• 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

It’s not a penalty if …

Posted by Michael Martinez on January 3, 2008 in Seo Myths

Google unleashed significant churn in its search results this week. There are four possible reasons for why I might be seeing that churn (those four reasons should be permanently etched into your heart). SERP churn happens every week, every day in some cases. It’s normal, ordinary, everyday stuff in the world of search.

But once in a while something significant happens and when I see broad, sweeping changes across dozens of unrelated search expressions, I sit up and take notice. That’s usually when I hit the SEO blogs and forums to see who is whining and crying because whiners and criers inevitably tip their hand and reveal that they spammed or did something skanky.

If the only people complaining about being hurt are spammers, then everyone else who has been hurt will usually recover their lost rankings.

But what if people who have been keeping their noses clean for years start piping up and claiming they have lost rankings despite doing nothing? In my experience, 70-80% of all people who claim to have done nothing have usually done something to their Web sites: they’ve deleted pages, changed links, changed templates, etc. To most people, that is “doing nothing” because to most people all there is to SEO is creating links.

Relevance has about as much to do with your backlink profile as George W. Bush has to do with the internal decision-making processes of the Democratic National Party.

A natural backlink profile is diverse, mostly ineffective (I usually estimate that no more than 70-80% of your natural backlinks are passing value), and has relatively little impact on search results rankings. But people who optimize for search tend to rely almost completely on links. Hence, when they lose well-established rankings, they assume it’s not because of their links.

Think about it this way: you limp along for 8 years on weak, poorly optimized content and shore up your algorithmically irrelevant content with link spam (but to you that link spam is “thousands of quality backlinks”), and then one day you lose your link spam-based rankings.

Have you been penalized?

The issue of tripping a penalty is important in search engine optimization but most people who think they have been penalized usually have NOT been penalized. A lot of SEO forum admins through the years have had to watch newcomers show up and insist that some new penalty has been implemented because they lost rankings. It’s rare when such claims prove to be true. Maybe once or twice a year the SEO community effectively identifies a new search penalty. Some people might argue it happens perhaps four to five times a year. I don’t think anyone with any significant experience would claim you see more than four or five new search penalties in a year.

Still, a lot of people trip penalties. But most people trip filters when they do something wrong, and filters are not penalties. Search engines may introduce new filters on a pretty regular basis. They are constantly analyzing spam reports and aggregate data from their crawls. A slight change in filtering technology can have significant, far-reaching impact — although that depends on what the filter actually does.

If a filter simply prevents a page from appearing in crowded but relevant search results, but allows the page to remain in the index without penalty, the page can still pass value to other pages through its links. And the page can still appear in less crowded but relevant search results (or maybe even in less relevant search results).

Remember, a Web document is relevant to every word it contains (caveat: unless the document is stored in Google’s Supplemental Results Index, which most documents still appear to be). If your page loses position for expression 1 but performs well for expressions 2, 3, and 4, you’re probably not being penalized. You may be filtered or you may simply have lost ground due to a change: a change you made, a change your competitors made, or a change the search engine made.

It could be changes were made all around.

Changing things doesn’t mean you’ve tanked your rankings. It may only mean that the search engine just needs time to recalibrate its search results to reflect your changes, other people’s changes, or its own changes. But it’s important to understand that any alteration you made to a page or site — no matter how insignificant you deem it to be — could have disastrous effects on your rankings. It could be you mistyped something when making the change. More than one Webmaster has propagated bad code across thousands of pages.

So, anyway, this morning I checked out a favored forum for what I call “weather reports”. There was some minor screaming and yelling but nothing really significant. Instead, I found a discussion about some new proposed “penalty” where the person attempting to define the problem and identify potential causes offered only three possibilities: backlinks, backlinks, and backlinks.

With this kind of narrow-minded thinking, it’s no wonder the SEO community struggles to adapt to change.

Usually proposed new penalties are believed to be “search term specific”. It takes an inordinate amount of resources to hone in on one or a few search expressions in order to penalize a handful of miscreants. A search engine needs to justify that kind of effort. Now, it’s been done before. Google addressed link bombing by targeting about 100 expressions after being extensively rebuked and criticized for a couple of years. They also reportedly confirmed last year (2007) that certain highly competitive commercial expressions have been subjected to very specialized scrutiny.

So keyword-specific penalties and filters have been created and the notion should not be dismissed out-of-hand. Nonetheless, through the years I’ve looked at thousands of Web sites that supposedly suffered from “keyword-specific penalties” and found them wanting in so many ways there was just no reason to believe they had been penalized for competing in those search expressions.

There is currently a very popular SEO pundit who has been whining on a blog about lost Google rankings. Other SEO pundits are dutifully linking to the whiner post, suggesting that Google has inexplicably penalized the whiner’s blog for various expressions. However, when I looked at the posts in question I found no indication of a penalty. What I did find (which is fairly typical of high profile SEO bloggers who whine about lost rankings) was a serious lack of optimization for the expressons.

The lack of critical thinking in the visible SEO community is a serious detriment to the professional development of this community. If people devoted as much effort to questioning the crazy assumptions their favorite SEOs make as they do to supporting those crazy assumptions without asking questions, search engine optimization information on the Web would be considerably more reliable than it currently is today.

There is a pretty solid test that anyone can perform to determine if a search engine has implemented a penalty. Once you think you know what the penalty is concerned with (links, on-page spam, whatever) you need only create four new Web sites.

Two of those sites should violate your proposed penalty.

The other two sites should be clean, lean, and basically text-only.

You can build as many links as you wish or not. Whatever you do about links, you should do for all four sites.

If you feel the penalty is keyword-specific then you need to create sites for that expression. If it’s a heavily competitive expression you can still test the penalty.

How much time you need to conduct the test depends on what resources you have available. If, for example, you know how to get pages into a search engine’s index in a matter of hours, you should not need to wait more than a week or two to find out if you can trip a penalty. Okay, most people don’t know how to do that but it’s not as hard as you may think. You just need a backdoor.

In any event, if the pages you feel should trip the penalty lose their rankings or never rank, whereas the pages you feel should NOT trip the penalty perform well, then you have pretty good reason to believe there is a penalty.

You can try to look at what other people who have lost rankings may be doing but the problem with that kind of analysis is that most people in the SEO industry practice very poor SEO. After all, if you get enough “quality backlinks” you can spam your pages to the top of the search results without doing any optimization at all.

Over time, if enough people report the same symptoms across multiple blogs and forums, you may reasonably conclude there is probably a penalty, a filter, or a really bad SEO tactic at work. If you can afford to wait that long, you can find out whether you’ve been penalized.

In my opinion, it’s better to acknowledge that you probably changed something. The fastest way to determine if a change just alters the relevance scoring is to undo the change. If your good ranking returns, then implement the change again (exactly as before). If you lose your rankings, undo the change the a second time. If your rankings are restored again, then you know the change you made hurt your rankings.

Most people never bother to go that far. If they can just get their rankings back after undoing the most recent changes, they tend to be happy. But the sad point in that lazy approach to optimization is that it tells you absolutely nothing about why you lost your rankings or why you regained them.

If you cannot replicate the problem, you cannot identify it and therefore you cannot find a solution for it.

In which case, you might as well go back to building thousands more of your “quality backlinks”. At least you’ll feel like you’re doing something.