I know for announcements, HN prefers the coverage of others, rather than breathless press releases. For this topic though, maybe the original source is better?
From the article -- “We’ve discovered that it is one thing to build a search engine, and an entirely different thing to convince regular users of the need to switch to a better choice.”
I give these guys and gals full marks, and massive respect, for getting the full stack of a search engine integrated (front end, ranking, indexing, crawling).
They regularly dealt with the #1 impediment in monopoly search, Google buying traffic. It is harder to isolate just the money they pay to third parties to redirect traffic but based on their Q1 results for 2023 it is at least $20 BILLION dollars a year these days.
Google can't keep that up though, and given the increasing ways they have been cooking the books to make their numbers, I feel that within 5 or 6 years they will be dead or irrelevant. They certainly won't be able to pay Mozilla and Apple to send them all their traffic and they likely won't be able to force Android users to do the same. When a VC recently asked me what was going to happen to Google I suggested the likely outcome was that they sell themselves to Apple leaving behind just Apple and Microsoft, each with a platform, a hardware business, a search engine, and an OS.
Of course I figured this would have transpired by now :-) so clearly I'm pretty lame as a prognosticator.
That said, once the dust settles I think I know how a company could be formed around the technology known as search which would in fact be viable, but it would not have the margins that Google now enjoys.
For context, I'm not at all a fan of Google and try to avoid all of their products as best I can. However your prediction here seems very disconnected from reality.
Even if antitrust required Google not pay for search engine placement, they control the single largest browser (by far). And even when presented with a free choice on a browser they don't control, many users will continue using Google just out of habit. Most people use Google as a verb to mean "search the web" and these habits are firmly entrenched.
The most likely occurrence to disrupt their search market share has got to be AI, and they have a meaningful competitor in the LLM chat bot space along with significant resources to keep up with the others. So it seems like a stretch to even think that a new mode of search will completely obliterate them in a short time frame.
Even besides search, they have robust businesses with advertising, media, hardware, software services, and more. Admittedly these are smaller revenue-wise than search ads, but would still be a meaningful business if search somehow was snapped out of existence suddenly.
> Most people use Google as a verb to mean "search the web" and these habits are firmly entrenched.
within a year or two, which is a long time on the internet, bing could trench deep enough. i think bing chat outperforms google bard at the moment as well.. but who knows what could happen.
We just see different realities. My tenure at Google was 2006-2010, and they had just decided maybe the mini-kitchens were a bit much.
The challenge at Google is that the only business they have that makes money is search advertising. What is worse, the margins on search advertising have been eroding steadily since about 2008[1]. Google kept that from showing up in their results by first cutting down on the amount of money they sent to AdSense, then by increasing the number of ads on "Google properties", and then by constructing "revenue zones" around search that is considered "commercial" ("Where to buy <x>" is a commercial search, "What is the speed of light" is not). The result has been the "hollowing out" of their product.
When I joined Blekko in 2010 (acquired by IBM in 2015) the concept of a "spam free" search engine was new. It was a hit, especially with people who used search for their jobs, but there is a minimum level of traffic you need if you are going to support it with ads. We ran two versions of the service, the "labeled" version which had your standard web API and the "white label" version for folks who wanted search results but they wanted to monetize them differently. There are lots of things you can do there. For example you can put an Amazon affiliate ID for every result that meets the query but points to Amazon, you can boost a page in the results that is a link to another page in your inventory that has more profitable ads on it, you can absorb all of Wikipedia and show its pages with your own ads on them. We had people try all of these schemes and more. A number of things we thought of as "sketchy" like returning a first page of links that was nothing but ads "above the fold"[2] is something Google is now regularly doing.
The Google "product" that makes money has turned to crap. That makes them ripe for disruption but they do, as you note, have a ton of money on hand. They have tried unsuccessfully for the nearly 15 years since I left to come up with another product that makes money. Youtube is the closest bet, and they have done all sorts of crap to make it profitable (like literally putting ads on people's videos and taking the money for themselves) and yet it struggles to make a profit (I'm not sure if it has ever had a full year where it made more than it cost to run it).
So I see them as dying slowly have eroding margins while they fight against their own cultural biases in defining what is a "hit" and what should get canned. As a former insider (and knowing people who still work there) and looking at the problems that got them here that are not addressed, I am not confident they will be able to pull out of the dive. Its been over a decade and their descent rate hasn't changed AFAICT.
Yes, it is a different reality than the one you see. And I'm okay with it if it turns out my reasoning was flawed in someway and they pull out of the slump and go on to regain some former greatness. That will be a win because I will learning something from it. But as a guy who has spent a lot time in the operations role, of the future probabilities going forward for them, I think the most probable is an acquisition by Apple.
[1] 2008 was the year the Mortgage crisis hit but it was also the last point at which a step change in search quality was achievable due to algorithm and not just putting more machines on the problem. So from that point on, competing with Google "Search Quality" was a money problem not a technical problem.
[2] Web parlance that the "fold" is the second screen, so anything you can see in the browser window with scrolling down is above the "fold" (old newspaper term).
“Alphabet cash on hand for the quarter ending March 31, 2023 was $115.102B”
Bard also seems like a reasonable competitor to chat gpt.
As someone with kids YouTube is king.
Google search and ad network could be declining, Facebook did certainly enter their market and take some share. But the revenue per employee is high , and unlike say ibm of the past I don’t think google yet is in a decline like them - and ibms decline started in the 90s and they are still around.
I’m not the biggest google fan but even if they shrunk by half with their cash pile it will only take buying a growing company, or one more good investment to keep going.
Anti-monopoly authorities in the US — and in the EU! — would never allow Apple to buy the only other company with a mobile OS, not to mention the most popular browser and 90% of the search market.
And a company as large as Google is way too large to swallow even for Apple, or for Microsoft.
Google is much larger than Nokia ever was, and Android has a market share Nokia was not even able to dream of.
Also: I may be proved wrong, but I don’t think ChatGPT will have the effect on search that the iPhone had on mobile phones; and, in any case, Google is getting ready with Bard. My bet is that Google will be one of the world’s 5 largest companies even 5 years from now.
Man, I don't want to crap on someone, but Kagi hasn't impressed me at all. The need to pay for it eventually didn't help much either
It was a little better than Google for sure, and a lot better than DuckDuckGo, but I really really don't like having to log in. Every time I wasted one of my free trial searches by getting bad results felt really bad too
i tried it, and when i searched for how to build a website in 2023, it gave me a link to reddit on how to get a girlfriend... The top non discussion links were all the same as Google.
To take down an entrenched competitor, aiming at the center of their core competency doesn't work. You have to offer something better, something they can't duplicate because it doesn't fit well with their core competency.
> We’ve discovered that it is one thing to build a search engine, and an entirely different thing to convince regular users of the need to switch to a better choice.
Bold to take the stance of "woe to the users of an inferior product that just don't know they could be using us". If your search engine is that much better, you have to tell people about it, explain why it's better, and if you succeed users will indeed take the time to change their browser's search engine setting.
I imagine it would've been nice to get a bunch of users who go "hmm I guess i'll use this random search engine called Neeva" if all OSes/browsers had a list of search engines that the user had to select from upon setup, but the product failed more because you failed to either make a better product, or inform to the users of the benefits of switching. Lowering the barrier to switching makes it easier for people to switch to another engine just for the sake of it instead of switching to improve their own UX.
And the barrier of having to pay for a search engine is infinitely larger than the friction of going into your browser settings to change the default engine.
I think past hackernews comments and posts have proven that this is at least partially wrong. The drag is real. How many posts are there about duckduckgo -- in its golden days; that it was better and should be the default.
How much people preaching time and marketing has been spend on that? And yet the average user, even technical users does not care.
It has to be a vastly different experience, e.g. ChatGPT enabled search -- like Bing now say -- and even then, the monopoly moats placed by Google and the decades of inaction by DoJ have created what we have now.
Chrome should have never been allowed to be under Google or Alphabet. Paying to set a search engine to this scale should not have been allowed as a practice.
[Edit:] P.S. And to answer all the thousands of comments by users here of why Google has gotten worse: because it can; it can get worse while maximizing its ad profit; DoJ allowed it.
> And yet the average user, even technical users does not care.
If users do not care, they do not think the UX improvement is worth switching. So what is the goal here? Antitrust isn't happening here because the harm is only to competitors, with the UX being harmed if Safari, Chrome et al had to implement a popup like "duckduckgo.com would like to become your default search engine. switch? y/n".
At the time of my writing this comment (3 hrs after post; 14 comments), no recognition that the founding team is now doing crypto. Specifically, https://n.xyz for which they raised $40m in a Series A financing.
I’ve discovered saving by using the annual plan was a mistake. I’ll avoid that mistake if I try Kagi or something else.
But really miss Neeva’s integrations that were cancelled recently. Having the ability to search my GitHub repos & the internet simultaneously was pretty cool. Add to that Notion, and other services. It was great for the the 4 months it was around but with 8 months left, I feel ripped off but it doesn’t seem worth getting too worked up over.
That’s awesome though I haven’t received confirmation.
To me, it still seems less advantageous to try something that seemed as stable as Neeva with a 1 year subscription, month to month is okay. That’s what I’ll likely do for a Kagi subscription…
34 comments
[ 3.1 ms ] story [ 83.3 ms ] threadI know for announcements, HN prefers the coverage of others, rather than breathless press releases. For this topic though, maybe the original source is better?
I give these guys and gals full marks, and massive respect, for getting the full stack of a search engine integrated (front end, ranking, indexing, crawling).
They regularly dealt with the #1 impediment in monopoly search, Google buying traffic. It is harder to isolate just the money they pay to third parties to redirect traffic but based on their Q1 results for 2023 it is at least $20 BILLION dollars a year these days.
Google can't keep that up though, and given the increasing ways they have been cooking the books to make their numbers, I feel that within 5 or 6 years they will be dead or irrelevant. They certainly won't be able to pay Mozilla and Apple to send them all their traffic and they likely won't be able to force Android users to do the same. When a VC recently asked me what was going to happen to Google I suggested the likely outcome was that they sell themselves to Apple leaving behind just Apple and Microsoft, each with a platform, a hardware business, a search engine, and an OS.
Of course I figured this would have transpired by now :-) so clearly I'm pretty lame as a prognosticator.
That said, once the dust settles I think I know how a company could be formed around the technology known as search which would in fact be viable, but it would not have the margins that Google now enjoys.
But $50B/year profit doesn't look like a dying company auctioning off its parts.
Even if antitrust required Google not pay for search engine placement, they control the single largest browser (by far). And even when presented with a free choice on a browser they don't control, many users will continue using Google just out of habit. Most people use Google as a verb to mean "search the web" and these habits are firmly entrenched.
The most likely occurrence to disrupt their search market share has got to be AI, and they have a meaningful competitor in the LLM chat bot space along with significant resources to keep up with the others. So it seems like a stretch to even think that a new mode of search will completely obliterate them in a short time frame.
Even besides search, they have robust businesses with advertising, media, hardware, software services, and more. Admittedly these are smaller revenue-wise than search ads, but would still be a meaningful business if search somehow was snapped out of existence suddenly.
within a year or two, which is a long time on the internet, bing could trench deep enough. i think bing chat outperforms google bard at the moment as well.. but who knows what could happen.
The challenge at Google is that the only business they have that makes money is search advertising. What is worse, the margins on search advertising have been eroding steadily since about 2008[1]. Google kept that from showing up in their results by first cutting down on the amount of money they sent to AdSense, then by increasing the number of ads on "Google properties", and then by constructing "revenue zones" around search that is considered "commercial" ("Where to buy <x>" is a commercial search, "What is the speed of light" is not). The result has been the "hollowing out" of their product.
When I joined Blekko in 2010 (acquired by IBM in 2015) the concept of a "spam free" search engine was new. It was a hit, especially with people who used search for their jobs, but there is a minimum level of traffic you need if you are going to support it with ads. We ran two versions of the service, the "labeled" version which had your standard web API and the "white label" version for folks who wanted search results but they wanted to monetize them differently. There are lots of things you can do there. For example you can put an Amazon affiliate ID for every result that meets the query but points to Amazon, you can boost a page in the results that is a link to another page in your inventory that has more profitable ads on it, you can absorb all of Wikipedia and show its pages with your own ads on them. We had people try all of these schemes and more. A number of things we thought of as "sketchy" like returning a first page of links that was nothing but ads "above the fold"[2] is something Google is now regularly doing.
The Google "product" that makes money has turned to crap. That makes them ripe for disruption but they do, as you note, have a ton of money on hand. They have tried unsuccessfully for the nearly 15 years since I left to come up with another product that makes money. Youtube is the closest bet, and they have done all sorts of crap to make it profitable (like literally putting ads on people's videos and taking the money for themselves) and yet it struggles to make a profit (I'm not sure if it has ever had a full year where it made more than it cost to run it).
So I see them as dying slowly have eroding margins while they fight against their own cultural biases in defining what is a "hit" and what should get canned. As a former insider (and knowing people who still work there) and looking at the problems that got them here that are not addressed, I am not confident they will be able to pull out of the dive. Its been over a decade and their descent rate hasn't changed AFAICT.
Yes, it is a different reality than the one you see. And I'm okay with it if it turns out my reasoning was flawed in someway and they pull out of the slump and go on to regain some former greatness. That will be a win because I will learning something from it. But as a guy who has spent a lot time in the operations role, of the future probabilities going forward for them, I think the most probable is an acquisition by Apple.
[1] 2008 was the year the Mortgage crisis hit but it was also the last point at which a step change in search quality was achievable due to algorithm and not just putting more machines on the problem. So from that point on, competing with Google "Search Quality" was a money problem not a technical problem.
[2] Web parlance that the "fold" is the second screen, so anything you can see in the browser window with scrolling down is above the "fold" (old newspaper term).
Bard also seems like a reasonable competitor to chat gpt.
As someone with kids YouTube is king.
Google search and ad network could be declining, Facebook did certainly enter their market and take some share. But the revenue per employee is high , and unlike say ibm of the past I don’t think google yet is in a decline like them - and ibms decline started in the 90s and they are still around.
I’m not the biggest google fan but even if they shrunk by half with their cash pile it will only take buying a growing company, or one more good investment to keep going.
I think concerning part is that profit margin was falling.
And a company as large as Google is way too large to swallow even for Apple, or for Microsoft.
Also: I may be proved wrong, but I don’t think ChatGPT will have the effect on search that the iPhone had on mobile phones; and, in any case, Google is getting ready with Bard. My bet is that Google will be one of the world’s 5 largest companies even 5 years from now.
Google's current market cap: $1.56 trillion
It was a little better than Google for sure, and a lot better than DuckDuckGo, but I really really don't like having to log in. Every time I wasted one of my free trial searches by getting bad results felt really bad too
https://wikipedia.org/wiki/Searx
Blocks tracking + search aggregation + self hostable
Bold to take the stance of "woe to the users of an inferior product that just don't know they could be using us". If your search engine is that much better, you have to tell people about it, explain why it's better, and if you succeed users will indeed take the time to change their browser's search engine setting.
I imagine it would've been nice to get a bunch of users who go "hmm I guess i'll use this random search engine called Neeva" if all OSes/browsers had a list of search engines that the user had to select from upon setup, but the product failed more because you failed to either make a better product, or inform to the users of the benefits of switching. Lowering the barrier to switching makes it easier for people to switch to another engine just for the sake of it instead of switching to improve their own UX.
And the barrier of having to pay for a search engine is infinitely larger than the friction of going into your browser settings to change the default engine.
How much people preaching time and marketing has been spend on that? And yet the average user, even technical users does not care.
It has to be a vastly different experience, e.g. ChatGPT enabled search -- like Bing now say -- and even then, the monopoly moats placed by Google and the decades of inaction by DoJ have created what we have now.
Chrome should have never been allowed to be under Google or Alphabet. Paying to set a search engine to this scale should not have been allowed as a practice.
[Edit:] P.S. And to answer all the thousands of comments by users here of why Google has gotten worse: because it can; it can get worse while maximizing its ad profit; DoJ allowed it.
If users do not care, they do not think the UX improvement is worth switching. So what is the goal here? Antitrust isn't happening here because the harm is only to competitors, with the UX being harmed if Safari, Chrome et al had to implement a popup like "duckduckgo.com would like to become your default search engine. switch? y/n".
https://www.linkedin.com/posts/sridhar-ramaswamy_web3-crypto...
I’ve discovered saving by using the annual plan was a mistake. I’ll avoid that mistake if I try Kagi or something else.
But really miss Neeva’s integrations that were cancelled recently. Having the ability to search my GitHub repos & the internet simultaneously was pretty cool. Add to that Notion, and other services. It was great for the the 4 months it was around but with 8 months left, I feel ripped off but it doesn’t seem worth getting too worked up over.
Just can’t let that happen again.
> If you have a paid subscription to Neeva Premium, you will receive a refund for the unused portion of your subscription.
To me, it still seems less advantageous to try something that seemed as stable as Neeva with a 1 year subscription, month to month is okay. That’s what I’ll likely do for a Kagi subscription…