• Register
• Log in

SEO Theory - SEO Theory and Analysis Blog

Algorithm analysis, Web community relationship analysis, SEO practices and techniques, industry news, etc.

• Home
• About SEO Theory
• SEO Glossary

How Microsoft May Change The Searchable Web

Posted by Michael Martinez on February 8, 2008 in SEO Case Studies

How much search market share does Google really control? Before you reach for your latest Compete, comScore, Hitwise, or Nielsen search market share report, stop and consider the fact that none of them accurately measure search market share. So far they’re not even coming close, although they have begun improving their analyses in the past few months.

It’s fair to ask how search market share should be measured. Why is number of queries performed in a month a poor measure of market share? Why does looking at number of visitors/users in a month not really improve the metric? Let’s look at a market in the offline world.

You have two fast food restaurant chains, Marckle’s and Buckley’s. Marckle claims to have served billions and billions. Buckley’s claims to be the favorite. How large is the market? Is it defined by the number of people who buy food at each restaurant? Is it defined by the number of meals sold? Is it defined by the number of dollars earned?

Of course, each of these metrics is measured by industry watchdogs because, well, let’s say that Marckle’s and Buckley’s are both public corporations. Stock analysts look at their financials all the time and want to know where growth will come in the next four quarters.

So let’s look at search engine market share. Ask, Google, Live, and Yahoo! all generate revenues, serve results to, and receive millions of visitors each month through search. But Ask is currently a subsidiary of IAC, Google is its own operating company with subsidiary or secondary operations, Live is a service provided by Microsoft, and Yahoo! is more of a media company than a search giant.

Finding a common ground between the four giants of search is not as easy as identifying what makes Marckle’s and Buckley’s competitors. Marckle’s and Buckley’s do offer similar menus, as do the four major search services. But whereas Marckle’s and Buckley’s mostly sell their food through their own restaurants, Yahoo! and Live both suffer from severe Search Integration while Ask suffers from Search Segregation (IAC still has not integrated Ask search into its properties).

Google builds or acquires new search technology faster than Ask, Live, and Yahoo!. Yahoo! squanders market position more than Microsoft and Microsoft can’t seem to figure out what it wants to do. Ask is the quiet kid in the corner who wants to be heard but so far it can’t figure out how to join the conversation.

We need to look at a combination of metrics to really understand each search engine’s market position: revenues, absolute unique visitors per month, recurring visitors per month, number of queries performed per visitor, and number of satisfied visitors (that is, number of visitors who found what they are looking for). I’d love to know if anyone can figure out how to capture that last metric.

But the other metrics are problematic. For example, Google’s revenues are a matter of public record but do they actually break out the revenues for search? Search Engine Land’s 2007 Q4 analysis for Google doesn’t provide such a breakdown. So how much money does Google actually make through search? If Microsoft breaks out advertising revenues for search, I don’t see where.

Of course, Microsoft’s $544 million in advertising revenues could be compared to Google’s $4.83 billion in revenues (is that all from advertising?) but Yahoo!’s financial numbers are even murkier. There is no comparing apples to apples when it comes to search-driven revenues for Ask, Google, Live, and Yahoo!.

Absolute Unique Visitors Per Month is interesting but not really a trustworthy number since people surf anonymously (through proxy services), from shared IP addresses (at work, school, home, and through wireless hot spots), and through model pools. Also, many people clear their cookies, block cookies, use multiple browsers, and in other ways obscure their online metrical personas.

Recurring visitors per month offers some useful insight because it at least demonstrates a minimal level of user loyalty despite issues with tracking metrical personas.

Number Of Queries Per Month is useless because some search engines induce search fatigue sooner than others. For example, Ask prides itself on NOT indexing content that is relatively obscure. Google, on the other hand, favors less relevant and less authoritative results because they are popular and users “don’t know any better”. Live and Yahoo! have pretty extensive indexes but they seem to favor broader, less tech-savvy audiences so they use different relevance scoring.

As near as I can determine, Google’s Real Search Market Share is probably in the range of 35-40%. That is, 40% of all searchers substantially use Google to find content. Yahoo! appears to have something like 25-30% Real Search Market Share. Microsoft seems to have 15-20% RSMS and Ask probably has between 5 and 10%. There’s a little bit of wiggle room for each of them, and somewhere between 1% and 5% of searchers substantially use other resources.

While many people have their own ideas of how to measure search market share, if we assume for the sake of discussion that my numbers are pretty close (and that they will remain fairly stable over the next 12-18 months), we can see that a Microsoft acquisition of Yahoo! would bring Microsoft into solid competitive reach of Google. In fact, they would be neck-and-neck.

But if Microsoft acquires Yahoo!, Microsoft will have to make some hard choices. Either the media-oriented concept is retained, in which case search becomes less of a priority for Microsoft, or the search functions will have to be broken out from the media functions. Microsoft’s Live.com already separates search from media to a certain extennt (although about as many people use MSN.com for search as use Live.com for search). So a blending of Microsoft’s media network with Yahoo!’s media network could be effected more easily if search is allowed to evolve on its own path.

But there you end up with two brands competing for the same traffic under one umbrella. Yahoo! solved that problem by devaluing the brands it bought up. Altavista’s superior search technology was apparently either abandoned or so watered down through integration with the Inktomi search engine as to be rendered completely invisible in today’s search results. Yahoo! operates multiple search brands but only the Yahoo! flagship brand receives any real attention.

In a Microsoft-Yahoo! combination the media brands would have to unite and that would mean the foundation for the search brands’ visibility would shrink. Consolidation can only hurt Microsoft and Yahoo!’s collective market share. In order to engineer a soft landing for the four brand values (Yahoo! Media, Microsoft Media, Yahoo! Search, and Microsoft Search), Microsoft would have to completely build two new brands and launch them with a smooth transition. Call a baby swap in the middle of the night.

Users would have to be immediately redirected to the full-service media or the full-service search portal. Sub-domains won’t cut the mustard on this kind of transition. The user experience would have to be very good, very satisfying. That means reducing Yahoo!’s in-your-face advertising presentations to human tolerances (the Mindless Ad-clicking Monkey model would have to completely go away). And it means eliminating Microsoft’s AJAX-driven interface.

In other words, if the goal is to retain and build market share the transition has to effectively replace four brand models in user’s minds with as little pain and sense of loss as possible. Glutting the browser window with more gizmos and features will only turn people off. Instead, intelligent choices will have to be made about what to emphasize on the critical entry-level real estate and what will have to be moved to secondary portals (where sub-domains can be leveraged for secondary branding).

I think it would take Microsoft 6 to 12 months to lose about 1/3 of its newly purchased market share. That gives them a window of about 6 months after merging with Yahoo! to build two great resources that currently don’t exist. And those great resources both have to engage users in ways that neither Yahoo! nor Microsoft sites do now. More engagement, less obstruction are the order of the day.

Ask might enjoy a brief opportunity to poach market share from a combined Microsoft-Yahoo! network as inevitable erosion sets in, but Ask lacks the resources to capitalize on competitor vulnerabilities. Ask needs to leverage its brand value across the IAC network. And Ask needs to be less restrictive about what kind of content it is will to share in its search results.

Other search resources that have recently sprung up might also benefit from the almost assured decline in Microsoft-Yahoo! market share, but there are no other resources that are ready to take on the challenge. Hakia, Wikia, and other fairly new, small search services are still proving themselves. And even if someone can and does poach market share from the Microsoft-Yahoo! combination, if they don’t find a way to take it from Google, too, Google will also be in there feeding off the collateral damage.

For search optimizers a combined Microsoft-Yahoo! network will simplify the process of optimization a little bit. Keyword research will become a little less complicated. We’ll only have to track three major search engines’ indexing of content instead of four. But then again, if we lose position in one search engine we only have two other competitive options to work with. As long as Microsoft-Yahoo! maintains its market position, that’s a good thing. But if Google grows its Real Search Market Share at the expense of Microsoft-Yahoo!, everyone (but Google) loses.

We need a robust search environment with real competitors. Google does a pretty good job of serving up search results for some queries but not all of them. Other search engines do better with other queries. People need alternatives. Different groups of people prefer different search engines. A Googleopoly won’t even do Google any good in the long run because it will just make them more unpopular as Google Fatigue sets in.

Nonetheless, now is the time for people to start mulling over ideas about how we as search marketers can adapt to a new search landscape if/when Microsoft and Yahoo! become one party.

Of course, if Yahoo! turns to Google for any reason we’re all doomed. But that’s a post for another day.