Best SEO Blog|Launch your site.

If anyone still doubts that Microsoft is the second most popular search engine today, the Real November 2008 Search Market Share Report knocks that doubt into yesterday. Microsoft served an estimated 100 million search visitors in November and that’s just too big a market for SEOs to ignore.

But how do you capitalize on a market that is so well obscured by the myths and preconceived notions of the SEO community? The trick here is to understand that most SEOs wrongly believe that Google controls about 70% of the search market — so they tend to focus their efforts on Google and just assume that basic SEO works well enough for Microsoft and other search engines.

That’s where you have the advantage.

Through the years there have been numerous studies concerning search index overlap. Researchers have shown repeatedly that each of the major search engines miss large portions of the Web that the others index. The probability of two search engines indexing entirely the same data is extremely low.

So armed with that knowledge, here is how you optimize for Microsoft:

First, start your linking resource research on Microsoft. Be careful to understand that I’m talking about linking resources. This is not backlink research (which is a waste of time). All you’re concerned with here is finding the types of Web sites you can use to develop launch links in Microsoft’s index.

Secondly, develop content that appeals to Microsoft’s user base. If you are working with existing Web sites you can look at your raw server logs to see which queries people use to find your content from Microsoft (Google Analytics won’t reveal this to you — nor will any standard Web analytics software). Knowing which queries Microsoft search promotes your sites for helps you zoom in on where you met Microsoft’s criteria.

Thirdly, take advantage of Microsoft’s site search functions. As long as you get your content indexed in Microsoft, using their site search ensures you have a testbed in which to alter your optimization techniques to ensure Microsoft positions the best pages first. You can watch as changes to your pages are indexed and gauge how fast Microsoft works to integrate your content into their index.

Now you’re ready to set a dual-path optimization strategy into motion. That is, you’re going to develop content and links for two Web sites and watch how your optimization works with Microsoft. Why two? Because you need to test your conclusions and assumptions. Microsoft is going to crawl your new content on the basis of its own priorities.

If you’ve been seeing relatively little traffic from Microsoft search it’s not because they don’t refer people to Web sites, it’s because you haven’t been optimizing for Microsoft search. My personal network receives a lot of traffic from Live.com and search.msn.com because I made the conscious decision to optimize for Microsoft search two years ago when everyone else was laughing at Microsoft’s decision to launch live.com.

In today’s search world I can live without referrals from Google because I still get substantial traffic from other sources, including Microsoft and Yahoo!. If you cannot say the same thing it’s time for you stop paying attention to bogus search market share estimates and start optimizing for the real search market. Don’t fuss over who is laughing at Microsoft’s market share — they’re obviously not getting a very large piece of Microsoft’s 100 million search visitor pie, either.

There seems to be a growing interest in SEO for subdomains. I find that ironic given that, after many years of tolerating subdomain abuse, Google has over the past 12 months or so tightened its requirements for subdomain visibility and inclusion. I would say a lot of people have missed the boat, but there are certainly still legitimate uses for subdomains and, when their content becomes large enough, there are legitimate uses for sitemaps on subdomains.

So let’s talk about sitemaps for subdomains. We’ll start with traditional HTML sitemaps.

Use HTML sitemaps on any subdomain with more than 10 pages - You can include an HTML sitemap on a smaller site but it doesn’t provide much value. However, once you get up to the 10-15 page volume for a site, your on-page navigation becomes clumsy and inefficient for your visitors. Yes, search engines can still crawl 30-50 links on a page and they are happy to do it, but you need to keep your users in mind. They tend to be overwhelmed by more than 10-12 navigational options.

An HTML sitemap can act like an extension of your primary site navigation. You link to the most important content in your sitewide navigation, and your HTML sitemap should be included in your idea of your most important content. Make it easy for people to find the HTML sitemap. These are among the few types of pages where people are mentally comfortable with navigating through dozens or hundreds of links.

Speaking of hundreds of links on a page…

Paginate large HTML Sitemaps - Just because you CAN put 300 links on a page doesn’t mean you should. It’s okay to paginate your HTML sitemaps when they become very large. Google recommends keeping your links to around 100 per page. That’s as good a limit as 150, 200, or 300 but it’s more user-friendly. Think of ways to organize your paginated HTML sitemap so that people can quickly and easily find what they are looking for.

If your site is extremely large, with hundreds of thousands or millions of pages, you might consider creating a subdomain that is used as nothing other than an HTML sitemap. Think of how you could structure and populate the pages so that people want to use them. You might use a date-oriented structure to archive aging content, a topic-oriented structure, an alphabetized structure, or a hybrid structure.

Include subdomains in your primary HTML sitemap - If your main site is large and has its own HTML sitemap, it’s okay to include at least the top-level content from your subdomains in your sitemap page(s). If the subdomains are integrated into your architecture and design, it makes sense to include them in your primary HTML sitemap. You don’t need to include everything on a subdomain in that highest level sitemap. You can deep-link to your subdomain’s HTML sitemap pages and its root URL and leave it at that.

XML Sitemaps for Subdomains - If you have enough content on your subdomain to justify an HTML sitemap, you have enough content to justify an XML sitemap. You can link to your subdomain pages from your primary subdomain XML sitemaps but you may find you have to authenticate each subdomain with the various search engines (Google, Live, and Yahoo! all allow you to do this).

Subdomain sitemaps can be hosted on the subdomain or primary domain, but again you may run into search engine restrictions. Be sure you authenticate your sites properly and try to manage your subdomain XML sitemaps on the subdomains themselves.

Paginate XML Sitemaps - There is divided opinion in the SEO community on whether you should include as many URLs as possible in XML sitemaps or if you should break them up. Think about this in terms of what is easiest for you to manage. On a large content site you’re most likely going to have sections by topic, date, or some other categorical structure. I would recommend dividing your URLs into XML sitemaps that match your major site structures.

You can do this on large content subdomains as well. The rule of thumb is, ask yourself which is more unwieldy: scrolling through 50,000 blocks of XML code or updating 1,000 XML files.

Only repeat your most important pages in multiple XML sitemaps - An XML sitemap helps a search engine identify URLs for crawling. If you have more than one XML sitemap on a site, I see no reason not to include the root URL and HTML sitemap in every XML sitemap. I do this myself. I feel strongly that covering those URLs as much as possible is better than assuming I have covered them exactly where they need to be covered.

Restrict access to your XML sitemaps - Quite frankly, there are a LOT of rogue robots out there. They gobble up XML files like locusts on a field of grain. You can protect your XML sitemaps from the rogue robots by using .htaccess files to restrict which search tools have access to your crawling lists. Let the major search engines see what you have on your site and tell the rogue robots to go fly a kite.

I have often used XML Sitemaps.com to create small XML sitemaps, but you may need to develop a custom solution or find another tool for truly large sites.

I think it’s interesting that people are now appending year numbers to the queries they type into Google and other search engines. I thought I was the only person looking for specific information within specific chronological periods. Great minds think alike, I suppose.

So here we are at the end of 2008 and people want to know how to build trusted links to their Web sites. And they don’t want yesterday’s SEO advice on the subject — they want current, 2008-era advice on how to build trusted links to their Web sites.

The whole question (”how do I build trusted links to my website”?) raises the issue of how obsessed people must still be with links. You would think that at least some people would have gotten the message by now that it isn’t all about links. After all, as more people look for sources of trusted links, they must understand that all those cheap, easy-to-obtain links they were building previously don’t pass value.

Okay, we can leave the lecture on SEO efficiency for another blog. If you really want a 2008-era method for obtaining trusted links to your site, here are some suggestions:

1. Avoid social media Web sites like the plague.
2. Create value in whatever sites you contribute to.
3. Reward the sites that help you.
4. Focus on creating content that helps the sites you want to link to you.
5. Think more in terms about building sites that can help themselves

Sites that search engines trust are trusted because so many people use them. It’s not that the search indexing algorithms are tracking visits (although that may be one factor taken into consideration when assigning trust), it’s that a heavily used site looks different from a site that people avoid like the plague.

Heavily used sites have a lot of activity — they publish original content often, they link out to other sites liberally, they have activity in their communities, they attract many low-value links naturally.

The less traffic a site receives, the less likely it is to publish new content. The less new content a site publishes, the fewer new outbound links it creates. The fewer new outbound links a site creates, the fewer referrals it sends to other sites. The fewer referrals it sends to other sites, the less likely other sites are to recognize its contributions to their communities, the less likely people are to mention the site casually.

You can’t even get low-value links if no one thinks you’re worth linking to.

Now, when I say “low-value links”, I’m not talking about links from article archives, press release distribution services, social media sites, and “seo-friendly” directories — those are all crap links. There is at least some value in a “low-value link” — a link from a Web site almost no one has heard of, which doesn’t have much content, and which only gets a trickle of traffic.

People in the SEO community all want to be on the front page of DIGG and Sphinn. Screw DIGG and Sphinn — the amount of time you spend submitting your whitless articles to those sites could be better spent cultivating relationships with small, low-visibility sites that are trusted because they’ve been around long enough to have established good track records in the search engines.

Of course, those sites may not link to crap content so you may be stuck submitting to DIGG and Sphinn until you figure out the difference between good content and cheap, off-the-cuff content.

You build trusted links to your site by earning the trust of Webmasters whose sites are deemed trustworthy by the search engines. Competing for links from the same influencers everyone else is chasing is just plain stupid. There are millions of trusted Web sites out there, and most of them are being ignored by the link-greedy SEO community.

Make sure you have some good content, and then share it strategically, helping to build value in sites that can help your sites. That doesn’t mean changing your tactics from submitting badly written schlock articles to archives to submitting them directly to unsuspecting Webmasters. It means you should think about ways you can actually help other sites — without lecturing them.

Those Web sites will link to your sites when they see the value you help them create for their visitors.

Many Web sites that mix dynamic data on their front pages discard massive amounts of indexable data on a regular basis. Go to any major news site like CNN and ask yourself: what happens to that front page image when the headlines are refreshed?

There is, in fact, historical value in preserving mashup page images as static HTML pages. These pages serve as special index pages to historical content on high production Web sites. The spirit of Web 2.0 has taught Webmasters to invest their resources in throwaway information that exists for a brief time and then vanishes completely.

The unique organization and compilation you produce in your on-site mashups of new content every day (or every hour) may, in fact, serve as exactly the kind of unique content you need to boost your site’s internal navigation.

If a site publishes more than 10 new articles a day, there is intrinsic value in preserving the original “front page” mashup that introduces visitors to those articles. However, in today’s framework-pattern of constantly scrolling sites, the story excerpts and images are typically dropped completely while headlines are added to scrolling “archive” pages.

Those archive pages slowly repaginate their contents into long chains of almost useless link lists.

If you have the ability to create a mashup of new content for your home page, you should have the ability to create a snapshot image of that page that can be archived for future indexing and reference. People do, in fact, visit such snapshot pages as they search for older content in its original context. You can depend on archive.org to do this for you on a sporadic and incomplete basis or you can do it for yourself.

There are undoubtedly some practical limits to how many mashup snapshots one should archive. For example, I am not sure how useful it would be to archive every change made to a mashup. If you can capture all or a majority of your scrolling content on a daily or weekly basis, I think that would be easier for your visitors (who must, after all, search through the archive — don’t make their task more difficult by overloading it with duplicate content).

Spammers have in fact been creating snapshot archives on their spam blogs for years. I find these types of sites all the time, but their snapshots are built around RSS feeds scraped from other people’s sites and are pretty easy to identify. Real searchers are not interested in that type of content. But if they visit your site to read your articles to begin with, then your articles must have sufficient intrinsic value to be worth indexing — and thus your front page mashups also have intrinsic value.

Preserving that value helps you ensure that older content stays well-connected in your site and available through site search and alternative on-site navigation.

Matt Cutts recently told WebPronews that Google is relatively satisfied with its current subdomain management in search results.

In other words, they set out to do something specific and they feel they have done a pretty good job. But subdomains continue to dominate some brand queries. A lot of SEOs were predicting as recently as only a couple of months ago that some day nearly all the subdomains would vanish from Google’s search results.

In fact, subdomains seem to be thriving in Google’s search results. It’s just that the subdomains we find tend to provide more unique or distinctive value than they might have in the past. Or not.

Okay, no system is perfect but it does look like Google has made a good faith effort to help searchers find differentiated content when subdomains still dominate the search results. The search engine optimization strategist can still rely on subdomains when taking uniqueness of content into consideration. So the old approach of using a CMS and template with keyword substitution may not work as well as it once did, but you can still target subdomains toward keywords provided you back off on your use of boilerplate.

Subdomains remain a two-edged sword for optimizers. They have been used for geotargeting essentially duplicate content, as well as for differentiating content by seller or provider. But subdomains can be used in many other ways.

For example, a complex ecommerce site can utilize subdomains to provide different presentational formats for its product categories. People do have different expectations when it comes to searching for health and beauty products than when searching for books or videos. Although it may be convenient to use only one CMS template for an entire site, you can distinguish your category brand values by giving each their own templates, their own site search tools, and their own subdomains.

The optimization benefit arises from the flexibility not the uniformity that subdomains offer you. You can create essentially unique Web sites under one domain umbrella. Some of those sites can provide support, community, and research information. Some of those sites can focus on sales, marketing, and building business relationships.

Subdomains should not be viewed as mere extensions of Web site root directories. They offer legitimate platform branding value that cannot be matched by subdirectories. You use use subdomains to make it easier for people to find the right content by branding the content with a recognizable subdomain.

Just make sure the content really deserves its own subdomain.