<< by on June 8th, 2010
Since Google has the lion’s share of the search market, it is often tempting to focus solely on SEO for Google. But then MSN became Bing, announced a partnership with Yahoo, and started growing its own share of the search market. Bing is now a bigger player than it used to. Furthermore, there have always been certain demographic differences between the users of Google and Bing. Depending on your company’s target audience, Bing may be an even more attractive channel than Google.
For all these reasons, it is now even more important to know how to optimize your site to perform well on both Google and Bing. To do that, you have to know how the two engines differ. The panelists at this SMX Advanced 2010 session break those differences down.
The speakers today on the panel of “SEO for Google vs. Bing: How Different Are They?” are:
- Matt Cutts- Google
- Janet Driscoll Miller- Search Mojo
- Rand Fishkin- SEOMoz
- Sasi Parthasarathy- Bing
Janet Driscoll Miller started off the session with an overview of basic ways in which Bing differs from Google. Why worry about Bing? By joining with Yahoo, Bing is essentially creating a two-engine space. This makes it harder to ignore Bing. The biggest challenge Janet faces with Bing is getting the kind of search traffic she needs for clients, but in certain industries Bing can offer some valuable traffic.
One way in which the two engines differ is in their treatment of Flash. With either engine, you must ensure that your Flash content is accompanied by indexable content. In addition, the same XML sitemap file can be submitted to both engines. However, Bing (unlike Google) does not accept video or news sitemaps.
Bing actually fleshes out certain local search results like weather, movie times, etc. in a format that many people find more user-friendly. However, Bing does not offer the capability for webmasters to remove undesirable sitelinks.
Based on the “page score” that Bing provides in the Backlinks section of Webmaster Center, we can tell that Bing really values links from authority sites in your industry (with similar content to your own) as well as educational institutions. This is a tool you don’t necessarily have with Google.
On the Shopping front- Google allows e-commerce users to submit a feed of products to show up in Shopping results. Bing, however, only accepts paid inclusion in Shopping listings. It also recently did away with its Cashback program for merchants.
One great feature that Bing offers is the preview pane when you roll over certain listings. Bing now also offers playable video embedded in the rollover preview pane- but only for YouTube videos. If you have content on your page that you DON’T want shown in Bing preview panes, you can disable it by adding this meta tag to your source code:
<meta name=”msnbot” content=”nopreview”>
Rand Fishkin shared the results of some super cool research he has done on Google and Bing. The study looked only at first page results. Rand analyzed the correlation significance between different SEO elements (such as the value of h1 tags) and rankings. (A correlation significance of 0 would mean there is absolutely no relationship between the two variables, and a score of 1 would indicate a perfect correlation).
The first element Rand studied was query matching in the domain name (i.e.- including a keyword of focus in the domain name). Google showed a higher correlation for exact match hyphenated domains. Both engines showed extremely high correlations in cases of completely exact match domain names.
How about the effect of using keywords in a subdomain? There seems to be some correlation for Google, but almost NO correlation at all for Bing. The conclusion here is that keywords in the subdomain are not as important as keywords in the root domain. And Bing may be rewarding keywords in the subdomain less than it used to.
As for the effect of on-page keyword usage on rankings, Rand’s findings are somewhat surprising. H1 tags and the title tags do not seem to have any positive correlation for either engine. Period. Google seems to show a higher correlation for the remaining attributes (putting keywords in the alt attribute, main body copy, and URL). Rand concludes that optimizing keyword tags is likely widely employed, and therefore it’s hard to differentiate and get significant value from the best practice.
Rand found that the diversity of links (the number of domains linking to your site) shows a higher ranking correlation (for both engines) than merely the number of links. He also notes that Bing may be a bit more naive than Google when it comes to link valuation, because its correlation numbers are a bit inflated.
Another feature tested was the URL TLD extension (.org, .net, .com, etc.) By far, the strongest correlation was for the .org extension. However, this includes Wikipedia results which we all pretty much agree may have a skewing effect. Rand also notes that the strength of this relationship may have some lurking confounding variables. (Perhaps the content on .org sites is different from other extensions. OR maybe they get more diverse links).
Generally speaking, the longer the URL length the lower the correlation with ranking.
On Bing more so than on Google, there is a higher correlation between ranking with homepages of sites as opposed to deeper pages.
The Takeaway: The overall conclusions agreed upon by Janet, Rand, Matt, and Sasi were that you shouldn’t necessarily approach SEO as a way to optimize specifically for any one search engine. The underlying philosophy for the ranking algorithms of both Google and Bing are to find a way to match the most relevant content to user searches. Keeping that idea in mind for every SEO decision you make is probably a wise way to start ranking well for all major search engines. You don’t have to choose between them.