Optimizing site launches

The best search engine optimization begins before the Web developer ever puts two lines of HTML and CSS code together. Now, most people just don’t know that and they should not be faulted for creating the site first and then wondering what to do about search engine traffic.

Even though it’s not rocket science, search engine optimization is a specialized area of knowledge that is not intuitively obvious. There are certain things that should be done for every SEO campaign and yet, in many cases, they are either not done or they done in the wrong order. Many SEOs find themselves playing catchup simply because they have been brought into the process at the wrong point.

Ideally, before the Web site is built your SEO campaign should:

1. Define the target market
2. Identify the applicable queries
3. Organize the content so that it maximizes search visibility, crawling, and conversion
4. Set the checkpoints and boundaries so that you can determine a return on investment

Understandably, most people believe they know what their target market is when they decide to create a Web site. But that’s not how it works in search engine optimization.

An SEO market is not a Web site market. The Web site market should always be larger than the non-paid search engine referral market. If nothing else, you always have the option of building traffic (and visibility) through a pay-per-click campaign. That is now generally called search engine marketing, which together with search engine optimization is what I sometimes call search engine placement. But even search engine placement should constitute only a portion of your overall marketing strategy.

So your SEO market is very different from your Web site market. Your SEO market comprises the search engines you intend to market through and the people who use those search engines. Optimizing for different search engines is still more effective than utilizing a one-optimization-plan-for-all approach, although a unified optimization plan is easier to implement. You have to decide how you will tackle the search engine market. Which search engines will you track?

Don’t make the mistake of targeting only Google. A lot of people do that and they just end up crying and wailing because they cannot get into Google. But there are other search engines out there and while those search engines don’t even come close to matching Google’s volume of searches they still handle millions of referrals every day.

Why cut yourself off from all that traffic if it’s relevant to your marketing plan?

Query research (more often called “keyword research”) is one of the most common failing points for ineffective SEO campaigns. People who know their industry better than their customers usually pick the wrong keywords. You know the thing you sell is called a “Whizbang 256A Converter”. Your customers all think it’s a “Gizmo ProLonger”. You know that’s a specific brand name.

Well, Xerox Corporation and Google, Inc. both got it wrong. People do use brand names as verbs and generic product descriptions. Although you need to respect trademark law, you also need to understand that you may have to chase some trademark-dominated queries. Regardless of whether you have to tread carefully in the lands of trademark protection, you absolutely need to know what words people use to find sites like yours.

And to make your query research more complicated, you may have to learn about seasonal queries, regional queries, sometimes even “shift” queries. There are, for some products and services, peak times of the day when queries surge and other times of the day when those queries taper off. If you don’t understand the patterns of searcher behavior, you may waste a lot of time looking for ways to improve unimprovable valleys in your referral data.

People search for stuff using their own words, or words that are common in their area, or words that are common only at certain times of the year, or words that are more often used at certain times of the day. It’s always best to accumulate 1-2 years’ worth of query research so you can make adjustments to your SEO campaign as it progresses.

Organizing content for SEO is not as simple as I would like it to be. Most people know you need to think about title tags and meta descriptions. Some people know that both Ask and Yahoo! still use the keywords meta tag to determine relevance. But you need to plan out your pages as if you were setting up a long row of dominoes so you can knock them over in a cascading process.

You need to design a pathway between your pages to ensure that crawlers get to the most important pages as quickly as possible, so that they can in turn fan outward from those pages to the rest of your site. You need to design your internal navigation to pass value to the right pages (and that includes link anchor text).

You need to design cross-promotional tools and strategies to help drive converting traffic that comes in from an unexpected direction.

You need to make sure your copy both informs your visitors and teaches the search engines what your pages are relevant for. You should have all this clearly in mind before you start writing copy.

And you need to know how to gauge return on investment. If it costs you $1,000 to launch your site, how do you get that $1,000 back? If you costs you $25,000 to launch your site, how do you recover that $25,000? How many sales do you need? How many convertible actions do you need to break even? How do you get past the break-even point?

The plan should begin before the work. And if it happens that you come into an SEO campaign after the site has been created, then make an effort to relaunch the site. Sometimes that works better than trying to save a sinking ship. You won’t always be allowed to do the right thing, so of course you need a backup plan.

But you should always treat every entry point into an SEO campaign as if you are at the very beginning of the complete process. It’s never too late to launch a successful Web site no matter how old it may already be.