• Register
• Log in

SEO Theory - SEO Theory and Analysis Blog

Algorithm analysis, Web community relationship analysis, SEO practices and techniques, industry news, etc.

• Home
• About SEO Theory
• SEO Glossary

How search engine optimization works

Posted by admin on December 21, 2006 in SEO Theory

The Internet marketing firm I now work for has been interviewing people for various positions. When I sit down with someone to talk about an opening on my team, one of the first questions I ask is, “have you ever heard of search engine optimization?” Another variation is, “Do you know what search engine optimization is?”

Not every position we hire for is about SEO, but I ask anyway. Although there may be no right answer to the question, it’s interesting to hear people try to describe it. One recent interviewee immediately said, “That’s where you get links from other sites to your site so that your site ranks well in search results.”

I wish it were really that simple. After all, getting links is easy.

But search engine optimization is not about getting links. It’s not about any one thing. There are four factors that determine any page’s position in search results:

1. What you do with your page
2. What other people do with their pages
3. What the search engines do
4. What people search for

When you pick a specific expression to target (hopefully it’s a good one), you only nail down one factor. The other three factors are still very much unknowns.

What you do with your page

The majority of people seem to understand that you have to do something with title elements. Each page should have a unique title element that accurately and briefly describes the content on the page. Many people still believe that meta tags matter. Well, they can help in some ways but setting meta tags is not optimizing on-page content.

There are many things you can do with your pages, including:

1. Put indexable content on them
2. Put unindexable content on them
3. Point links to them
4. Link out to other pages
5. Optimize the indexable content
6. Optimize the outbound links
7. Optimize the inbound links
8. Misconfigure your server
9. Configure your server properly
10. List your pages in directories
11. Hide your pages

For future reference, I will not discuss link optimization on this blog. That would be giving away too much to competitors, but I’ll confirm that there are definitely methods for optimizing linkage that most SEO link gurus have never considered, and none of them have discussed such techniques in either blogs or forums (so far as I am aware).

Optimizing on-page content has nothing to do with “keyword density”, “keyword placement”, hiding text, or writing compelling copy (but compelling copy is vital for the success of a page that needs to make conversions).

On-page optimization is all about making it clear to search engines that your page is relevant for the topic you are targeting. If you place indexable text on the page, it will be relevant to a large number of topics you don’t target. But its relevance for the majority of those topics will be relatively weak. The purpose of on-page optimization is to emphasize relevance for the most important topic.

Many people take a formulaic approach to on-page optimization. They tick off features like they are buying groceries from a list. Such pages rarely meet my expectations of on-page optimization. You emphasize importance in one of two ways: repetition or use of HTML elements. Too much of either and your page looks ugly. Worse, it may look spammy. So you have to develop a non-formulaic approach to writing copy that favors the topic you want to target.

But people often overlook copy in all the obvious places. The most frequently used navigation link anchors are “home”, “next”, and “prev”. I’ve often used them myself. Nonetheless, in a day and age where everyone sings the praises of link anchor text, why would you want to put out a 500-page Web site that externally boosts the relevance of the root URL for “home”? Not only is that not a word most of you would want to optimize for, it’s an extremely crowded word space.

Think of how you would write copy if you could only deal it out to each visitor one line at a time. Each line has to be written on an index card and it has to compel the visitor to take one of two actions: click on a link or go to the next line.

That’s all I have room to say about on-page optimization tonight.

What other people do with their pages

Assume that regardless of what you do, there are people out there targeting your precious word space who follow my advice and do everything I mention above. They are optimizing their content. The search engines have to reward that work in some way by recognizing the relevance. Now, each search engine measures relevance in a different way, and they throw in other factors. But the point is that you can sink or float solely on the basis of what other people do regardless of what you do. They may have more advantages than you.

Don’t assume you lost your rankings because you were penalized, or because Google implemented a new filter. Investigate. See what you can learn about other people’s pages. But you should also understand that truly savvy search engine optimizers can practice sleight-of-hand tricks intended to fool only you, the competition. Don’t get caught up in copying their keywords, or their anchor text, or their on-page copy, etc.

In fact, you should assume that your initial analysis of your competitors’ higher ranking pages is completely wrong. If you’re convinced they got to where they are based on links, history shows us you’re most likely wrong. Oh, they may have lots of links, but how many actually help? Those are the only links anyone should care about.

If you’re convinced they have the greatest copy in the world, you’re going to find out the hard way that emulating their copy won’t help you. You need your own copy.

What I am saying is that, if you are not coming in first, you won’t figure out how to beat the competition by trying to do whatever you think they are doing. They may have seeded their pages and link footprint with false leads. So, be aware that other people may outperform you, but don’t kill yourself trying to do things their way. Their way may not be a good way, and it may not be so obvious.

What the search engines do

Search engines do a lot of things. They refresh their databases (meaning they drop sites and recrawl the Web), they change their ranking criteria, they change their crawling priorities, they change their filtering, and they occasionally penalize and/or ban pages.

A penalty simply prevents a page from showing up for any words other than the page URL. The page remains in the database. It’s indexed. But it’s penalized.

A ban removes the page from the database completely. You won’t see it show up even for the URL.

Filters look at many different things. To avoid being filtered, you want your content, your outbound links, and your inbound links to be as clean and as reputable as possible. Do other people slip past the filters? Sure they do. Search engines are not perfect. So what? If you don’t respect the fact that search engines are moving targets, you’re going to come in last every time you enter the turkey shoot.

A good methodology, and one which many experienced SEOs advocate that beginners employ, is to go for the low hanging fruit. Try optimizing for easy-to-capture expressions first. Yes, you need the traffic (and the money) from the big keywords like everyone else, but if you don’t know what you’re doing, you’re just going to get the crap beat out of you and you won’t learn anything from the lesson.

Take on the fights you can win and learn from the easy campaigns first. Build your skills and experience. You won’t become very good at reverse engineering the algorithms. After all, most very experienced SEOs are very bad at reverse engineering algorithms. Algorithm chasing is an exercise in learning how to follow, because you can never stay a step of ahead when you are constantly having to figure out what the search engines just did. But search engine optimization is an exercise in learning how to lead. You can’t look back or someone will pass you.

Algorithm chasing can be fun but most people fail at it because they buy into the nonsense that is passed around SEO tutorials, FAQs, blogs, and forums. Some people get at least some of the facts right, but they aren’t who you think they are. The best algorithm chasers rarely reveal the goods. They have no incentive to tell you or anyone else what really works and why.

But if you want to chase algorithms, then start by going after the low hanging fruit. And start with the assumption that links are not nearly as powerful as everyone says they are. You’ll be closer to the truth that way, your analyses won’t be as flawed as the big name analyses tend to be, and it will be easier for you to make adjustments as you collect data.

But you also have to understand that algorithms perform differently for poorly populated queries from the way they perform in highly populated queries. See if you can figure out why that is so. It’s not a very deep, fundamental principle, but neither is it immediately obvious.

What people search for

You might believe that you have no control over what people search for, but that’s not always true. If you can afford to buy advertising of any kind, you can use that advertising to tell people to search for specific things. When you cannot grab control over the “pizza” name space because it’s too competitive, teach people to search for “Michael’s Italian Sausage Pies” instead.

When I launch a new site that is important enough to require promotion, I create a graphical advertising campaign that teaches people about the site. They know what to look for. They see my advertising over and over again. I brand my keywords into their brains. And I use non-graphical promotional copy in various places as well.

You cannot force people to search for your keywords, but you can give them compelling reasons to do so: offer rewards, tease them, use a little mystery, stick an annoying monkey in their faces, be monotonously repetitive. And design your Web content to cross-promote for those keywords. Seize control over any word space for which you build a market.

Understand, also, that Google has reported that as many as 20-25% of all monthly queries have never been used on their service before. Every month, 1 out of every 4 or 5 searches on Google is for an expression that Google has never seen before. You have plenty of opportunity to build brand recognition, keyword domination, and word space control.

Recap

Search engine optimization works when you take all the factors into consideration. You experiment, evaluate, and adjust. Over and over again.

You’ll never get it just right for very long. Each page, each targeted expression will be different. You’ll make it different just by creating content for that expression. As soon as you change the mix, everyone has to go back and re-evaluate: you have to re-evaluate to improve your results, your competitors have to re-evaluate to maintain their results, the search engines have to re-evaluate to maintain or improve their quality, and searchers have to re-evaluate to ensure they find what they are looking for.

They may not be looking for what you have to offer. Don’t be disappointed by rejection from searchers. Learn from the rejection. Find something useful in the experience and use it.

Experiment, evaluate, and adjust. That is how search engine optimization works. It doesn’t work any other way.