In search engine optimisation, cloaking is essentially the process by which the search engines will be shown a different version of content to the end user. So, you might click a link thinking you are about to visit a website that offers an illuminating and scholarly reflection on the early works of Fyodor Dostoevsky, only to be presented with a video of a cat solving a Rubik\’s cube on a unicycle…Okay, that\’s a silly example, but in extreme cases cloaking can cause the searcher to click on something they think is a good result, only to be shown something that has no relevance to their search query, gambling, pornographic or pharmaceutical content, for example.
Traditionally, there are three methods that can be employed to achieve this effect:
1. Identifying the User Agent Name of the Search Engine bot
2. Identifying the IP address of the SE bot
3. Detecting both the User Agent Name and IP address of the SE bot
There are a couple of good examples I\’d like to share that illustrate how content cloaking can be used for (arguably unethical) SEO gain and how it can get you in big trouble. Amazon, the huge online retailer use a form of cloaking, but get away with it, whereas BMW had a rather public run in with Matt Cutts and his merry band of spam fighting men.
It seems that the key to determining where cloaking is acceptable, and where it is not, is intent. If the search engines feel that you are trying to manipulate ranking and results through showing different content to bots and users, then you face penalisation. If, however, the cloaking doesn\’t interfere with the search engines noble objective to produce the most relevant search results, then you are less likely to incur a penalty…as long as you stick to the rules.
Still, there are certain occasions when it is necessary to show different content to the search engines and users
Site Design Testing
This is also known as Multivariate Testing or A/B Testing. An example of when this type of testing might occur is if a website owner is creating a landing page for a paid ad. The site owner might create two versions of the landing page to test which one generates a higher conversion rate. In this case, you might display different content to the user using JavaScript, cookies, session ID\’s, or a combination of the three.
However, in order to avoid the risk of being penalised for showing duplicate content you would need to nominate one page to Google (using the rel=\”canonical\” tag) for indexing. In this circumstance there would be occasions where the users are being presented with different content to the search engine bots, but because it\’s going to benefit the human user, there is very little risk of incurring a penalty. Google also had its own Website Optimiser Tool to help website owners perform this action.
Unspiderable Navigation
As mentioned in another blog post, if content is formatted in a certain way it is rendered unspiderable to the search engines, these include JavaScript, Java applications or Flash. In order to retain the visual benefits of Flash and JavaScript and still have their site fully accessible for the search engine crawlers, site owners might create a crawlable equivalent of the content in the HTML. The easiest way to do this is by using CSS layers; using these you can create a layer that is human-visible (using Flash and JavaScript etc.) and another layer for the search engines (with text based links in the HTML). Again, this is technically showing different content versions to search engines and visitors, but because there is no malicious intent behind these actions, the risk of penalty is minimal.
Duplicate Content
Creating duplicate content is almost inevitable when websites are created. In a previous blog post I detailed some of the ways in which duplicate content can occur, and the consequences. Again, it can be related to intent, duplicate content can and will occur naturally, but it can also be forced with malicious intent. Imagine you own a website that sells textiles. On your site you might list the same product several times on different pages; on one page it could be listed by size and on the other by colour. If this kind of situation is replicated on a larger scale then the site might risk a penalty for duplicate content appearing on their site. There are several ways to counteract this risk. One might be to use the robot.txt file to restrict the search engines access to a page. Similarly to the previous example, a solution can also be reached using the rel=\”canonical\” tag and nominating only one version of the page to be indexed by the search engines.
These examples are technically cloaking, although, perhaps the more suitable or \’politically correct\’ phrase might be conditional content delivery. If content delivery is conducted in a way that can be interpreted by the search engines as an attempt to manipulate rankings then it is likely that the website in question will face penalisation. If conditional content delivery is used to improve user experience, and to improve the crawlability of a website, then that is unlikely to be perceived as a form of malfeasance. I suppose the final warning I would give to site owners is to be extra diligent when a situation occurs that will require different content to be shown to the search engines and visitors. The strategies listed above are only a few that can be employed to segment content delivery positively; but great care should always be taken when implementing these techniques in an SEO strategy and an awareness of the risks involved is also important.
