No business plunges blindly forward when it comes to big plans. If they did, the results would likely mean the end of the business. Planning and testing are vital components of any business operation and the same is true when it comes to SEO.
As any professional will tell you, search engine optimisation isn’t an exact science. Many sites succeed thanks to the experience of the person doing the optimisation, rather than to any formulas or calculations. If this experience is missing, testing is the best way to plot your way forward.
What to test
Your main objective with testing is to check on your keyword selection, placement and design factors. Your main keywords should have been selected on a fairly solid basis. Your related keywords, however, might be a little less certain. These lesser keywords and the placement of your keywords on your pages require thinking about.
How to test
You might want to begin testing with a large number of options. To test these, you could try multivariate testing, which involves setting up a number of options in different variations across a number of pages. You then test each page and see which elements combined for success. SEO commentators also often talk about A/B testing. This method comes in handy when you have narrowed down your options and wish to see which is the most effective.
With A/B testing, it’s important to have some control areas of your pages. In other words, don’t change the design and the content and the navigation all at the same time. A/B testing works best when there is a specific area that you want to test.
You can perform your tests in two ways:
- Set up two versions of your pages so that users are randomly directed to each version. The statistics from each page will tell you which was more effective. It’s important to note that this method is not optimal from an SEO point of view, as it may annoy the search engines
- Publish one version one week, and another the next. This method is slightly less scientifically rigorous, but can still provide valuable information. It also puts your ranking at less risk than the first alternative
With any kind of test, it’s a good idea to perform the test several times. This produces more reliable results than those collected from a single test. Testing only once leaves you very vulnerable to unknown factors. With multiple tests, you can be much more confident that your decisions are the right ones.
When you don’t have time to test
SEO doesn’t often allow time for thorough testing. Usually, the test stage comes after optimisation, when initial results are analysed and alterations made. It takes experience to test on the run in this way, and if you don’t have an SEO company on board it can put your optimisation at risk. You can discuss post-SEO analysis with us at SEO Consult. It’s a good idea for every business to know how to undertake the testing process when it comes to search engine optimisation.
