Search Engine Optimization Track: SEO For Google Vs. Bing: How Different Are They?

It’s day one of SMX and we’re ready to begin in the SEO track with a head to head, no holds barred battle between Google and Bing. Blood will be shed. Marketshares will be insulted. Mothers will weep and hide their children.

Or perhaps I’m just feeling a bit bloody minded because we’re down two livebloggers this week.  Both Lisa Barone and our own Virginia Nussey are on the DL for this conference so I’m going to do my best to be in three places at once. We’ll have some emergency liveblogging help though, so keep an eye out for coverage from familiar industry faces. Barry Schwartz has decided to step up and liveblog as well so there will be coverage at SER for the ‘watch it as it happens’ fans. The SMX team has reserved the entire front row for livebloggers and press so I’m sitting pretty with coffee, a bagel and power.

But enough set up. Here’s the main event.  For the first session of the day, we’ve got moderator Danny Sullivan, Editor-in-Chief, Search Engine Land, ably assisted by Q&A moderator Duane Forrester, Senior Program Manager SEO, Microsoft. (Hey, isn’t it a conflict of interest to have an MS guy doing the Q&A?)

Our Speakers:

Matt Cutts, Software Engineer, Google Inc.
Janet Driscoll Miller, President and CEO, Search Mojo
Rand Fishkin, CEO & Co-Founder, SEOmoz
Sasi Parthasarathy, Program Manager, Bing, Microsoft

This fine cast of search brains will explore the battle of the only two engines that will soon matter once Yahoo goes away (sorry, Ask).

Danny starts off — “I feel like Liza Minelli in Sex and the City 2.” Silence. “Half of you saw that, ten percent were dragged to it.” Hee.  He teaches us how to use the Conference Guide (Read the page that says “Read this first” first.) and how to use hashtags on Twitter #smx for the conference, #1a1 for the particular session.

Janet Driscoll Miller is up first.  She’s going to be doing the very high level overview of the session, setting up for Rand to get into more detail.

Why worry about Bing?

We’re turning into a two engine world. Bing and Google have a very similar usability profile.  But Bing outperforms Google in many areas — except for search volume. They still don’t send a lot of traffic but they do send good traffic.

Bing v Google for SEO tactics

Similarities:

• Both don’t do well indexing Flash still.
• XML Sitemaps: You should be using them. They’re a universal standard and you should learn how to use them. You can submit your Sitemap to Bing and Google through their webmaster center/central.
  • Bing does not currently accept video or news Sitemaps.
• Local indexes are easy to submit to.
• Geolocated results: Bing will tell you the source, you can use that to figure out where you need to be to appear in their results (snow conditions from OntheSnow.com, etc.)
• Both have sitelinks but only Google allows you to edit them.

Differences:

• Bing Webmaster Center: Sitelinks, outbound links, link value on inbound links — gives you insight into what Bing values in links. (You can only download the first thousand results.)
• Getting into Bing news: There’s no formalized process yet. You have to email bns@microsoft.com and create an RSS content. It may take a long time to get a response. Keep pinging.
• Google Shopping/Base = Free
• Bing Shopping = paid only. Also, Cashback is going away.

New opportunities for Bing:

• Social sharing and results: Search for polar bears brings up images and there’s a “share on” various social networks functionality.  However, the links are back to Bing, not the original image. To post on Facebook, you have to grant access to the Bing application on Facebook.
• Document preview: Allows you to put more marketing data in there. You can put really great information in the preview and entice people to click on rollover. YouTube videos will play on roll over.

How do you optimize the preview? Bing pulls information in this order:

1. Takes the H1 if it doesn’t match the Title tag
2. First paragraph of information
3. To add contact info, just add that information to the page — Address, phone, email. They’re good at figuring it out.

To disable the preview, add <meta, content=”no preview”> to the page.

Rand is up next to explain the nitty gritty differences between the engines.  He’s covering ranking factor correlations: Google vs. Bing.

Methodology: 11,351 SERPs via Google AdWords Suggest, 1st page only (usually 10 results a page), Correlations are with higher position on page 1, Controlled for SERPs where all or none of the results matched the metric (for example, controlling for .org or .edu.) Data came from commercial terms, almost no brands.

Understanding correlation significance: 1 to 1 is perfect correlation, ie that’s the only thing the engine cares about. They didn’t get that. Mostly it was less than .3

Note: correlation does not equal causation. Just because something has negative correlation it doesn’t mean that you need to change it or that you should ignore it.

Query matching in the domain name:

Their conclusions: It appears that Bing has slightly less correlation with exact match in the domain name overall. Exact match domain remain powerful in both engines (anchor text could be a factor, too.)

Hyphenated versions are less powerful, though more frequent in Bing (G: 271 vs. B: 890)  (IE, they show up on the first page more often in Bing but they rank better in general on Google when they do show up.)

Just having keywords in the domain has a substantive positive correlation.

Exact match domains by TLD – .net vs .org, etc.  If you’re going to register an exact match domain, go with the .com. The others don’t have as high a correlation.

Keywords in subdomains: Higher correlation for Google than Bing (in fact, almost inside the standard deviation) but nowhere near as high as ExactMatchDomain.*  Keywords in the subdomains are not nearly as powerful as in root domain name. (There is a typo on his slide and it’s driving me crazy. Less THAN, Rand. Not less then.)

On-Page Keyword Usage: (scale changes here – maximum of 0.06 with a negative correlation in some cases)

• The ALT attribute of images is interesting — seems like it should be part of best practices.
• Putting KWs in the URL is probably also a best practice.
• Everyone optimizes Titles, differentiating is hard.
• (Simplistic) On-Page Optimization isn’t a huge factor [Um….no. It means everyone on the first page has relatively equal on-page. Not that you can ignore it. That’s why it’s called a best practice.]

Links:

• High correlation in # of linking root domains to URL.
• Links are still a major part of the algorithm.
• Bing may be slightly more naïve in their usage of link data than Google, they seem to care more about raw links.
• Diversity of link sources is important.

TLDs:

• .org has high correlation, .edu has low correlation.

Length:

• Content length: not substantial
• URL length: negative correlation
• Raw content length seems marginal in correlation
• Long domains might not be ideal but aren’t awful
• Shorter URLs are likely a good practice, esp on Bing.

Website home pages: Bing likes home pages more than Google does (over internal pages).

Anchor text:

• # of links with exact match anchor text — very low correlation for both
• # of linking root domains — again supports diversity. Very high correlation.

Features with highest correlation:

• Link attributes have a much higher correlation w/ rankings than on-page or domain-related elements.
• Exact match is still a powerful influencer
• Google and Bing are remarkably similar — you don’t need to optimize differently for them really.

Now it’s time to go to the search engines and hear their take.

Matt ums and uhs for a moment and calls Rand’s data “interesting”. He says you shouldn’t chase after search engines, you should chase after user experience.  They joke about how they should have done a correlation on Wikipedia in the number one position and Matt thinks that Bing shows Wikipedia than Google.

Matt also cautions that SEOMoz’s data is very niche and not at all broadly applicable. If you’re thinking long tail or non-commercial spaces, you need to think more critically about the data. Matt also tweeted this: “I’m on an #smx panel about the differences between Google and Bing. Here’s my quick take on the topic: http://goo.gl/tRdE”

Sasi agrees with Matt and says it’s not about more “bing-like” or more “google-like”. It’s about what’s right for the user. They would focus on different metrics than SEOMoz,  They would focus on things beyond seo metrics. Query rate, long tail, etc.

Rand says that if you’d like to have more information on any metric, you can email them to have them add it to their sample.  This data came after the Mayday release (it was done this past weekend).

Someone asks if Bing will kill Yahoo Site Explorer. Sasi says they’ll have the functionality still and that SEOs are users too.  Danny asks if people would use Bing to get Yahoo Site Explorer data, about half the room says yes and Danny tells us that if it goes away it’s because we didn’t raise our hands.

Matt and Danny clarify again that Yahoo and Bing will have identical organic results (refer back to SMX West).

Matt: Numbers at the beginning of your URL is harder for branding. If you can find a nice brandable domain, that would be a little bit better.

Matt: Google is going to be looking at Video Sitemaps more closely and they will be really trying to get videos indexed particularly with Google TV coming in the Fall.

Susan Esparza is former managing editor at Bruce Clay Inc., and has written extensively for clients and internal publications. Along with Bruce Clay, she is co-author of the first edition of Search Engine Optimization All-in-One Desk Reference For Dummies.

See Susan's author page for links to connect on social media.