One of the biggest issues SEOs face is dealing with various types of coding problems. Today, we'll explore three common practices that can be used quite legitimately, yet still could look like spam to search engines.
These can result from a content management system (CMS) that uses SEO-hostile coding practices (not on purpose of course, it's just ignorance) or an in-house engineer who's simply not familiar with SEO. Sometimes these problems are extremely hard to deal with. Here's an overview of some of the coding techniques that might be used completely innocently, but are potentially risky:
This popular technique for embedding collapsing menus on a Web page is helpful when you have large, complex menu systems. Let's take a look at an example:
As used on Amazon, the display:none attribute allows you to store within the HTML of the page the submenu items for all the major categories without having them shown all the time. This provides a better user experience, where the selected category shows its sub-categories, but the other categories don't. The user can then mouse over another category if they want, and they'll see its subcategories.
However, this technique can also be used to embed spammy text content in the page.
SWFObject is used together with Flash to provide search engines with HTML text content that they can read. While search engines can actually look inside Flash these days, their capacity to understand what is used inside there is limited. The search engine gets something easier to chew on with SWFObject.
However, there's nothing that ties the content of the two together, so the search engines have no assurance that the text content it sees is representative of what's in the Flash file. Thus, the potential to abuse this technique exists as well.
In contrast, there's another technique used with Flash known as sIFR (Scalable Inman Flash Replacement). sIFR is designed for improving font quality on a Web page. It's meant to be used in a limited way, such as improving the look of your header elements on your Web page. The improved font gets implemented in Flash. The search engine visible text is used as the input to sIFR, which then renders that exact text in a much nicer font. There's little room for misinterpretation of sIFR by the search engines.
And now, with this week's news that Adobe is providing Flash technology to Google and Yahoo for better indexing, Web developers and SEOs may soon not need to develop a workaround for Flash files.
This one's a bit more controversial. However, programmers will use it to solve specific types of problems. For example, a Web site application can detect if a person's IP address is from France, and if so, send them the French version of a Web page.
However, search engines have taken a strong stance against any form of IP delivery, and refer to it as cloaking. Cloaking has been used for a long time as a spamming technique, and is the riskiest of these three techniques.
Consequences and Risk Management
These three examples are all used by programmers for a variety of legitimate reasons. Yet, they're subject to misinterpretation by the search engines, even if a programmer uses it without knowing any better.
Ultimately, it's best to find safety in numbers. Common techniques such as display:none bear little risk of being flagged by an algorithm, provided you use it in a common way. SWFObject is also useful and likely to be safe when used properly.
However, any technique is subject to potential misinterpretation by search engines. The key is to use these techniques in moderation, and to use them in ways commonly found on the Web.
So, as soon as your programmer comes to you all excited about this novel new approach to implementing something, review it closely to see if the technique could be used by spammers. This is where your greatest risk comes from. Even if your intent is pure as gold, stick to commonly used techniques, and your risk of being accidentally flagged by a search engine will be greatly reduced.