More writings on gTLDs and SEO

I recently wrote about why there's good reason to believe that controlling your own gTLD can positively impact SEO.  This socialmediatoday post covers much of the same topic, and SearchInsider goes into depth on the idea that you can create an entire gTLD that is optimized to do well in search.

Finally, I went into detail in my first post about why to believe that quality content in a TLD might be rewarded in search.  We also have seen the opposite, in which exceptionally poor content can punish an entire TLD.

New gTLDs and SEO, part 2 of 2

Last time I described the reasons to be optimistic that a well chosen gTLD filled with quality, relevant content can become an SEO enhancer.   I don’t know that  your own TLD will immediately cause your pages to rocket up the rankings on the first day you go live (although I don’t know it won’t either), but no matter how Google treats your TLD in and of itself, there are definite additional opportunities to improve search engine optimization if you control your own domain space.

Opportunity #1: Increase search term density in domains

One well understood SEO technique is to maximize the density of search terms in the domain.  That’s the idea behind using subdomains for SEO purposes.  By eliminating useless words like .com, you can increase that density.

Opportunity #2:  Place the most critical search terms in the second level domain

A common Google behavior is to grant better position to pages with the searched term in the second level domain.  That’s why microsoft.com does so well on the term microsoft.  Unfortunately it’s a difficult fact to make use of since any term with competition for search position most likely doesn’t have the appropriate domains available.

Once you control your own domain space, that all changes.  You can generate as many pages as you need to, focusing each on the most perfectly relevant term you seek to win.  You can add and remove pages and content as you will to optimize your results.

Opportunity #3: Generate more clicks in any given search position

I’m not aware of any research looking specifically into this question, but it seems likely that a friendly domain will gain more clicks in any given search position than an old-fashioned TLD will.  Let’s say you search on laptop.  The page at laptop.bestbuy is highly likely to contain useful product information and be a place you can purchase.  That’s a strong cue for a consumer to choose this result over other results.  Or let’s say a search turns up login.usbank (or for that matter login.paypal).  I can imagine the typical visitor feeling more confident that she’s visiting the real bank and not some phishing site - and therefore choosing this option over others.

Remember, nearly 100% of the extensive SEO effort that goes on in the world is for purposes of bringing more visitors to your site.  The traffic is the true goal.  Search position is just a method of gaining this traffic.  If there turns out to be a method of increasing clicks from the same search position, that’s tantamount to improving your position in search results.  Being in the #9 spot but getting the same number of clicks as the guy in #8 is every bit as good as being #8.  Over time, especially as consumers learn to look for your friendly domain, I expect that preference will just continue to increase.

Note also that these three opportunities apply to your paid search results as well.  SEO value influences search placement, potentially putting your listing higher than just your pure bid value would lead one to expect.  And increased preference for your listing can increase the return on your SEM spend.

Future proof your SEO strategy

Of course, we’ll have to see exactly how everything shakes out.  The algorithms don’t exist today, as do neither the TLDs nor the content under them.  So we’ll all be exploring these ideas together.

One thing about which there is no question, however, is that only those of us who obtain our own TLDs will have the opportunity to participate in this upside.  Whatever SEO improvements the world identifies, only those on control of a domain space will be positioned to take advantage of them.  Right now companies must decide if they want to be a part of that progress, however it shapes up, or if they prefer to sit on the sidelines and let others overtake them.

New gTLDs and SEO, part 1 of 2

There is a great deal of interest and debate surrounding the question of how the new, upcoming gTLDs will affect search engine optimization (SEO).  With quite a bit of help from some of my new colleagues at Melbourne IT Digital Brand Services I’ve looked into this question, specifically in regard to gTLDs that are in the control of the organizations who are using them (as opposed to new community domains like .bank).  I am confident that properly used TLDs represent an opportunity to improve search engine performance.

First the caveats:

• All SEO techniques are just tools, and they need to be used well to yield good results.
• No SEO technique is a miracle worker.  Some ambitions will be outside your grasp no matter what you do.  Nobody but Apple Corporation will be ranking first on the term ipod any time soon.
• Results may vary.  Some people will get better results than others.  That’s how it goes in SEO-land.
• SEO takes time.  Results are never instant, so you have to be patient and determined.
• SEO is a subtle and elusive practice.  You’ll have to get your hands on your own SEO program with your own content and your own target keywords to optimize it for your needs.  There is no other way.

Google keeps mum on its search algorithms for very good reason, but we can draw some sensible conclusions based on our reasoning and behaviors we can observe in Google search.

For starters consider these facts:  Google is in the business of using all available information to offer the most accurate relevancy possible in its results.  TLDs have the potential to be giant screaming indicators of a site’s content.  Therefore it’s hard to imagine that the rocket scientists in Mountain View will categorically ignore this indicator without at least checking it out to see if it makes results better.

We have good evidence that Google is using TLD information to influence search results today.  For example, you can easily find a myriad of instances in which preference appears to be given to .gov and .edu sites for appropriate search terms.

Search on texas.  The top result is the state of Texas’s site at www.texas.gov.  Second is Wikipedia’s entry on Texas.  Now, there’s no way that the state of Texas has optimized its site to the level where it can go toe to SEO toe with Wikipedia.  What’s happening here?  Google is giving great weight to the fact that this site has a .gov domain.  (Note that texas.com doesn’t show up until #30.)

If you believe that in Texas maybe civil servants are SEO savants, then I’ll point out that the same happens for new york city, memphis, phoenix, dade county florida and many, many other place names.  The common thread?  Place names and .gov.

Search on harvard and you see the same exact thing with .edu.  This time Wikipedia is #5, behind four .edus (and with more .edus below).  Again, we see the same for princeton, ucla, johns hopkins, and lots of other educational institutions.  Again, it’s the .edu addresses.

I consider that pretty compelling empirical evidence that Google is prepared to use TLD information to influence search results when it feels they improve the results’ accuracy.  But on top of that Google has, in fact, applied for a patent on using TLDs as a contributor to search relevancy.  Patent application 12/468,195, filed by Google Corporation on June 17, 2009 states in part,

These pages can be differentiated and identified by, for example, a list of domain names; top level domain extensions, such as .biz, .com, .org, .edu; or web sites.

These two pieces together tell us that Google explicitly has considered how TLDs enter into determining search ranking and that the search engine uses this information today.

Now, how do we know what Google will do with any individual TLD?  We don’t.  But returning to the earlier point, we know the search engine’s mission is to index and present all available information in the manner that’s most useful to searchers.  If the entity that controls a certain domain space uses it well, we should expect in the long run that this TLD will become a useful indicator.  Likewise, if the entity floods it with useless content, then that TLD may actually be punished by Google.  For companies that control their own gTLDs that’s good news since they can ensure that this domain space is filled with relevant, useful content.  In fact, the content they would include even without SEO considerations is very likely to fit that description.

In my next post I’ll explain other opportunities to improve search engine performance using your own TLD.

The one standing order that all knowledge workers should have

In my career I've held and managed a lot of knowledge worker positions, particularly positions where a considerable amount of strategic thinking is required for success.  Over the years I've evolved a single standing order that everyone who works for me has.  It's quite simple:

Always know and be able to articulate the reasons you have for making the choices you do.

That's it.  As I've written in the past, you hire brains, not bodies, so it's essential that these brains are switched on and doing what they do best.  It's easy for anyone - especially as we become comfortable in our jobs - to kind of turn it off and fly on autopilot.  But since knowledge workers essentially have been hired to think, if you go onto autopilot you're not doing your job anymore.

Note that the rule is to know and articulate your reasons, not anticipate what reasons I would come up with.  That's a critical point.  I don't have the reasonable expectation that nobody who works for me will make a mistake ever (and I wouldn't want them to be that cautious anyway), and I also don't expect anyone to channel my own decision making and become a Tim Callan satellite.  Not only would these expectations be impossible to deliver on, but they wouldn't yield the best results anyway.

But it is reasonable for me to expect that you're always doing the best you can, that you're taking the knowledge, experience, and reason available to you and using them make the best decisions you're able to make.  So long as the people who work for me are doing that, we're good.

New gTLDs are on the way

We're about to reach a sea change moment in internet naming.  Starting early next year, organizations will be able to apply for their own Top Level Domains (TLDs).  TLDs are the rightmost strings on internet addresses.  We're all familiar with .com, .net, .edu, and the like, but if you include country code TLDs (ccTLDs), there are nearly three hundred of them in total.

This system has clear inherent weaknesses.  How do I know if I need to find a specific domain at .com or .net?  Do I look for a non-US business under .com or the appropriate ccTLD?  What is the relationship between physical geography, national jurisdiction, and an internet domain?  Is bit.ly really in Libya?  What percentage of .co, .me, and .tv domains are for entities located in Columbia, Montenegro, and Tivolu, respectively?  And how come nobody seems to have a .us domain?

Not to mention the bigger problem of what happens when all the useful words are taken.  Believe it or not, we're facing that problem on popular domains like .com and .net.  In particular, all legal strings of five or fewer characters are owned by someone on the .com TLD.  That presents problems for businesses, charities, individuals, etc. who are looking to create sensible web addresses to communicate with the public.

I've long imagined that this naming system would require some kind of alteration as time went on.  We're seeing a step in that progress (though surely not the last) coming up next year.  When internet naming corporation ICANN opens for applications early next year, interested parties will be able to apply for their own TLDs.  Since they won't be ccTLDs, they will go by the term general TLD, or gTLD.

New gTLDs could serve a number of functions.  The obvious ones are naming spaces to serve a specific customer or type of content, such as .bank or .secure.  There also is a great deal of interest in TLDs for cities (.paris) or for specific companies.  It's easy to imagine the benefit of, let's say, .sony.  In this case Sony could stand up sites at viao.sony, playstation.sony, movies.sony, customerservice.sony, etc.

Pundits are speculating that there will be thousands of gTLD applications, and I believe it.  I, for one, am going to be fascinated to see how this whole things plays out.