Since the beginning of search, keyword optimization has been king when it comes to SEO. Whether it was keyword stuffing in the early days, or counting words for the right keyword density, keywords are the obsessive bane or joy of website optimizers. Even less understood advanced concepts such as latent semantic indexing (LSI) still at their core are all about keywords and keyphrases. But in the next few years search engines are going to add a new component that will make SEO even more exact and potentially difficult. Both Google and Yahoo have filed patents concerning “Phrase Based Indexing”. Phrase Based Indexing is an attempt to recognize a relevant set of grouped terms in a section of text which represents a known concept stored in a central concept “dictionary”. What this means is search engines will potentially be able to determine a pages relevancy to a particular keyword without actually containing the keyword. For instance, a page containing the words “brakes”, “engine”, and “cylinders” would be recognized by the search engine as a page that relates to the keyword “cars” and return the page to searchers who search for “cars”. For SEO, not only will this change the way we write and organize our content, it could also blur the obvious relevancy we see today when doing competitive research for high value keywords. Following the previous example, not only could a website rank for a keyword that it doesn’t actually contain, due to other factors (traffic, page rank, etc…), the site could rank higher than many other sites that have a more direct relevancy. Also, such a concept dictionary could reveal common phrases and concepts that are present in advertising and spam websites which aren’t in strictly information sites. Search engines would have much more control over particular classes of sites then they do now, and I doubt it would be a good development for SEO. For more information about this topic, click here to read an in-depth article covering phrase based indexing on seobythesea.com

Technorati Tags: google,subdomain,host crowding,search algorithm

You may have already read that Google’s Matt Cutts went into more detail about changes Google made concerning subdomains. At first, many Internet marketers thought Cutts’ original comment at PubCon meant that no more than one subdomain would appear in the SERP’s for each domain. It turn’s out that the change was actually implemented several weeks before Matt made told audience members in Vegas and has gone pretty much unnoticed. Matt posted a new blog entry Monday that clarified the point of the change. Google is trying to fight ‘host crowding” - many SERPs coming from a single domain. Matt emphasized that the changes won’t effect every search. If a subdomain is relevant to a query it can still show up even if more than two subdomains are already included in the results. Here’s the relevant quote from Matt’s post:

For several years Google has used something called “host crowding,” which means that Google will show up to two results from each hostname/subdomain of a domain name. That approach works very well to show 1-2 results from a subdomain, but we did hear complaints that for some types of searches (e.g. esoteric or long-tail searches), Google could return a search page with lots of results all from one domain. In the last few weeks we changed our algorithms to make that less likely to happen in the future.

This change doesn’t apply across the board; if a particular domain is really relevant, we may still return several results from that domain. For example, with a search query like [ibm] the user probably likes/wants to see several results from ibm.com. Note that this is a pretty subtle change, and it doesn’t affect a majority of our queries. In fact, this change has been live for a couple weeks or so now and no one noticed. The only reason I talked about the subject at PubCon at all was because someone asked for my advice on subdomains vs. subdirectories.

To read the full subdomains and subdirectories post on Matt’s blog click here.

After all the noise Matt’s comments made you may still be confused about when to use subdomains and subdirectories even after the clarification. I think a big area of disconnect is between the way Google expects subdomains to be used and how they’re used by Internet marketers. From personal experience, many marketers use subdomains to host separate sites with different content such as: ringtones.domainname.com, dateingoffers.domainname.com, insurancedeals.domainname.com. Google expects subdomains to offer specific content around a similiar theme or single provider like images.google.com, video.google.com, or groups.google.com. In the first case, marketers shouldn’t be affected since the SERPs those subdomains show up in will come from very different queries. But what about subdomains like ringtones.domainname.com, superringtones.domainname.com, bestringtones.domainname.com, and hotringtones.domainname.com? This is where Google doesn’t want to see a search for “ringtone deals” to return the ringtone subdomains for “domainname.com” in 7 of the top 10 results. If you’ve been doing this, stop. It probably wasn’t working anyway.

Former Googler Vanessa Fox has posted an article that goes into great detail about subdomains and subfolders that should answer all your questions. Check out Vanessa’s post on how Goolge’s subdomain changes could affect you here.

In the past, if you had a great domain and you wanted to take advantage of the name on a new site, creating a subdomain was the obvious solution. Tag on a keyword to your main domain and Google would treat the subdomain as a separate site that could rank well right along with your root domain. Apparently those days will soon be over. At the end of PubCon in Las Vegas, Google’s Matt Cutts announced that the search engine will begin treating subdomains the same as folders. That means instead of showing up as a separate SERP result, a subdomain will appear as an indented result under a primary domain. For SEOer’s who use subdomains liberally (such as me), this news really stings. Click here to visit searchengineland.com and read more about the announcement and the possible issues Google could run into when they implement this change.

Without knowing exactly how Google views websites hosted on a shared IP, in the beginning of my own affiliate efforts I took the risk that many SEO gurus warn against and used shared hosting plans. I did so for one simple reason - I didn’t have any money starting out and dedicated or virtual hosting was not an option. During that time I always had a nagging worry that, as several experts cautioned, Google could inadvertently penalize one of my sites if another site on my shared IP was given a Google smackdown. Luckily, Mark Berghausen, a Google Webmaster Central rep, addressed the controversy in a Google Groups thread and allayed any fears that Google penalizes websites on shared IPs or favors dedicated IPs. Via seoroundtable.com, Berghausen said:

Lots of sites are hosted on shared IPs. If this had a negative effect on ranking, it would harm most of the sites on the web–and that’s not good for small webmasters or for our users. So, understandably, sharing an IP should not have an effect your ability to rank.

My advice: don’t worry about it. Host your site however you think is best, and instead spend your time focusing on the content–making sure your pages are easy to navigate for users regardless of their browser or hardware, and ensuring that the content you provide is informative and unique.

So follow Berghausen’s advice and stop worrying. Spend that time building links instead.

While I was checking my daily RSS feeds, I ran across an article on lifehack.org entitled “11 Causes of Procrastination and Their Cures”. As someone who falls into the procrastination trap every once in awhile I clicked on the link. The site isn’t actually the source of the article, but gives visitors a quick blurb about the post and then provides a link to the site where the article was written (johnsplaceonline.com). The article is worth your read, but the reason why I’m pointing it out to you doesn’t have anything to do with content of the article. Take a look at this screenshot from lifehack.org – the site that links to johnsplaceonline.com:

At a quick glance, do you see the link to the article or did you pause for a second, unsure which link will take you to the post? Even after all the articles I’ve read about AdSense ad placement and the time I’ve spent optimizing my own AdSense layout, I had to stop for a moment and make sure I was clicking on the right link. For lifehack.org, this is a standard post. Find an article with a hook, write a synopsis of the post and surround a similarly-formatted link with AdSense ads. Google has pretty strict guidelines about using AdSense code and in opinion lifehack.org is really pushing the boundaries of what’s allowed under Google’s TOS. I can’t help but feel just a little manipulated, but at the same time I admire the creativity lifehack.org employs here. I wonder how much money they’re making from webmasters who are paying for clicks from visitors who had no intention of clicking on the AdSense ad? By the way, I’m not knocking lifehack.org. I read they’re site daily and enjoy many of the articles they find about personal productivity that I wouldn’t have come across otherwise. But how do you feel about they’re use of AdSense code – brilliant marketing or an underhanded feint?

If you don’t already know, Search Marketing Expo Advanced is currently underway in Seattle. Hosted by Danny Sullivan, all the usual suspects are there like Rand Fishkin, Aaron Wall, Bruce Clay, Greg Boser, Vanessa Fox and Matt Cutts. One of the most interesting blog posts that have come out of the expo so far is on Search Engine Roundtable, which covered a question and answer session with Mr. Cutts. What Matt answers isn’t as nearly telling as what he doesn’t. One immediate thing stands out for me is acknowledgment that Google, using its access as a domain register, does take into consideration the quality and nature of all the sites an individual owns. Of course Matt wouldn’t answer exactly how the ranking of a particular site is affected by the ranking of other sites managed by the same person  -nor whether registering your sites with Google webmaster tools or analytics helps Google in this regard. Read the full SMX Q and A with Matt Cutts here.

Today’s New York Times published an article that takes an insiders look at the Google search team headed by Amit Singhal, and discusses the challenges and changes the group has gone through over the years. From an SEO standpoint, the article doesn’t cover any unknown aspect of the algorithm (as if Google would suddenly decide to open their code to the Times), but it does show a small snapshot of what its like to work at Google on the search engine itself. For general readers, it’ll be an eye-opening read about the technology and methods behind search. Hopefully, that will mean a lot fewer quizzical looks from friends and family when we try to explain how SEO improves the ranking of a website. Two things that stood out for me from the article were, first, a look at recent algorithm changes to improve ‘freshness’ – the speed at which topical content is indexed, and second, the revelation that head search engineer, Mr. Singhal, shares an office with two other Google employees – one of which is Matt Cutts! Read the New York Times article on Google and Search here. 

First, through the nature of the Google algorithm or by human manipulation, Wikipedia articles began to rank at the top of nearly every search. A few weeks ago, Google, implemented ‘Universal Search’, pushing organic results further down the page. What’s the latest difficulty for SEOer’s? It appears that ‘Google Trends’, a Google Labs project that shows the popularity of a search term over time, is now showing up in the SERP’s. Aaron Pratt of SEO Buzz Box has written about this change on his blog.  Instead of battling each other for rankings, we may find ourselves in a war with Google itself, and guess who will be on the losing side.

You have probably already noticed that Google recently implemented what it calls “Universal Search”. Now related image, video, and news results show up before website results. How is this going to affect you? Just being on the first page of the SERP’s will not work as well as it has. The first, second, and third results are now effectively the sixth, seventh and eighth result. Consequently, leveraging and optimizing image and video content will become more and more important. If image and video content are not important components of your site, now is the time to start figuring out how to integrate new media into your layout.

Google’s not going to dilute they’re bread and butter, though. Despite the hit organic search optimizers are going to take – PPC campaigners will still find their adwords ads showing up above universal and organic search results. Many SEO experts recommend combining PPC and natural search together. If you’ve avoided PPC until now, universal search is a further reason to take a serious look at adding PPC to your search tool bag. To read more about universal search and SEO you may want to check out pandia.com’s interview with SEO guru Kalena Jordan. 

Nearly everyone involved in search optimization and marketing are becoming increasingly concerned over how personalized search is going to change SEO strategy and effectiveness. I think a big reason why such worry exists is because no one has a clear idea how this change will affect them. When users searching on the same term receive different results, what methods will we employ to maximize our search visibility?

But before we can begin to attack this problem, we have to understand exactly how search engines are going to implement personalization. At this point, it’s still unclear what form personalization is going to take, not only from the perspective of SEO guru’s trying to anticipate the next search engine changes, but from within search company’s themselves. The leader in personalization, Google, has just added web history to personalized search for Google account holders. What does this mean? Nothing right now, but it gives Google a data set to begin analyzing associations between search terms and actual user intent. How effectively Google can determine user intent will become the overriding factor that leads to the extent Google implements personalization.

Though its pretty difficult to see exactly how personalization will play out over the next few years, I think we can take a few guesses about how search will evolve. The picture that is slowly coming into focus points towards the development of a recommendation engine, rather than a pure search engine. It’s no coincidence that Google was in a heated battle with eBay over the purchase of social media site StumbleUpon. Google lost out and is now trying to add StumbleUpon-like features internally. Features that will link sites together through user click patterns rather than algorithm determinations. The strategies you already employ in your social networking campaigns are the same strategies that will make their way over to your future search engine campaigns. If you are not using, or at least paying attention to, social media sites, now is the time lest you find yourself behind the curve when these changes become widely woven into search engine algorithms.

Is it time to panic? I don’t think so. At least not as far as we can predict at this point. Instead of an earth shattering change that will require new SEO methods, I believe what we’re facing is simply a merging of search and social media into a single product. The methods we use now will continue to be effective, but we’ll use them in a more cohesive fashion than we’re currently doing. The key to succeeding in the future will not be through technique but through knowledge.