5 ms·

High-Profile Study Turns Up the Antitrust Heat on Google

78 points by IBM 11y ago

AVTizzle 11y ago
>> "Wu co-authored the study with Michael Luca, an assistant professor at Harvard Business School, and data scientists at the local reviews site Yelp, which has been one of Google's primary opponents in the global antitrust fight. Wu was paid by Yelp for his work on the paper." That's a fun fact.
- mda 11y ago
  But it is a high profile study :)
- logn 11y ago
  "A 2011 Harvard study by Michael Luca found that there was no significant statistical correlation between being a Yelp advertiser and having more favorable reviews." - https://en.wikipedia.org/wiki/Yelp#Alleged_manipulation_by_Yelp https://en.wikipedia.org/wiki/Yelp#Alleged_manipulation_by_Y...
  - snsr 11y ago
    Skewed reviews or not, I tend to avoid Yelp search results. They intentionally degrade the web experience in favor of repeatedly promoting their native application; the opposite of useful content when searching for a restaurant etc.
- littletimmy 11y ago
  It is an unfortunate fact of academic life that you need a grant to complete your study. Take a look at studies that come out of HBS or Stanford GSB; more often than not they are sponsored by big banks and investment firms. I do feel as uncomfortable about this as you do. This sort of sponsoring ruins academic objectivity.
  - CrazyCatDog 11y ago
    There is no need for grants in top b-schools. Whereas other academic departments pay faculty 9-months out of the year, even third-tier b-schools will pay for 2/9 or 3/9 of your salary in the summer if you publish just about anything. Larger studies? No problem, the dean has you covered. The brilliance of b-schools is that they can mint $ with certificates and exec Ed, all of which is available to them at low overhead (namely the university collects a lesser cut). It's preposterous to believe that faculty at the very top need to whore themselves out to industry for funding. What they will do to get their hands on data, however, is unspeakable. Yelp, Dropbox, eBay, etc. are merely providing data to the researchers, and more often than not, they have a say on what gets published. The truth always does eventually come out (e.g. negative company information) but it does so behind conference/seminar doors, not in print.
    - meepmorp 11y ago
      I'm just guessing here, but you don't seem like you have much familiarity with academia and how funding works.
  - j_baker 11y ago
    How is this any different from when tobacco companies fund studies that say that smoking isn't bad for you?
dylanjermiah 11y ago
"The new study, which was presented at the Antitrust Enforcement Symposium in Oxford, U.K., over the weekend, says the content Google displays at the top of many search results pages is inferior to material on competing websites. For this reason, the paper asserts, the practice has the effect of harming consumers. "
hahainternet 11y ago
I'm just beginning my reading of this 'paper' but so far I'm not convinced. From what I can tell, they compared Google's pages with and without Yelp reviews, but Yelp won't give this data to Google. Therefore it's complete bullshit, they're comparing a hypothetical product that cannot exist and arguing that Google is being anticompetitive because users prefer this nonexistant product.
- threeseed 11y ago
  Whether Yelp supplies data to Google is irrelevant. The point of the study was to show that users don't like ads masquerading as organic content and neither do competitors. Therefore if Google is the only winner then their motives are anticompetitive. The logic is sound. Yelp doesn't need to be complicit in it to disagree with it. Nor frankly should they. It's a dumb feature that actually makes Google less functional.
  - zaphar 11y ago
    No that wasn't the point of the study. They didn't remove ads. They removed knowledge graph data. Knowledge graph results aren't paid placement they are a collection of facts that Google has discovered through various sources, much like organic search results. In this case the facts come from Google+ reviews. Google is just as likely to display Yelp reviews if they were able to collect that data. Yelp however doesn't supply it.
    - jdmichal 11y ago
      > Google is just as likely to display Yelp reviews if they were able to collect that data. Yelp however doesn't supply it. I'm pretty sure arguing for this kind of solution would be silly if Google were to, say, also require payment. I wonder, is this really so much different without money changing hands? Why is it Yelp's responsibility to provide Google access to their data in order to get a fair playground in "Search"? The same Google, by the way, which offers a directly competing product "Locale"? It's a pay-to-play model, just without any money in order to seem innocent. Except everyone knows that data is Google's real currency, so that whole innocence goes out the window and the true cost to pay-to-play is revealed. I think Yelp is exactly right to want to protect their data from Google, and I think they're right in fighting this.
jacquesm 11y ago
Any google code that explicitly mentions another google product in their search engine rankings algorithm would be evidence of anti-competitive behaviour. Absent that I don't see how you could make an outside observation stating that google engages in anti-competitive practices. There may be a lot of smoke but that's the only thing that I would call a fire. So if they want to prove this google will have to open up their search engine code base, which I really don't see happening.
- threeseed 11y ago
  This happens today though. Google search for "bars" shows a custom component at the top which links through to what is a mini Yelp site complete with reviews. It isn't aggregating reviews from the web they are specifically "Google Reviews" from Google+ users. This sort of behaviour is very similar to what got Microsoft in so much trouble. Going from one market "search engine" to another "reviews". Normally adding features like this would be fine but because Google is so powerful as to make/break web businesses it can easily be seen as anticompetitive.
- jfuhrman 11y ago
  From http://www.wsj.com/articles/how-google-skewed-search-results-1426793553 http://www.wsj.com/articles/how-google-skewed-search-results... >A previously undisclosed report by staffers at the Federal Trade Commission reveals new details about how Google Inc. manipulated search results to favor its own services over rivals’, even when they weren’t most relevant for users. >In a lengthy investigation, staffers in the FTC’s bureau of competition found evidence that Google boosted its own services for shopping, travel and local businesses by altering its ranking criteria and “scraping” content from other sites. It also deliberately demoted rivals. >For example, the FTC staff noted that Google presented results from its flight-search tool ahead of other travel sites, even though Google offered fewer flight options. Google’s shopping results were ranked above rival comparison-shopping engines, even though users didn’t click on them at the same rate, the staff found. Many of the ways Google boosted its own results have not been previously disclosed. >One way Google favored its own results was to change its ranking criteria. Google typically ranks sites based on measures like the number of links that point to a site, or how often users click on the site in search results. >But Marissa Mayer, who was then a Google vice president, said Google didn’t use click-through rates to determine the ranking for its own specialized-search sites, because they would rank too low, according to the staff report >Instead, Google would “automatically boost” its own sites for certain specialized searches that otherwise would favor rivals, the FTC found. If a comparison-shopping site was supposed to rank highly, Google Product Search was placed above it. When Yelp was deemed relevant to a user’s search query, Google Local would pop up on top of the results page, the staff wrote. >Other regulators have found similar practices. European antitrust authorities in 2013 said Google had a different, “specialized” search algorithm for ranking its own content. >To bolster its own listings, Google sometimes copied, or “scraped,” information from rival sites. According to the FTC report, Google copied Amazon’s rankings of how well products were selling, then used that information to rank its results for product searches. Amazon declined to comment. >While Google promoted its own results, it sometimes demoted rivals, the FTC staff found. For example, Google compiled a list of comparison-shopping sites and “demoted them from the top 10 web results,” staff wrote. According to the report, Google users in tests didn’t like the changes; only after Google tweaked its search algorithm at least four times, and changed the ranking criteria, did the new results get “slightly positive” feedback, the staff said. >Google’s efforts paid off, the FTC found. It said Google’s maneuvers reduced Web traffic to rivals, and increased traffic to Google sites.
hussong 11y ago
There are some areas (e.g. weather), where Google's own content is a lot more useful (because of its simplicity) to me than third-party sites. On others (e.g. shopping in DE) not so much.
- greggman 11y ago
  Except at least in my experience their results are Awful. I'd be really curious to know how often their predictions actually matched reality. Not to mention flat out bugs http://techsuxor.blogspot.jp/2011/05/google-weather-on-mobile-is-missing.html?m=1 http://techsuxor.blogspot.jp/2011/05/google-weather-on-mobil...
fauigerzigerk 11y ago
>“Google appears to be strategically deploying universal search in a way that degrades the product so as to slow and exclude challengers to its dominant search paradigm, ” the paper concludes So Google is hurting competitors by degrading its own product (if you believe this study commissioned by a competitor). This sounds to me like a situation that is perfectly self limiting and self correcting.
norea-armozel 11y ago
I'm split on this issue. To me, it should be Google's job to both provide a list of viable candidates from its advertising segment and that which is found on the wider web. I just wish that Google would return to the way it use to mark the difference between the two. Anymore, it seems they want to confuse users between their advert search results and the actual web results. But it's clear that this is an inherent conflict of interest that can't be solved, at least in my opinion, by using a sledgehammer approach of antitrust laws.
at-fates-hands 11y ago
It's interesting that for years people have been hammering them about being a monopoly and looking for angles to try and get the DOJ to go after them. While it's clear they're giving preference to their own content, it's a shaky argument since I know I can go to multiple other search engines and compare results. You have a harder time convincing me as well since this is not like the Walmart in BFE North Dakota where people have to drive an hour before they can find another competitor - it's the internet. A few clicks here, a few clicks there, and you have plenty of other options. Sorry, but you can't call it a monopoly if people are just too lazy to find another search engine to compare results with.
- jdmichal 11y ago
  People really need to understand that antitrust and monopoly do not mean the same thing. The existence of a competitor does not automatically make antitrust behavior OK. The main focus of antitrust law is how an action effects consumers; effects on competitors are only interesting in the sense of effecting consumers. (With the understanding that more competition generally means more choices for consumers, which is held as a Good Thing™.) I would suggest giving the Wikipedia antitrust article a good read, as a start: https://en.wikipedia.org/wiki/United_States_antitrust_law https://en.wikipedia.org/wiki/United_States_antitrust_law
Grue3 11y ago
Great, now there's scientific basis behind what I've been observing for the last 5 years or so. Google search results are actually really, really bad. There are ads everywhere and everything is designed to make you click them. I'm not at all surprised by the findings. And it doesn't matter that Yelp paid them, after all it's Google who is unfairly competing against Yelp here and hurting their business.

78 points by IBM 11y ago

>> "Wu co-authored the study with Michael Luca, an assistant professor at Harvard Business School, and data scientists at the local reviews site Yelp, which has been one of Google's primary opponents in the global antitrust fight. Wu was paid by Yelp for his work on the paper." That's a fun fact.

mda 11y ago
But it is a high profile study :)
logn 11y ago
"A 2011 Harvard study by Michael Luca found that there was no significant statistical correlation between being a Yelp advertiser and having more favorable reviews." - https://en.wikipedia.org/wiki/Yelp#Alleged_manipulation_by_Yelp https://en.wikipedia.org/wiki/Yelp#Alleged_manipulation_by_Y...
- snsr 11y ago
  Skewed reviews or not, I tend to avoid Yelp search results. They intentionally degrade the web experience in favor of repeatedly promoting their native application; the opposite of useful content when searching for a restaurant etc.
littletimmy 11y ago
It is an unfortunate fact of academic life that you need a grant to complete your study. Take a look at studies that come out of HBS or Stanford GSB; more often than not they are sponsored by big banks and investment firms. I do feel as uncomfortable about this as you do. This sort of sponsoring ruins academic objectivity.
- CrazyCatDog 11y ago
  There is no need for grants in top b-schools. Whereas other academic departments pay faculty 9-months out of the year, even third-tier b-schools will pay for 2/9 or 3/9 of your salary in the summer if you publish just about anything. Larger studies? No problem, the dean has you covered. The brilliance of b-schools is that they can mint $ with certificates and exec Ed, all of which is available to them at low overhead (namely the university collects a lesser cut). It's preposterous to believe that faculty at the very top need to whore themselves out to industry for funding. What they will do to get their hands on data, however, is unspeakable. Yelp, Dropbox, eBay, etc. are merely providing data to the researchers, and more often than not, they have a say on what gets published. The truth always does eventually come out (e.g. negative company information) but it does so behind conference/seminar doors, not in print.
  - meepmorp 11y ago
    I'm just guessing here, but you don't seem like you have much familiarity with academia and how funding works.
- j_baker 11y ago
  How is this any different from when tobacco companies fund studies that say that smoking isn't bad for you?

dylanjermiah 11y ago

"The new study, which was presented at the Antitrust Enforcement Symposium in Oxford, U.K., over the weekend, says the content Google displays at the top of many search results pages is inferior to material on competing websites. For this reason, the paper asserts, the practice has the effect of harming consumers. "

hahainternet 11y ago

I'm just beginning my reading of this 'paper' but so far I'm not convinced. From what I can tell, they compared Google's pages with and without Yelp reviews, but Yelp won't give this data to Google. Therefore it's complete bullshit, they're comparing a hypothetical product that cannot exist and arguing that Google is being anticompetitive because users prefer this nonexistant product.

threeseed 11y ago
Whether Yelp supplies data to Google is irrelevant. The point of the study was to show that users don't like ads masquerading as organic content and neither do competitors. Therefore if Google is the only winner then their motives are anticompetitive. The logic is sound. Yelp doesn't need to be complicit in it to disagree with it. Nor frankly should they. It's a dumb feature that actually makes Google less functional.
- zaphar 11y ago
  No that wasn't the point of the study. They didn't remove ads. They removed knowledge graph data. Knowledge graph results aren't paid placement they are a collection of facts that Google has discovered through various sources, much like organic search results. In this case the facts come from Google+ reviews. Google is just as likely to display Yelp reviews if they were able to collect that data. Yelp however doesn't supply it.
  - jdmichal 11y ago
    > Google is just as likely to display Yelp reviews if they were able to collect that data. Yelp however doesn't supply it. I'm pretty sure arguing for this kind of solution would be silly if Google were to, say, also require payment. I wonder, is this really so much different without money changing hands? Why is it Yelp's responsibility to provide Google access to their data in order to get a fair playground in "Search"? The same Google, by the way, which offers a directly competing product "Locale"? It's a pay-to-play model, just without any money in order to seem innocent. Except everyone knows that data is Google's real currency, so that whole innocence goes out the window and the true cost to pay-to-play is revealed. I think Yelp is exactly right to want to protect their data from Google, and I think they're right in fighting this.

jacquesm 11y ago

Any google code that explicitly mentions another google product in their search engine rankings algorithm would be evidence of anti-competitive behaviour. Absent that I don't see how you could make an outside observation stating that google engages in anti-competitive practices. There may be a lot of smoke but that's the only thing that I would call a fire. So if they want to prove this google will have to open up their search engine code base, which I really don't see happening.

threeseed 11y ago
This happens today though. Google search for "bars" shows a custom component at the top which links through to what is a mini Yelp site complete with reviews. It isn't aggregating reviews from the web they are specifically "Google Reviews" from Google+ users. This sort of behaviour is very similar to what got Microsoft in so much trouble. Going from one market "search engine" to another "reviews". Normally adding features like this would be fine but because Google is so powerful as to make/break web businesses it can easily be seen as anticompetitive.
jfuhrman 11y ago
From http://www.wsj.com/articles/how-google-skewed-search-results-1426793553 http://www.wsj.com/articles/how-google-skewed-search-results... >A previously undisclosed report by staffers at the Federal Trade Commission reveals new details about how Google Inc. manipulated search results to favor its own services over rivals’, even when they weren’t most relevant for users. >In a lengthy investigation, staffers in the FTC’s bureau of competition found evidence that Google boosted its own services for shopping, travel and local businesses by altering its ranking criteria and “scraping” content from other sites. It also deliberately demoted rivals. >For example, the FTC staff noted that Google presented results from its flight-search tool ahead of other travel sites, even though Google offered fewer flight options. Google’s shopping results were ranked above rival comparison-shopping engines, even though users didn’t click on them at the same rate, the staff found. Many of the ways Google boosted its own results have not been previously disclosed. >One way Google favored its own results was to change its ranking criteria. Google typically ranks sites based on measures like the number of links that point to a site, or how often users click on the site in search results. >But Marissa Mayer, who was then a Google vice president, said Google didn’t use click-through rates to determine the ranking for its own specialized-search sites, because they would rank too low, according to the staff report >Instead, Google would “automatically boost” its own sites for certain specialized searches that otherwise would favor rivals, the FTC found. If a comparison-shopping site was supposed to rank highly, Google Product Search was placed above it. When Yelp was deemed relevant to a user’s search query, Google Local would pop up on top of the results page, the staff wrote. >Other regulators have found similar practices. European antitrust authorities in 2013 said Google had a different, “specialized” search algorithm for ranking its own content. >To bolster its own listings, Google sometimes copied, or “scraped,” information from rival sites. According to the FTC report, Google copied Amazon’s rankings of how well products were selling, then used that information to rank its results for product searches. Amazon declined to comment. >While Google promoted its own results, it sometimes demoted rivals, the FTC staff found. For example, Google compiled a list of comparison-shopping sites and “demoted them from the top 10 web results,” staff wrote. According to the report, Google users in tests didn’t like the changes; only after Google tweaked its search algorithm at least four times, and changed the ranking criteria, did the new results get “slightly positive” feedback, the staff said. >Google’s efforts paid off, the FTC found. It said Google’s maneuvers reduced Web traffic to rivals, and increased traffic to Google sites.

hussong 11y ago

There are some areas (e.g. weather), where Google's own content is a lot more useful (because of its simplicity) to me than third-party sites. On others (e.g. shopping in DE) not so much.

greggman 11y ago
Except at least in my experience their results are Awful. I'd be really curious to know how often their predictions actually matched reality. Not to mention flat out bugs http://techsuxor.blogspot.jp/2011/05/google-weather-on-mobile-is-missing.html?m=1 http://techsuxor.blogspot.jp/2011/05/google-weather-on-mobil...

fauigerzigerk 11y ago

>“Google appears to be strategically deploying universal search in a way that degrades the product so as to slow and exclude challengers to its dominant search paradigm, ” the paper concludes So Google is hurting competitors by degrading its own product (if you believe this study commissioned by a competitor). This sounds to me like a situation that is perfectly self limiting and self correcting.

norea-armozel 11y ago

I'm split on this issue. To me, it should be Google's job to both provide a list of viable candidates from its advertising segment and that which is found on the wider web. I just wish that Google would return to the way it use to mark the difference between the two. Anymore, it seems they want to confuse users between their advert search results and the actual web results. But it's clear that this is an inherent conflict of interest that can't be solved, at least in my opinion, by using a sledgehammer approach of antitrust laws.

at-fates-hands 11y ago

It's interesting that for years people have been hammering them about being a monopoly and looking for angles to try and get the DOJ to go after them. While it's clear they're giving preference to their own content, it's a shaky argument since I know I can go to multiple other search engines and compare results. You have a harder time convincing me as well since this is not like the Walmart in BFE North Dakota where people have to drive an hour before they can find another competitor - it's the internet. A few clicks here, a few clicks there, and you have plenty of other options. Sorry, but you can't call it a monopoly if people are just too lazy to find another search engine to compare results with.

jdmichal 11y ago
People really need to understand that antitrust and monopoly do not mean the same thing. The existence of a competitor does not automatically make antitrust behavior OK. The main focus of antitrust law is how an action effects consumers; effects on competitors are only interesting in the sense of effecting consumers. (With the understanding that more competition generally means more choices for consumers, which is held as a Good Thing™.) I would suggest giving the Wikipedia antitrust article a good read, as a start: https://en.wikipedia.org/wiki/United_States_antitrust_law https://en.wikipedia.org/wiki/United_States_antitrust_law

Grue3 11y ago

Great, now there's scientific basis behind what I've been observing for the last 5 years or so. Google search results are actually really, really bad. There are ads everywhere and everything is designed to make you click them. I'm not at all surprised by the findings. And it doesn't matter that Yelp paid them, after all it's Google who is unfairly competing against Yelp here and hurting their business.