Everything In Between

Just a note before bed:

• Google search ranking for a search on my last name: 9^th
• Google search ranking for a search on my first name: 7^th
• Google search ranking for a search on my nickname: 1^st

Hmm….

Spammers have been trying to fool the search engines for years. Trouble is, no one except the search engines know how web sites are ranked in their result lists. Consequently, dozens upon dozens of different tactics have been employed to try and get web sites to rank well.

This is changing.

Google’s PageRank™ Algorithm Public

Today Google’s site ranking algorithm was made public via a public patent filing. This will change lots of things, but not in the ways some people might expect. As it turns out, Google’s algorithms are set up in such a way that makes it incredibly difficult to spam the index with the kinds of techniques spammers are currently notorious for.

Nick Finck of Digital Web Magazine cites:

Now my first thought was “oh great, now every manipulative marketer on the block is going to abuse this info,” but if you look closely you’ll see that if they, in fact, followed these site rank guidelines to their best advantage, their site will become less of a link farm, less of a re-blog, less of a link exchange, less of a faux landing page. And who knows, maybe some more useful content will be generated out of this. [via Jordan Rule]

I’m sincerely hoping that will happen. Just today, I’ve received over three hundred spam comments on my blog from various spammers attempting to raise their rankings with backlinks and targetted anchor text. None of them made it through to the site past my moderation system, but the point is they tried and I had to clean up the mess.

Ethical Internet Marketing Increases Google PageRank

The spammers and the search engines have always been caught in a perpetual arms race. With Google’s ranking algorithm public knowledge, in what direction will the balance of power tip? Does this give the spammers a bigger, juicier target?

I’m inclined to believe that, in fact, this is bad for spammers because it’s good for everyone else. I think this is good for SEO professionals and marketing folk because now we know precisely what works and what doesn’t (and what works discourages spamming) and good for web searchers because more content will be findable. In other words honest SEO works better than spamming, and now we can prove it.

This doesn’t mean spammers will stop what they’re doing, or won’t find new ways of abusing the algorithm, of course. However, it does give every honest internet marketing agency the validation they deserve, and now it also gives them a broader set of tools with which to work. I, for one, can already think of several enhancements I can make to my site to get better rankings—and none of them involve spam.

What You Should Do to Rank Better in Google

Darren Yates proclaims that the days of Spamming Google are drawing to a close. Some of the facets of Google’s algorithm were somewhat obvious, but some of them were very surprising. For instance:

• The longer the duration you have registered your domain name for, the better pages in your domain will rank.
• If links to your site propogate around the Web too quickly your pages will rank worse. (Be very careful who you exchange links with.)
• Fewer higher-quality in-bound links with varying anchor text will make your site rank better than many in-bound links with identical anchor text. (That’s considered “anchor spam” and can take months to recover from.)
• Delivering regular, fresh content is not ncessarily a requirement for every site to rank well. Whether or not your site should update regularly depends on your market. The idea is to keep information accurate and reliable, regardless of how old it may be chronologically.
• Site click through rates are monitored from a variety of sources including the Google Toolbar and Desktop Search products, as well as the Google web site. The amount of time it takes for a visitor to return to Google after a click-through is also an indicator of the quality of content; the stickier you make your site, the better.
• The whois records for your domain name are checked for consistency between the technical and administrative contacts since these records are often faked for spam domains. Make sure the proper information appears in your domain name registration.
• If your web site is hosted on a shared server, then you share the same IP address as (potentially) thousands of other web site owners. If even one of them gets caught for spamming, you get penalized as well since you share the same Internet address.

Much of this information is a goldmine for web developers and internet marketers. Taken in conjunction with proper coding techniques (such as semantic HTML, proper separation of presentation and content with CSS, external and light-weight JavaScripts, etc.) it can deliver a powerful one-two punch for new and existing web sites that want to increase their Google PageRank.

The bottom line, however, is to ensure that a site grows as organically as possible. Make sure you register new domain names for a minimum of at least two years (I’d recommend three or more), be very careful who you exchange links with, vary your anchor text, and depending on your market, make sure to add chunks of high-quality content to your site every so often.

Finally, make your site is as “sticky” as possible. Encourage visitors to bookmark your pages and to return regularly. It is far more beneficial for a site to get a small amount of repeat traffic than an influx of traffic that never returns.

What’s a CMS?

Much of the Web these days is organized within Content Management Systems (CMS’s) which help site owners and authors easily produce new content while managing the old. A good example of a CMS is blogging software such as Moveable Type or, my personal favorite, WordPress. Other examples include PHP-Nuke, PostNuke, Drupal, and Mambo. CMS’s are so widely used that you’ve probably interacted with one whether or not you run a web site of your own.

SEO and CMS

So with all this content being generated, site owners and CMS developers and users need to know: What can I do to better my search engine ranking through my CMS? Here are a few tips:

1. Reduce code-bloat.

   Both end-users and developers can and should look over their page templates and cut out any unnecessary or extraneous (X)HTML that’s only adding to page weight. PostNuke and PHP-Nuke, for instance, are notorious for generating old-school nested tables through their various modules. Do everything you can to eliminate this tag soup.

   Also, take a look at the head section of your template. If you see a lot of JavaScript or CSS in there, you should cut it out and attach it to your page via an external document link or <script src="..." /> tag.

   Ensuring that your content is near the top of your document will help keep the spiders from getting annoyed. Plus, the reduction in bandwidth is often surprisingly impressive.

2. Dovetailing off the previous tip, use semantic markup.

   It shouldn’t come as any surprise that a p tag tells the browser that the its content is a paragraph. Despite the obvious importance of this fact, all too often this paragraph tag is replaced by a semantically meaningless div or td element. When considering headers, this is especially important because search engines weigh the text between h# tags heavier than they do other elements of your page. So make sure your templates are using the proper element for your content!

3. Use your meta tags to your advantage.

   In lots of cases there doesn’t seem to make a difference whether or not meta tags are accurate, or even exist. This is the result of search technologies getting better and of search engines distrusting authors (thanks mostly to spammers and porn mongers) from providing accurate information in their meta tags.

   Still, if you can use them, do so. Most of the well-ranked pages are well-ranked not because of their tags but because of their content and their referers, that is, because people linking to them. However, even they were once not-so-well-ranked, and meta tags do seem to make a difference when gauging smaller sites’ pages.

   As for how specific one’s keywords need to get, I think that’s a judgement call based on how specific the majority of your content for said page is. If you have a web site about cars with some general pages about how cars work and why you love them, then using “cars, Ferrari, Porsche” is fine as long as there’s at least a passing reference to Ferraris and Porsches. Then on your child, more-specific page, do the same, but omit “Ferrari” on the Porsche page and vice-verca.

4. Create friendly URLs to encourage spiders to index dynamically generated pages.

   This one is aimed at developers of CMS’s, though sometimes a saavy end-user can accomplish this as well. I’m not sure how prevalent it is anymore, but I know that at least in the past some spiders wouldn’t index page content if it were generated by a query-string. That is, if your page’s addres ended with something like ?n=v&n1=v1 then the resulting page may not be indexed by search engines at all. Furthmore, that’s not the easiest page name to remember. All that extra stuff from the question mark on is called URL cruft.

   The Apache Webserver has an insanely powerful tool called mod_rewrite (documentation) which many sites (including this one) use to clean up that URL cruft and transform it into a clean address string for search engines and visitors alike.

   This tool is also exceptionally helpful because it means you can embed a page’s keywords directly into its URL, providing an exceptionally helpful hint for search engines.

There’s much more you can do, and certainly more developers can do to help CMS’s better integrate with search engines. Hot Banana is one such recent CMS endeavor that has generated some buzz lately.