Beyond SEO

A video post about using free Google tools to understand your website and customers. Focuses on Google Webmaster Tools, Analytics and Website Optimizer.

This is part of my Beyond SEO series where I look at the myths and realities behind search engine optimization, with practical tips about publicizing your site and building your personal brand. Read all of my Beyond SEO articles.

The Google blog asks for user input into what makes a good SEO and reports that they've just rewritten their page that warns against rouge SEO artists and gives recommendations about what to look out for. It starts with their definition

SEO is an acronym for "search engine optimization" or "search engine optimizer." Deciding to hire an SEO is a big decision. Make sure to research the potential advantages as well as the damage that an irresponsible SEO can do to your site. Many SEOs and other agencies and consultants provide useful services for website owners.

The blog asks "how would you define SEO? What questions would you ask a prospective SEO?" I've been doing a lot more optimization for clients lately. What's particularly fun is running across the work of the SEO scam artists their competition have brought in. I've seen many instances where the other SEO firm has stepped over the bounds of fair practice and been penalized by Google.

Google's job and our job

I've always taken the approach that it's Google's job to give people the most useful and relevant return for their search and our job to make sure we have useful and relevant material and arrange it in such a way that Google can access it.

SEO is important but only in the context of smart web design and a coherent and well thought out internet marketing strategy. Firms that claim to do SEO without checking the analytics data and consulting with the client about their business strategy will not help the site in the long run.

What your SEO expert should be doing

I would agree with most of Google's recommendations of what to look out against. But what to look for? A quick list would include:

• A SEO consultant that looks at analytics data before making any changes. If the client doesn't already have Google Analytics running on the site I install it and wait a month before doing anything. I do that because you want:
• Quantifiable results. You should be able to see shifting use patterns if the optimization is working. The internet gives us precise figures and it's often very easy to demonstrate the value of the work you've done. Clients should have full access to the analytics and be trained enough to be able to independently verify the results.
• A consultant that frequently answers questions with "Hmmm..., I don't know." No one knows what Google is doing. You try something, then you try something else. Anyone who claims to know everything is scamming you.
• Someone who looks at your entire business model and asks hard questions about your internet strategy. What do you hope to accomplish with your site. Are there specific goals that we can measure?
• Think about your your Inbound and Outbound strategies. Google will send people your way if you have useful material so think about what compelling content you can offer the universe. And once people come to the site you have to make it compelling for them to stay a while, subscribe, etc.
• The SEO consultant should make you sweat: anyone who says they can significantly boost your site without you having to lift a finger is fooling you. You will almost always have to add compelling content and it will take you committing staff time to the project (a good development team will look for ways to make this fit into your existing staff routines so that it's as painless as possible!).

Any others suggestions for what to look for in potential SEO consultants?

I was recently working with a client who has a large Google Adwords campaign, with an annual ad budget in the low six figures. He's been very careful about the keywords he's chosen and we've both poured over the Google Analytics figures to see how the campaign progressed.

It took a third party keyword tracking system to discover that many of the ads were being served up to wrong keywords in the Google searches. I want to keep the client's identity private, so let me use an analogy: say you're a boomerang maker and you've bought a campaign intending ads to show up for those who search "boomerang" in Google. What we discovered is that Google was serving up a large percentage of these ads for searchers of "frisbees" -- close, but not close enough for searchers to care. Few people clicked on the misplaced ad. We're talking serious money wasted on ads served up to the wrong target audience.

How did a carefully constructed ad campaign get on so many poorly-targeted searches? Google allows fuzzy matching under their broad match guidelines:

For example, if you're currently running ads on the broad-matched keyword web hosting, your ads may show for the search queries web hosting company or webhost. The keyword variations that are allowed to trigger your ads will change over time, as the AdWords system continually monitors your keyword quality and performance factors. Your ads will only continue showing on the highest-performing and most relevant keyword variations.

You can disable these broad searches using negative keywords (i.e., "-frisbee") and with specific keywords ("boomerang").

But Google does not make it easy to see just where your ads are going. You have to set up a special Search query performance report. It's really essential that anyone doing a large Google Ad campaign set up one of these searches and have it automatically emailed to them every month. Google clearly wasn't tracking the "performance" of its broad search on this client's ad. I'm particularly disturbed that we didn't see these misdirected keywords listed in the Google Analytics tracking reports. It is dangerous to use the same company to both sell you a service and to report how well it's been doing.

Credit where it's due: it was the excellent long-tail blog content service Hittail that gave us the information that Google was misdirecting its ads. See my previous Hittail coverage.

Whenever I talk with fellow web designers, the issue of "SEO" invariably comes up. That's techie slang for "search engine optimization," of course, that black science of making sure Google lists your site higher than your competitors. Over the years a small army of shady characters have tried to game the search engine results.

Just Give Google the Content It Loves

In theory it's easy to make Google, Yahoo, MSN and the other big search engines happy: give potential visitors site they'll want to visit. Forget the tricks and spend your time putting together an amazing site. Search engines like text, so write, write, write.

I'm looking to join a web design house, which means I've been interviewing with slick web developers lately and whenever they ask me the best way to increase SEO for their clients, I tell them to start a blog. They look at me like I'm an idiot but it's absolutely true: two blog posts a week will end up being over 100 pages of pure content. All of these sites full of Flash animation get you nowhere with Google. (An aside: I've been having fun going after the keywords of some of the design firms I'm interviewing with. I've actually been pretty successful. I am the king of Minotola Web Design, muhahahaha!)

Just a note that any kind of text-rich web system can achieve many of the same results--blogs are just the easiest way yet to get content on your site.

Presenting What You Already Have: Blog your Water Cooler Chat

When I talk to people about starting a corporate blog they quickly start telling me how much work it will be. Bah and Humbug--your company's life is probably already filled with bloggable material!

I used to work in a bookstore where I did most of the customer service, much of it by email. About two or three times a week I'd get a particularly intriguing query and would spend a little time researching an answer (mostly by looking through the indexes of our books and searching the arcane sites of our niche). This research didn't always pan out to a book sale, but it marked our bookstore as a place to get answers and gave us a competitive advantage over Amazon and its ilk. Each of my email answers could have easily been reformatted to become a blog post. By the end of a year, I'm sure the volume coming from these obscure searches would be quite high (see yesterday's Long Tail Strategy post on the HitTail blog for an account of how attention to search engine's one-hit-wonders helped achieve a widespread keyword dominance).

Whenever something new happens that breaks you out of your routine, think about whether it's bloggable. At the bookstore, a new book would come in and we'd spend ten minutes talking about it. That conversation reached half-a-dozen people at most. In that same ten minutes we could have written up a blog post saying much the same thing.

Last Spring a controversial article appeared in the local newspaper that tangentially involved my employer. That morning my workmates gathered together in the reception area for the better part of an hour trading opinions and wisecracks. After about five minutes of this, I slipped back to my office and wrote my opinions and wisecracks down into my blog. I hit post and came back to the reception area--to find my workmates still blathering on, natch. My post reached hundreds and took no more time out of the work day than the reception pontifications.

Humans are social animals. We're always blogging. It's just that most of the time we're doing it verbally around the water cooler with three other people. Learn to type it in and you've got yourself a high-volume blog that will add invaluable content and SEO magic to your site.

Mix up your content: Tag Your Site

Lastly, a point to webmasters: it usually pays to think about ways to re-package your content. My most recently experience of this was tagifying my personal blog over at "QuakerRanter.org." Every time I post there a Movable Type plugin fishes out the key words in the article and lists them afterwards as tags. These tags are all linked in such a way that results send the term through the site's search engine to give back an on-the-fly index page of all the posts where I've used that term.

Tags are like categories except they pick up everything we talk about (when we use them aggressively at least, and especially when we automate them). We don't necessarily know the categories that our potential audience might be searching for and tagifying our sites increases our keyword outreach exponentially. My personal blog has 239 entries but 3,860 pages according to Google. It's the parsed out and re-packaged content that accounts for all of this extra volume. This doesn't increase traffic by that nearly that much, but last month about 30% of my Google visits came from these tag indexes. More on the mechanics of this on my post about the tagging.

Every web designer under the sun talks about search engine optimization (SEO), but it amazes me to see how often basic principles are ignored. I'm in-between jobs right now, which means I'm spending a lot of time looking at potential employers' websites. I've decided to start a series of posts on SEO myths and realities that will talk about designing for maximum visibility.

One of the neatest observations to gain popularity in the last few years is that of The Long Tail, first coined a few years ago by Wired magazine editor Chris Anderson (here's the original article). He noticed that the internet had opened up access to niches--that searches and national distribution networks had given new markets to obscure and small-market products. The classic example is Netflix, the direct-mail movie rental service, that has a huge catalog of titles, the great majority of which are so obscure that no local video rental store could afford to carry them. But Netflix actually rents them all and if you add all these low-volume rentals together you'll find the total volume exceeds that season's blockbusters. I learned just how strong the long tail can be a few years ago when I worked on Quakerfinder.org, a meeting/church look-up service. For the first year, the site got moderate traffic from search engines. Google wasn't able to index the actual church listings because users were required to type towns and postal codes in to get results. The only search engine visitors we got came in on very generic phrases like "find quaker meetings."

Suspecting we were losing a large potential audience, I redesigned the site so Google could index each and every meeting (adding a few tricks so each listing traded links with half-a-dozen other listings). Once the change was in effect (help from our programmer), those old generic search phrases were still the most popular. But now we got small numbers of visits on thousands of terms which we hadn't hit before: "Quakers Poughkeepsie" and "Quaker Churches in San Francisco," etc. This was the long tail in effect. Our visits jumped fourfold within a few months (see chart). The long tail made us much more visible. (More on the Googlization effort in that year's analytic report.)

A great new traffic analysis service is called HitTail. Like many other programs it tells you what search phrases have brought traffic to your site. But what's cool is that it gives suggestions--keywords it thinks will bring even more visitors in. Some of the suggestions are funny. For example, it thinks I should post about "traditional sweat lodge songs," "ticklish armpits" and "how to dress with personality" over on Quaker Ranter. But it also thinks I might consider posting on "small church local outreach ideas," "new online magazines" and "christian quakers."

If all one was worried about was sheer traffic volume, then a post on each keyword might be in order. But this would bring a lot of random traffic and dilute any focus the blog might have (I already get a lot of traffic on a particular non-typical post that I wrote partly as an SEO experiment). My guess is you should go through the HitTail suggestions list to find topics that match your site's focus but do so in language that you might not normally use.

I might try some experimental posts on my personal blog soon. I'll definitely report back about them here on the MartinKelley.com design blog. In the meantime, check out HitTail's blog, which has some good links.