Why the discrepancy in search engine market share results?

February 24, 2006 7:04 PM

Every time i see statistics like this one which list market share for various search engines, I'm somewhat surprised. According to the article/graph, in mid-2005, Google had a 36.5% market share in online searches, with Yahoo and MSN at 30.5% and 15%, respectively. But on every website I've managed for myself or others (and on the sites of everyone I've ever spoken to about this), Google consistently accounts for 80 - 90% of all their search referrals. So, what accounts for the discrepancy? Is it just because my evidence is anecdotal, or is there something special going on in which Yahoo/MSN searchers get sidetracked before ever reaching their search results?

I understand that, for example, Yahoo! Sports searches count toward Yahoo's overall total, as explained in the article. But I would assume that would be evened out by the alternative searches Google provides as well (groups, froogle, print, news, etc).

Nielsen claims Google has a 48% share, which is closer, but nowhere near 80% - 90%.

posted by helios to computers & internet (18 comments total)

Of the few people I've known who used non-Google search engines, most were fairly casual web users who spent their time on a small number of very high-profile websites (CNN, New York Times) etc. I think a significant group of Internet users never go much behind major media sites, and I'd also expect this group to have a higher proportion of MSN Search and Yahoo users.
posted by IshmaelGraves at 7:13 PM on February 24, 2006

What is the subject matter of the sites you and your acquaintances have managed? Audience might account for the skew.
posted by Good Brain at 7:31 PM on February 24, 2006

I just launched a new site, and I'm seeing a lot of MSN search referral traffic for some reason. There's a lot of Google and Yahoo too, so my money would be on anecdotal skew in your data.
posted by willnot at 8:11 PM on February 24, 2006

George Will once told the allegedly true story of the woman who was heartbroken on Election Night 1984 to learn that Walter Mondale lost the race for president. "But everyone I know voted for him!" she cried. Whatever you think of the people mentioned, I do like that story for how clearly it demonstrates the danger of extrapolating from one's own circle.

It's easy to understand MSN -- it's the default home page for IE, and some people just don't care or don't want or even know that they can change it. Google attracted a smart group from the beginning. And Yahoo -- well, Yahoo was a lot of people's introduction to the Web. Plus Yahoo Mail has a lot of users (deservedly so).

Which is pretty much a longer-winded version of what IshmaelGraves and Good Brain have said, I guess.
posted by pmurray63 at 9:12 PM on February 24, 2006

Market-share statistics for particular countries are often different than worldwide statistics, too. (The ones you gave were from a survey of "English-speakers worldwide.")
posted by mbrubeck at 9:30 PM on February 24, 2006

Google had a 36.5% market share in online searches, with Yahoo and MSN at 30.5% and 15%, respectively. But on every website I've managed for myself or others (and on the sites of everyone I've ever spoken to about this), Google consistently accounts for 80 - 90% of all their search referrals.

You're comparing apples and oranges, it seems to me. Or perhaps I mean you've answered your own question?

Google only has 36% of searches, but produces 80% of results. Where's the discrepancy? A result comes from someone saying "yes, that's the thing I wanted" and clicking on a link. Perhaps that just happens more with Google than with the others because Google is better.
posted by AmbroseChapel at 9:34 PM on February 24, 2006

When you search with Yahoo! or MSN, often the search results simply aren't as good as Google's (although both have narrowed the gap somewhat). So maybe your click-throughs reflect that the people who found you from Google and were looking for what your site offers.
posted by evariste at 9:44 PM on February 24, 2006

Damn, AmbroseChapel beat me to the punch. And here I thought I was a clever boy.
posted by evariste at 9:44 PM on February 24, 2006

AmbroseChapel : "Perhaps that just happens more with Google than with the others because Google is better."

And you would expect users to start recognizing that and switch over, hence converging the two measures.
posted by Gyan at 10:59 PM on February 24, 2006

I only use Yahoo for searching their groups, but they are are really useful and I use it quite a bit. Maybe that accounts for some fo the difference?
posted by fshgrl at 11:10 PM on February 24, 2006

After years of working in undergraduate computer labs and public libraries, I'm firmly in the IshmaelGraves/pmurray camp.
posted by box at 5:32 AM on February 25, 2006

Yahoo usage is higher among women and also higher among people who are in e-commerce window shopping mode.

Google is used by the more technical, for finding a specific item and research and business related activities.
posted by Mick at 8:08 AM on February 25, 2006

One or both of two possibilities:

1. You're higher ranked on the google searches, so the MSN and Yahoo searchers never reach you.

2. Google searchers are of a demographic more likely to be searching for what you're offering on your website. (Others have suggested this above.)

NB that your website is not a microcosm of the web as a whole.
posted by commander_cool at 8:21 AM on February 25, 2006

I also see consistently that Google accounts for 80 - 90% of the search referrals, and everyone I know uses Google.

If you'd check the Alexa traffic details you can get a better picture, which confirms the 80%-90% traffic.
(search.)google.com 76%*271,900=206644
search.yahoo.com 8%*276,650=22132
search.msn.com 8%*263,100=21048

Conclusion: "There are three types of lies - lies, damn lies, and statistics."
posted by Sharcho at 9:17 AM on February 25, 2006

Yahoo! handled 19 percent of global Internet searches in November, a drop from 27 percent a year earlier, according to Web tracker ComScore Networks Inc.

Google's share, by contrast, rose to 60 percent from 47 percent.

BLOOMBERG NEWS Jan 2006
posted by Lanark at 1:33 PM on February 25, 2006

Yahoo is very large overseas....
posted by crewshell at 2:04 PM on February 25, 2006

you would expect users to start recognizing that and switch over, hence converging the two measures.

No, I wouldn't. Most people aren't like that. They use whatever search engine they're used to. They don't follow tech trends, they don't know that one search engine might be different to another or know that there even are different search engines.
posted by AmbroseChapel at 2:46 PM on February 25, 2006

Just for reference, the "80-90% google" that I'm referring to is not just for my own site (which is ~96% google), it covers sites of friends, co-workers, and ones I've managed in the past, covering law, music, blogs, personal pages, sports, "talk about anything" forums, etc.

AmbroseChapel has a good point - perhaps Yahoo! searchers, for example, simply see the results and don't find anything meaningful, so they stop searching.

And IshmaelGraves, perhaps you're right. Google has much less incentive to keep searchers within its own network than Yahoo, which has a site for almost any conceivable topic. Maybe Yahoo/MSN are just skilled in keeping people within their own little world.

It'd be interesting if some group did a more in-depth analysis on this topic than a simple "xx% market share."
posted by helios at 4:59 PM on February 27, 2006