Conducting SEO experiments is important for success because SEO is always changing and your results always fluctuate. However, the idea of “testing” versus “experimenting” is slightly different, despite the fact that the two terms are often used interchangeably. The differences are, however, very slight:

  • Testing. Typically testing means having an idea and giving it a try while monitoring results. Usually you still have your constant, or normal SEO routine, in place so that you don’t fall behind and so that you have something to compare your tests to when the time comes.
  • Experimenting. This is more of a gamble when it comes to SEO. If you’re going for an experiment, you usually are trying something completely new and cutting edge, so you aren’t sure what result to expect. In other words, it’s often something you thought up on your own.

Understanding the definitions of each will help you understand how you can achieve balance in your SEO strategy. Testing has been a common SEO practice (a common practice for all aspects of business, actually) for quite some time, but experimenting is a bit more nerve-racking. You have to first ask yourself: What makes up a successful SEO experiment, and what must you avoid?

Different Elements When Experimenting with SEO

Most companies find that whether you’re testing a well-known method or experimenting with something new, many elements crossover. For example, everyone knows that having a goal and an approach for conducting the experiment is important. However, there are a few things that make up a successful experiment that are not so obvious. If you’re jumping into an experiment with too much excitement and not enough planning, you could be missing a few key elements to success:

  • Measurement. You cannot go into an experiment blindly. You might have goals set up and a plan to execute, but it won’t mean much if you don’t have any type of measurement strategy put in place. The biggest mistake here is the companies often figure out how they are going to measure something, but it’s really the “what” about measurement that’s tricky. Are you trying to measure the behavior of search visitors, or can information about non-search visitors be helpful?
  • Goals Change Procedures. The way in which you conduct your experiment could very well change based upon your goals, so this is something you have to consider. Don’t come up with one way of doing things without considering other alternatives. For example, experimenting with changes in SERP ranking, indexing, or crawling will require you to split up your data by device (smartphone, tablet, etc.) or even location. Isolating your search-referred traffic probably wouldn’t make sense; however it would make sense if you were experimenting in changes in traffic.
  • Double-Check. You cannot perform and experiment and then just run with it. Try it over and over again to make sure you have all the kinks figured out. You don’t want to base any other projects on something that was just a fluke (and it’s more common than people might think).

Understanding what makes up an “SEO experiment” is also important before you dive into the deep end. Essentially, anything involving traffic involving search falls under the SEO experiment umbrella. According to Michael Martinez of SEO-Theory, conducting any sort of test that compares online data to non-search traffic is considered a “conversion experiment.”

Have you conducted an SEO experiment? Were you successful or unsuccessful? Let us know your story.