The most important on-page factor for SEO is the title tag -- that bit of text between the <title></title> tags. There are already some outstanding articles out there on how to craft successful title tags -- specifically, Netconcepts' own Brian Brown has a two-part guide on successful title tag strategies (part 1 and part 2) which just came out recently, and I posted some quick title tag tips a while back.

You may have your title-building strategy down, but have you given serious consideration to title length restrictions? Titles look different to spiders than they do to humans. SERPs only display a maximum of 65 title characters, so that's all visitors will see, but search spiders record up to 120 characters or more. Some initial tests on lengthy Amazon.com titles reveals that some engines count keywords long past the 120 character mark, opening up an unfortunate opportunity for exploitation by spammers.

The people who find you on any SERP (search engine result page) will make their click decision based heavily on the first 65 characters of your title. Though the rest of the snippet (often times taken from your meta description tag) will also play a significant role, the title is the most influential piece of your search listing. Oddly, the title is ephemeral from the visitor's perspective; once on your page, most visitors will quickly forget about the title, which is relegated to the top border of the browser. Search spiders, on the other hand, have traditionally listed 120 as the maximum number of characters they will index.

If spammers are not yet taking advantage of these differing limitations by putting a normal, user-oriented title into the first 65 characters and tons of keyword spamglish into the remaining 55, then I'm sure they will be.

This assumes that search engines give any credence to those last 55 characters (and that they are truly limited to 120 characters maximum, which at least Google no longer seems to be), knowing that searchers will not be able to see them in SERPs, and that the title is only important to visitors before they click a link.

Or perhaps this "hidden" portion is completely discounted?

If not, I anticipate that search engines will adjust their algorithms to count only the characters viewable to searchers. At least we can hope that search engines will address the issue before it becomes common practice to keyword stuff the "hidden" segment of page titles.

With two natural listings in the top 10 on the Google SERPs for “dating” it’s hard to argue with Match.com’s SEO tactics. It works well for them - a flashy front page with a novel of text below the fold. Since this has worked so well for Match.com and has been talked about on several popular blogs it seems that others are following suit and using this same format.

I came across Patagonia.com recently and low and behold I found a near replica of Match.com’s tactic – An image and a simple selection form. Scroll down a little, however, and we find keyword-stuffed gibberish text and lots of it. This is disturbing because it feels lazy. Is this the future marriage of usability and SEO? It works, it is easy to duplicate and one doesn’t even need to write good content to get decent results. The only thing this tactic requires is a bare-bones layout built on a foundation of spam.

My instinct tells me that this tactic will fall out of favor with Google in the near future as the spiders advance and learn how to detect it. Until then, however, I expect this trend to continue to grow as more and more snake-oil SEO’s fall in line with what Match.com has made popular.

I always enjoy ranting about the dangers of black hat SEO. Gray hat SEO too, for that matter. But what I enjoy most is the challenge of disecting sites that are using black or grey hat SEO techniques (Fresh Pair, for example) and peeling away the layers of the onion. I feel like a detective... like Robert Langdon (from the Da Vinci Code) trying to piece together a complex puzzle. The part of the sleuthing that I find the most challenging and the most satisfying is uncovering sophisticated link schemes -- from aggressive link buying like what Fresh Pair and H&R Block are doing, to the most egregious spammers who obscure their ill-gotten links through cloaking and sneaky redirects. I recall back in 2004 figuring out that Findgiftcards.com was funneling link gain to their network of sites from legit sites using their free hit counter hosted at 123counters.com as the vector. That hit counter spread their spam by embedding keyword-rich links underneath the hit counter in the HTML code that the webmaster was supposed to copy-and-paste. Wow, that was like a scavenger hunt! I wrote about that one in Catalog Age. Funny, it wasn't long after that they disappeared from Google. ;-) Immediately prior to that they were #1 for "gift certificates." Oops, sorry guys! :-D

If "deconstructing" sophisticated search engine spam doesn't sound like fun to you, then you probably won't appreciate the blast I had reverse engineering As-Seen-on-TV-Store-1.com last year in my sixth installment of my SEO Report Card column for Practical Ecommerce magazine. This black hatter had no idea my published critique was coming; I'm sure they didn't appreciate me airing their dirty laundry in public! But the sins of others can serve as a great teacher.

That affiliate site was a house of cards ready to fall. Here are a couple of the reasons why (I elaborate more in my full article)...

These guys have inbound links and link text -- in spades! Yahoo! Site Explorer reveals over 6,500 inlinks to the site, excluding internal links. These links include some very reputable sites such as unesco.org/wfeo. Often times the link text is great too -- full of keywords. But the linking sites aren't relevant. Upon closer inspection, the links have been obtained by duping webmasters into posting a hit counter (e.g. from freestatscounter.com, freehitcounters.net, etc.) that contains links to doorway pages. (Hmm... sounds familiar, eh!)

"But wait, there's more!" (I'm using my infomercial voice while saying that!) -- a veritable minefield of bogus feedbacks, link farming, spam blogs (splogs), and doorway pages that have lost link popularity. 

For the rest of the findings, read the full article. Now. I promise you'll like it!

I blogged last month about Wikipedia and SEO. There are a number of considerations when making edits, creating entries, and passing the "Notability" test -- practices to avoid so you don't run afoul of their guidelines and so on.

Well folks, the game has changed. Wikipedia just instituted nofollows on all external links. This had already been in place for a while on some of their sister sites. This effectively removes a lot of the incentive to contribute to Wikipedia. Or does it? It does if your end goal is receiving PageRank to your own sites. But not if your goals are traffic (a top ranking Wikipedia page that links to you will still drive plenty of direct clickthrough traffic your way), credibility (companies with entries give the impression of being bigger and more legitimate), and reputation management (because a favorable Wikipedia entry for your company will probably occupy a spot in the top 10 in the SERPs for searches on your company name).

So are legitimate SEOs going to give up on contributing to Wikipedia? I hope not -- at least for the ones who are adding value to Wikipedia. I think we'd all like the spammers to leave (I certainly would!), and I know that is Jimbo Wales' intention, but I doubt that's what will transpire. Nofollowing blog comments didn't drive the spammers away; I can't see it working for Wikipedia. Especially as long as Wikipedia holds the top spot for important keywords such as "marketing" in Google. (sigh!)

More discussion on this development at SEOMoz.

"Gray hat SEO" is that fuzzy area of search engine optimization between ethical SEO (i.e. "white hat") and the really naughty stuff that you'll get banned for if you get caught (i.e. "black hat"). Some say that white hat SEO is idealistic, whereas gray hat SEO is pragmatic, employed by SEOs with keen business acumen. I say it's simply pushing one's luck.

The difference between white hat SEO and black hat SEO is profound and obvious. But the gray area in between the two is not so easy to define. Cloaking and pagejacking are obviously black hat. But what about stuffing the same keyword dozens of times into dozens of links on a web page? Or tucking a keyword away in the top left corner of the page in order to maximize its keyword prominence?

Consider the following examples, illustrated with screenshotted excerpts of three different homepages...

HOMEPAGE #1:

• Are there too many links?
• Is there too much repetition of the same keyword in the anchor text of these links?
• Is the link text too light? (the rest of the text on the page is markedly darker)

HOMEPAGE #2:

• Are there too many keywords stuffed into title attributes of links? (an example of which was made visible in the above screenshot by mousing over one of the links)
• Are there too many links?
• Is the color of the text too similar to the background?

HOMEPAGE #3:

• Is the keyword phrase in the top left of the page too obscured?
• Was it excessive for them to have applied this tactic to over a thousand pages?

Clearly these companies are into aggressive SEO. But have they crossed the line? What do you think??

What I find most interesting is the fact that all three of the sites rank really well for keywords they've targeted, and it appears due, at least in part, to these aggressive tactics. Top ten rankings in Google for many of the keywords targeted by the anchor text of homepages #1 and #2, and for many of the keywords targeted in the top left corner of a thousand+ pages of site #3. I can see the allure of these tactics -- after all, they seem to work!

You can't just ask a Search Engine Optimization vendor if they are ethical. Of course they will say "yes." So if you are shopping for some SEO help, how do you screen out the baddies?

A while back I blogged about how to be objective with your SEO vendor selection, but I didn't specifically cover how to screen out the unethical ones. I will do that now.

First off, interview the vendor extensively. Get them to explain the techniques they will be using. A "yes" from them to any of the following questions is a warning sign:

• Do your techniques involve any kind of deception?
• Do you use proprietary techniques?
• Do you use doorway pages or anything similar?
• Do you do deceptive redirects?
• Have you ever had sites banned?
• Do you offer rank guarantees? (You can’t guarantee something you have no control over. The only way you can get a guaranteed rank is through pay-per-click.)
• Do you send email to prospects with whom they do not have a prior existing business relationship or permission from those prospects in advance? (If so, that's spam! Never do business with a spammer.)

During your discussions with the vendor, if they describe their SEO tactics as short-term, you might want to reconsider. SEO, when done right (i.e. when following "best practices"), has long-term sustainable impact — for years, in fact. For proof, just read this.

After you're done quizzing the vendor, talk to their clients. Ask those clients:

• Does your SEO vendor teach you how to fish, or do they always do the fishing for you?
• Have your traffic and sales gone up a lot because of the vendor? If so, do you believe the increase to be sustainable?
• How long have you worked with the vendor? How long do you plan to continue working with them? Any idea what the vendor's client churn rate is?

Then you'll need to do some of your own investigating. Check the HTML code on their clients' sites for hidden text, hidden links, and so forth. Also examine what their clients' websites are serving to the search engines. There are a couple different ways to view a website through the eyes of a search engine spider: one is through a Firefox browser extension called User Agent Switcher; the other is through the cached version of the page that was indexed by the engine, available from the Cached link in the search results. Compare and contrast the page meant for the search engines to that corresponding page off the native website as seen by a normal visitor. If the content served up to the search engines is something completely different than what is served up to visitors, then they are spamming. Things to look for when making your comparison: if the title tag is significantly different, and if keywords have been stuffed into the body copy, the meta tags, and into parts of the website to help the version that was shown to search engines rank better. Finally, search the online forums and SEO directories like SEOPros.com and SEOConsultants.com with Google for complaints about the vendor.

Got any horror stories or lessons learned to share from dealing with a less-than-stellar SEO vendor? Post a comment.