) owned by the user. In the very worst case, the final bitmap being displayed could be modified to remove ads; this crude approach could not be stopped short of a totalitarian state (Google isn't quite that strong) or mangling web pages into unreadability (in which case they wouldn't get traffic so ads would be pointless).

But it would never get that far, for the simple reason of accessibility (i.e. for people with disabilities). Accessible websites are machine comprehensible, to a greater or lesser degree. What the machine can comprehend (in form, if not meaning), it can edit.

Imagine the arms race with a super fast 30 minute release cycle. Every 30 minutes a new updates comes, 30 minutes later a new counter measure.

During those 30 minutes of vulnerability Google is losing only the marginal cost of serving an ad-free search, a rounding error.

During the other 30 minutes ad-blockers will lose users. Users of ad-blockers are not fanatics, they have no moral issues with ads. They are normal people who are using ad-blockers because it improves their browsing experience. So what happens when these ad-blockers temporarily ruin the browsing experience? They'll temporarily turn off said ad-blocker. By the 87th time what percentage will have given up on ad-blockers for good? If that percentage is anything greater than 0% then Google is winning.

The race is not like the DRM arms race, it is more akin to the virus arms race. Users want to run untrusted code and said untrusted code wants to do something the users do not want it to do.

I doubt Google would be willing to do 30 minute release cycles. Like any large piece of code search needs to be tested before it is released, the tests alone probably take 30 minutes to run. Google probably could not update their code more frequently than once a day. Risking breaking the search webpage is too high of a cost. On the other end ad blockers don't have to worry about lossing millions of dollars a second if they introduce a bug so they can respond much quicker.
People wouldn't use a product that changed that quickly. People complain enough as it is.
(comment deleted)
(comment deleted)
Actually, there's no way to really block ad-block.

It wouldn't be too hard to make an ad-blocker that used CSS and Javascript to do all of the blocking during page render. It'd be an inefficient use of bandwidth and rendering time, but it'd be able to block 100% of ads and undetectable from the web server.

In fact, before Opera had a built-in ad-blocker, I think there was an ad-block extension built on its "User Javascript" and custom CSS abilities.

Seriously?

  function verifyAds() {
    if (!adContentsVisible()) {
      loadAdContents();
    }
    setTimeout(verifyAds, 1000);    
  }
  verifyAds();
Yes, but it's easy for a browser to just override your verifyAds function.
Huh? Extension can override JavaScript on the page, too, you know:

window.adContentsVisible = function () { return true; }

  (function() {  
    // local scope, can't override
  })();
My point is that I agree with grandparent, that they just allow ads to be shown, and it's not technically impossible to reload ads, or prevent page contents view, if no ads visible.
The plugin can absolutely mock the ad element(s) and make them seem to be visible.
only because sites generally don't do so; if necessary, extensions could get much more aggressive with website scripts, current user script limitations aside.
If it's client-side, it can be defeated. Let's not enter into an arms race of larger and more bandwidth-intensive checkers/blockers to try to get the upper hand. If someone ad blocks, they're not likely in your target demographic anyway.
(comment deleted)
greasemonkey:

    function adContentsVisible() { return true }
not if it's defined in local context.
I can modify my browser's javascript interpreter if I need to.
Also, it is possible to block 3rd party ads without touching the browser by going straight into the underlying IP routing tables.
"During page render".

Render the page, identify the regions of the resulting bitmap which are ads, and blank them out. The browser can always just lie at the end, even after running all of the JavaScript and marking all of the DOM elements to visible.

If they find a way to force the visibility of ads, we still could move them to the bottom of the page.
Mobile Safari successfully refrains from adblocking.
There is another factor that people are ignoring for all the talk of technical possibilities...

Those who are minded to go out of their way to install something that blocks ads probably aren't very likely to click ads anyway.

They are also probably technology 'leaders' who help sites to bring in all the regular folk who do click the ads.

Google is probably the least likely company to lose users this way, but they have another factor that they want a good reputation with that community because they want to hire lots of them.

If google becomes too annoying people will leave. There are alternatives.
I think it's more a matter of simple customer segmentation (same principle behind outlet malls: charge price-insensitive customers a premium, without losing the price-sensitive but hassle-insensitive customers who are willing to schlep to the outlet mall for a discount.)

AdBlock allows Google customers to self-segment the same way. Win-win.

Google makes money from people clicking on ads. I'd wager that people who install adblock, don't tend to click on ads, especially if you force them to disable it.
Also about 14m users [1] of Ad Blockers for Chrome. Those numbers are nothing to sneeze at, I wonder how it will play out in a few years if more and more people default to no ads.

[1] https://chrome.google.com/webstore/category/popular

I was wondering what the author was going on about until I remembered Ad-block. It's still a very clean interface and the distinction between ads and search results is pretty clear. Doing a quick informal search, the terms that the author chose seems to be a large outlier in terms of numbers of ads.
bing seems to be doing better in this regard.
ok granted it is doing better by allocating more space to the search results. But how about the search result quality? How's it doing there?
(comment deleted)
> Is Google becoming a Yahoo?

I suspect that all ad-revenue sites are subject to the same economic pressures. Almost all such companies eventually succumb to these pressures.

All such companies will tend to cater to the lowest common denominator. They will tend to put their own interests (that of showing ads and increasing ad effectiveness) over the interests of users.

Perhaps there is an opportunity for a paid integrated search/email/calendar/reminders/maps/events application? You pay them $5 or $10 per month, and they do their best to create the best possible experience for you. Apple and Microsoft are well positioned to become this company as well, as is Google.

Wouldn't it be ironic if Yahoo! became this company?

> Perhaps there is an opportunity for a paid integrated search/email/calendar/reminders/maps/events application? You pay them $5 or $10 per month, and they do their best to create the best possible experience for you.

You've just described Google Apps for Business (well, except for search).

The failure of A/B testing. More ads => more revenue. People won't abandon the site immediately just because a new ad or two showed up, like a slow boiling pot, so short term effects are always positive.
Just like the sesame-seed hamburger bun: http://www.joelonsoftware.com/items/2007/09/11.html

(Though, personally, in this case, I don't think Google's getting enough credit. The added space for search tools is designed to make search better for most users, and probably do, even if most of us don't need them.)

I've had trouble staying focused when using Google a while back and decided to write a Chrome extension for improving the Google Search experience. Here's the URL for those interested: http://restfulpanda.com/post/25658805059/cleaning-google

EDIT:

Turns this: http://f.cl.ly/items/2y1F0d2d0a2K3L1n0Q2m/1.png

Into this: http://f.cl.ly/items/2P3B2c0X0V0Q3U0h3e3A/2.png

A big improvement, but as with ad blockers, you still have search results crammed into a fairly narrow box. Obviously, you won't be able to get Google to show longer titles and descriptions, but you should be able to widen the box so that the two-line descriptions end up on one line, which would fit more results on each page.
I just wish you could get it to expand rightwards and get the longer descriptions back, the amount of space they've wasted to 'integrate' G+ really annoys me.
This observation made the rounds at Google a few months ago. Note also it depends on the topic of your search. 3 ads in the yellow box are reserved for highly monetized queries - "flowers" or "car insurance" for example.

Queries about programming topics - "sinatra post parameters" is one I tried a few minutes ago - usually don't have ads.

are the "highly monetized queries" queries also the most common queries?
They can be -- but not necessarily. It has a lot to do with conversions. Someone searching "car insurance" or "cheap flights" is most likely on the hunt for something they are willing to pay for; while someone searching "football scores" is most likely just looking for free information.

Of course this doesn't even scratch the surface of running a site for ad revenue that provides free information to common search questions, e.g., "how do I change the oil in my car"

Wow I just searched NCAA rankings and saw no ads.
Ads are actually more likely to provide the answer to your question, or the solution to your task, when you're searching for something to buy. Which may weigh into why Google puts so many of them on the page.
It's one of the things that prevents me from ad-blocking sites like Google. Some of the time I will actively be searching for the ads, and avoid the natural results, because the natural results for some product searches can be extremely spammy with "review" sites and comparison shopping sites with horrible user experiences dominating the results, while the paid results are usually decent retailers or distributors.
Not to be a shill for Duck Duck Go, but the amount of search information displayed is refreshingly dense:

https://duckduckgo.com/?q=saas+help+desk

https://duckduckgo.com/?q=travel

First time I have seen mapquest used anywhere other than mapquest.com.
I made duckduckgo.com my home page several months ago after I saw http://donttrack.us/, and it has been working fine. The presentation is nicer than google. I have AdBlock enabled, so I don't see the ads on either site.

I like the infinite scroll on duckduckgo; I have AutoPager plugin on chrome to get something similar with google.

I'm not bombarded by fifteen different ads, who are all trying to pretend like search results?

The results are hit or miss...I probably !g about half my queries. But the other half are entirely pleasant experiences and I'm hopeful DDG continues to get better.

The ads on the Google result page are probably better links than most of the organic results. People who pay for traffic usually make the payoff worthwhile.
Nonsense. All three ads on the Google page are to thin domains with little more than a contact form to harvest your info and spam you to death.
At first I thought "well, if you're using an ad blocker, it's different", but in Google's case, you just end up with white space instead of ads because the search results are still crammed into a 42em wide box. I wonder if there's a user script or chrome plugin to fix that.
Some results are even worse. There were keywords where you could see only 7 organic results.

The rest was AdWords.

To make things worse, Google adds more and more "product search" results (paid placement).

With each day, there's less and less traffic for those people who don't pay Google. I suspect that the paid placement trend will increase until it is possible.

To make things worse, Google not only decreases the SERP "real estate". They also introduce frequent algorithm updates which tend to drop down websites which earn on search-engine traffic. Leaving those which may have lesser quality but don't earn on their traffic ( no monetization ). And if you do it to a million websites, you will gain quite a lot of new AdWords buyers. Some of them have to try your PPC ads. Let's say that 10% of them will try to get traffic out of PPC - now you've got yourself 100.000 paying customers.

It's obvious they want to force people both to buy and to click ads. They don't gain anything by sending free traffic to other affiliate websites ( which will then convert to sales/leads ).

The question is: Where do you draw a line between a search engine and an "ad engine"?

To many people, especially those less tech-savy, Google = Internet.

I personally think, it's pretty bad for the overall state of e-economy to have this one big mogul who administers the majority of targetted traffic. He wants his share of the pie from everyone. If you are too small to pay, fuck you. They don't need you. The consumer won't notice you are not in the search results, there are 50 other websites happy to take your place. Either you are at the top or you don't exist.

It'd be great if there would be something like Google but with a more "socialistic" approach. Instead of having a few websites occupying the top place, provide the traffic to other, smaller players. So the e-economy can expand more.

> Instead of having a few websites occupying the top place, provide the traffic to other, smaller players. So the e-economy can expand more.

How would you do that? Throw in some random numbers to shuffle the search results?

In the image where you show the 18.5% organic footprint vs 81.5% ad footprint, you are lumping in a lot of page real estate used for navigation and search tools. The entire left hand side is not add related.

I'd be a more accurate/fairer way to represent the info by providing 3 categories: organic, ads, navigation/misc. If you did this, you'd provably see a closer to 40/60 split, still in favor of ads. Similar to what you showed in your screenshot from the past.

I looked for "81" in the page and found nothing. Apparently, he never said that 81.5% of the area is made up of ads. He may have edited it but I think his criticism was towards the unused space for results, not especially towards the ads.

And if that is the case, I must concur. I fail to understand why Google added a large side bar instead of adding more top bars in the unused space next to the logo (perhaps right under the search text box). The designers at Google may disagree but the Jitbit guy sure has a point.

There is a graphic that shows 18.5% for the organic and 81.5% for the ads/rest of the area.

The main reason I disagree with the conclusion is that the area used by Google, or any other company for that matter, for navigation/other tools is hardly equivalent to space used for ads.

That in no way is intended to say that the space is used "efficiently".

He didn't go to Google looking for navigation, he went looking for the results of his search.

So the way he discusses the real estate devoted to fulfilling his purpose is quite reasonable.

Except you use the navigation to improve your search. No one wants a page completely full of organic results.
I do.

Well, except perhaps my query at the top so I can refine it if necessary.

So turn on adblock or get one of the scripts that strips down the UI. Mountains of testing have shown that you're in the severe minority here.
I do use scripts for that. I was just providing a counter point to the parent's "No one wants". "Most people don't want" would be more accurate.
So you would prefer something like the this:

http://i.imgur.com/ryiPq.png

It seems like poor UX to me. It's a little ugly how line length isn't constrained and you're unable to do anything but type a query into the box. Special queries would require memorizing commands or just doing without.

I'd prefer a multi-column layout, 'cause you're right that long lines are hard to read.

I don't use that many special queries, except sometimes "site:", and frankly if I have to type (or paste) the entire domain anyway, "site:" is way less effort than switching to a different text field, even if it were one of the defaults.

I'd be OK having a "More Options" link or something for the rare case when I want to search for images by color or whatever.

Having said that, I'm aware this is very much a personal opinion. Google can design their site for the majority. I don't mind.

I kinda like the collapsable sidebar present in Google Maps. However, I rarely miss the extra space taken up by the current sidebar on the main search results page.
Exactly - I use the "search within date range" functionality a lot. No, I don't want to type out that in the search box the entire time - esp. when the default facets (1 day, 1 month, 1 year back) are so convenient.
What I want, ideally, is exactly a page of organic results that fulfills my search query with perfect relevance.

Ideally, in the perfect world, I don't need the navigation, page 2, or even a place to refine my search. Ideally, I don't need any of that.

If we're talking about what I want, that's it.

Surely the ideal would be a single result with exactly the information you need. But because the search query is seldom perfect, it's good to have several results - and by extension, some paid results too.
In a perfect world, you don't even want page 1 -- why show ten results when you really only want one?

And behold, that perfect world exists now: "I'm Feeling Lucky".

You joke, but yes, that's exactly what I want. Why would I not want exactly what I'm looking for delivered directly to me?

Also, "I'm Feeling Lucky" isn't an option for me, not in the version of Chrome I have anyway. Even if I explicitly load the www.google.com which shows the feeling lucky button, as soon as I type 1 letter in the box the screen changes to an interactive search, which no longer has the lucky button.

Until people show me that paid results aren't relevant (hint: they usually are), then most of the real estate is fulfilling his purpose.

Also, if Google decided to some how change to using 2 columns of organic search so they could squeeze 16 results into a single page, you'd hear just as much, if not more, complaining.

Arguably, from a design perspective, a wider column of text is harder to read than a narrower column of text. I personally find Google to be more legible than it was before at 53% result space. That's just me, though.
I'm not sure it is just you. Shorter columns are easier to read, which means it serves users even better; our eyes don't have to track a massive row of characters to see what the search result is.
Question for people more familiar with google ads that me:

I quite often type in a search term for something that I imagine must be a very common search term and I don't get any adverts whatsoever.

Does this mean that nobody has bid on this keyword (because it's too expensive?) or are some search terms simply not allowed to be advertised on?

Some terms don't display advertisements - if the combination of relevance and google's profit potential on a specific query is not above a certain level (which is always changing I'm sure based on yield management and user data tests), then no ads will display for that query.

Most navigational queries fall into this class.

By definition if no one has bid on the keyword, you can easily win the auction by bidding the minimum amount. It is likely that the terms you were searching for are not ideal targets for effective ads. Can you give an example of such query?
I can't remember which terms I searched when I discovered this, but I tried it just now.

"computer networking" yields 1 ad. "networking" yields no ads.

The search in our isn't considered 'search'?!
(comment deleted)
For a query like "credit card", those non-organic results seem much more, if not just as, relevant than the organic results. Why do people assume that if it's an ad that it must be non-relevant? Advertisers wouldn't be advertising on these keywords unless they were converting a good percentage of users ... that means a good percentage of users are finding those links relevant.

I don't use AdBlock, don't really see the reason to as I find most of the search ads to be relevant whenever I'm searching for a term that actually shows search ads (which isn't really that often). Hate for search ads just because they are ads is kind of an odd imo ...

"For a query like "credit card", those non-organic results seem much more, if not just as, relevant than the organic results"

That's almost certainly on purpose to get better ad click ratio. Google manually inspects all major keywords, you can bet "credit cards," "mortgage" and "insurance" are tweaked and tweaked to Google's specification. Showing "bofa.com" when searching for mortgage isn't that useful to the average user, so people click on ads.

Can we please change the headline to something a bit less invalid and full of sensationalism? This is a perfect example of distorting statistics to serve ones purpose.

* Author has a small screen.

* Author compares results space vs everything else, and then calls it "ads"

On my 1920x1200 screen, overall ad space and results space actually take up the same amount of space. There's a ton of UI and white-space though. While Google could arguably make better use of their empty space, suggesting the remainder is entirely ads is absurd.

Screenshot with results in red and ads in blue: http://i.imgur.com/YTspx.png

> * Author has a small screen.

Author has a big screen. His screenshot shows five results above the fold. I can barely see two on my 1280x800.

The argument works both ways.

1280x960 is not a "big screen" by any modern definition.
(comment deleted)
The CRT I threw out years ago was 1280x1024. It wasn't big then, and a smaller screen isn't big now.

Even if you're on something like an iPad you still have 1024 vertical and end up with plenty of content. Only netbooks and/or low-quality screens have this issue.

Is that actually so different from the way things used to be? If you run the example 2008 query, "open source social networking," the number of ads hasn't changed.

It seems like most of this complaint should be directed at the navigation sidebar to the left. I'd imagine Google tested the sidebar ad nauseam, though, and is leaving it there for a reason (probably increased engagement in other search products and more readily exposes advanced refinements).

Ironically (or not) jitbit.com is a similar percentage of content above the fold, with almost identical content width.

I say 'or not', because there are well established readability reasons for having a narrow content width. Would many people genuinely prefer the google of the 2008 screenshot? The days of sites with full page width text are largely gone, and for good reason.

I'm sure google would also suggest that the ads are content.

The author is only analyzing content above the fold. There's no conclusion to draw from this, other than that Google used to cram more content above the fold, like all other sites.