Dweebs, horndogs and geezers
December 19, 2006
Now this is mind-boggling. Check out how radically different the top ten search terms of 2006 were for Google, Yahoo and AOL.
3. World Cup
1. Britney Spears
4. Jessica Simpson
5. Paris Hilton
6. American Idol
7. Beyonce Knowles
8. Chris Brown
9. Pamela Anderson
10. Lindsay Lohan
4. American Idol
There's only a single overlap: "American Idol" appears on both Yahoo and AOL. That's it. I would have thought that, given the sheer number of searches done through each engine, there'd be a lot more similarity in the results. I guess it means that very different types of people use each of the three engines.
Looking back over the results, I think I can suggest the following market segmentation: Google users are dweebs. Yahoo users are horndogs. And AOL users are geezers.
At first, the appearance of "Tattoo" on the AOL list threw me. It didn't seem to fit with the Geezer demographic. But then I realized that the old folks are just trying to figure out what the hell's going on with their grandkids these days.
UPDATE: Allen Stern provides tables showing the top ten searches on Google and Yahoo for every year from 2003 through 2006. The Yahoo searches don't change much from year to year. But the Google searches go through a radical transformation. In 2004, Google's top searches looked a lot like Yahoo's top searches look this year. The four most popular Google searches in 2004 were Britney Spears, Paris Hilton, Christina Aguilera, and Pamela Anderson. At some point over the last couple of years, in other words, Google users went from being horndogs to being dweebs. This seems to confirm what I've long suspected: Googling is bad for the libido. Consider yourself warned.
The lack of overlap might be odd, but it probably has more to do with the way each engine is filtering it's list than the raw results. Odds are "the", "a", "sex" and "mp3" are in there as the true leaders, but that's not exactly interesting and obviously each list has been filtered somehow. Google's is odd, try comparing Bebo to YouTube on Google Trends and you'll see some other side of Google happens to rank the later far higher than the supposed #1 term...
Clearly the algorithm/humans making those lists are run slightly differently than both each other and other doorways to the same data...
so perhaps the categories have a bit more to do with the characters of the organizations' staff than their users?
Posted by: Abe at December 19, 2006 02:13 PM
Abe's right, there is some serious tinkering here. This is more like what I would expect if the search giants got together and subtracted what they had in common.
You mean there might be some fiddling going on here? I'm shocked! Shocked! I thought search results were, above all other earthly things, sacrosanct.
In any case, I'm going to stick with my own explanation. It's funnier.
Posted by: Nick Carr at December 19, 2006 03:57 PM
They say everything we buy tells a little bit about ourselves. Marketers are starting to use technology to predict what we like and think based on these little tidbits of information. A recent article in Fortune magazine describes this nicely http://money.cnn.com/magazines/fortune/fortune_archive/2006/11/27/8394347/index.htm
Maybe you've stumbled on to another predictor - which search engine you use.
Posted by: Michael Schaffner at December 19, 2006 05:06 PM
I've never seen such blatant nonsense released by companies as fact. Ditto other posters on the more salacious aspects of web search as being obviously filtered out.
Posted by: John Koetsier at December 19, 2006 05:29 PM
Are you a Gmail user? A Google rep at SES Chicago suggested that Google is starting to auto-target search results for people logged into their Gmail account. Yahoo! could be doing the same for their profiled users (MyYahoo, Yahoo Mail, etc.). I've had instances recently where my clients see different search results for the same terms than I do.
In other words, maybe Google things you're a dweeb and Yahoo! thinks you're a perv :)
Posted by: Dr. Pete at December 19, 2006 06:08 PM
Further evidence that the ranking algorithms are crazy: this article is #1 on Techmeme. Of all the interesting articles that Nick writes, this fiddly thing bubbles up to the top like "Britney Spears."
Nick - looks like we were on the same page, I posted a similar report (though mine is in a chart) to show the differences. One of my readers believe Google's users are more sophisticated and that's the reason behind the lack of entertainment searches.
I also created charts for 2003-2006 on Google Yahoo and AOL.
Posted by: Allen Stern at December 19, 2006 10:17 PM
but wait--i'm a geezer and i always google. i'd never use either of those others. so now what does that do to your demographic?!
but seriously, that really is interesting. do you know the raw numbers, like what size population makes up each list?
Posted by: lotusgreen at December 19, 2006 11:22 PM
Google sucks for doing research. It's great for buying things. Yahoo, only used it by accident (I like Yahoo's new home page). AOL, really never used it.
I prefer using Ask for research. Google still gives too many garbage results (including blog postings, no offense!) :)
It will be interesting to compare and contrast the top 25 or top 50 entries for each search engine. Is any data available on that?
Posted by: sam mishra at December 20, 2006 02:23 AM
Another possible explanation for the difference between google and yahoo is that people use those sites for different purposes. To me google is an institution; if you need to search you go there, no doubt. Yahoo has a great music site and this is probably increasing its accountability on music. Therefore people go there for music and search for music stars. I think that AOL is really a specific case. It is like an archeological piece that goes back to the first internet era. AOL grew within its niche, people who started to use internet in the first place (several years ago). I guess people who are using it are probably older than the average Internet user.
Posted by: Marco Bettiol at December 20, 2006 05:01 AM
That's very good peiece of Information
Posted by: angelamwilson at December 20, 2006 05:38 AM
Is this a symptom of the toolbar? Many of the searches on Google seem to be websites which may be a result of people just banging in the website name and letting Google go find it. I wonder if MSN shows a similar effect.
Posted by: Chris Edwards at December 20, 2006 07:07 AM
That's very good peiece of Information
Posted by: angelamwilson at December 20, 2006 08:45 AM
I'd argue that Google is used as a quick launch for other sites because it's quicker than writing it to many.
Over half of Google's top 10 are brands – people aren't really searching for information about Wikipedia or MySpace, they want to get to that page.
Now that Google is the homepage of nearly everybody I know, and most browsers now have a search in them too, it's quicker just to type in what you want, press enter and hit the first result then to worry about whether it's .org or .com.
I'm one of those responsible for the "video" search, I used to use that to get to YouTube before I realised that I used it everyday I put it on my bookmarks bar. However, I am a geek and know how to use my computer. Most of my friends ignore all that "stuff" at the top of the screen, Google is all they need. It's like ODS all over again.
Posted by: Thomas Edwards at December 20, 2006 10:09 AM
Yahoo's landing page lists top 8 search terms on yahoo. Assuming that yahoo has found success with this valuable space on its landing page i.e. users click on these links and stay on yahoo site longer, yahoo's top search terms are more likely to remain the same over multiple years.
What is the big deal? Interesting yes! But every one monetizes traffic by directing clicks and optimizing the referrals. This is the nature of economic markets. You want neutral? Go to the Post Office. The Search Market won't be mature until Google is twice the size of Yahoo and three times Aol and four times MSN. The famous Zipf Curve.
Hi. Interesting. I think it could be due to the top results all being very close. Or perhaps each engine has manipulated the figures (chosen how they are measured) and wants to give out a certain image.
I have to disagree with Marco about AOL. AOL certainly originally was marketed at novice users. It had a reputation of being restrictive for the experienced user in attempting to simplify. To be fair, I don't know how much that has changed now, I have just always steered clear of them.
These lists are complete baloney and I think the PR firms were consulted on them. Google is trying to appear wise and geeky, Yahoo is trying to appear cool and trendy. AOL... might actually be somewhat honest.
I mean, seriously. "Weather" didn't make Google's top 10 but "wiki" and "metacafe" did? And I'd bet money that one of Yahoo's top 10 search terms is "Google" but that one didn't make the cut...
Posted by: Michael Moncur at December 21, 2006 10:06 AM
I agree with the posts about people using the each of the tools differently. Yahoo and AOL are destination sites people use as portals. AOL does tend to skew toward the newer users, younger users or people who don't visit the site just for search purposes. Their search needs are probably more lifestyle based. I would also categorize Yahoo as a lifestyle and reference destination site so the queries from Yahoo sites will also skew the way AOL's do.
Google is not a destination site (yet) but more of a launch pad or springboard to other sites: People don't go to Google's home page for editorial or promotional content, they go there to get somewhere else. As a consequence the most common search terms will be more varied and rooted less in "pop culture" or lifestyle.
Look at Google's statement : "we compared frequent queries this year against 2005". These are not the top ten queries, but the top ten *growth* (whether it's relative or absolute, I can't tell). Of course, "Sex", "Weather", "MP3" and "Video" are in the top ten. As suggested, a simple look at Google Trends would show you that.
This functionality is actually associated with any Googla account, and can be fine-tuned in the Google Toolbar: it's called Google Suggest.
I'm not sure to follow your interpretation of the observation of the Zipf's law. I would strongly challenge the idea that a mature market is characterized by a Zipfian distribution of its shares.
A real milestone would be to have in these results something not in latin alphabet.
I spent some time in Google Trends: my guess would be something like:
video, yahoo, sex, google, games, what, myspace, youtube, porn, mp3, weather, dictionary, phone, maps, blog, cars, wow
As you have to type in sugesstion and compare them, this list certaily misses some --- but two things strike me:
- "video" has replaced "MP3";
- The presence of "yahoo" argues in favor of having many people actually use Google as a launch-start, instead of bookmarks.
In my opnion, "google" must be here because of queries associated with product names: I can't beleive people look for the service they are using --- but once again, you wouldn't beleive how many people ask me for directions to the Louvres Museum when standing in front of the Pyramid.
Just weighing in late, however, my company released a study on the differences between google and Yahoo using emperical conversion data:
Might help to shed some light on the differences.
Posted by: jeffh at December 30, 2006 09:24 PM
The reason why the Google list is so different every year is that it is not actually a list of the Top 10 searches, but rather it is a list of the Top 10 growing searches compared with the previous year. As a result, one would expect the lists to change every year, as well as to be very different from a list of Top 10 searches at other sites.
Nonetheless, it's pretty interesting to see how quickly people want to blame differences in results on evil PR schemes, rather than trying to understand what was actually done.
Happy New Year, Joe
Posted by: Joseph Cerro at December 31, 2006 03:03 PM
Post a comment
Thanks for signing in, . Now you can comment. (sign out)(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)
"Riveting" -San Francisco Chronicle
"Rewarding" -Financial Times
"Ominously prescient" -Kirkus Reviews
"Riveting stuff" -New York Post