Which type of SEO Test should I run?

How do you know if the changes you make to your site and content have made your Google rankings go up or down?

Tracking clicks, rankings, and click through rate from Search Console is the best way, but pulling this data out manually can take a really long time.

Thankfully SanityCheck allows you to setup SEO Tests to track the results of your changes automatically. You can now confidentially say when your changes are having a positive impact, or roll back changes that haven't worked.

 SEO Tests are really useful to run experiments on a small set of pages and getting a positive result before making the change site wide.

SanityCheck now supports two types of SEO tests.

1, Simple SEO Tests - this is where you compare 2 periods for a single page, query or the whole site. The period before a change was made, compared to the period after the change was made.

2, SEO AB Split Tests - if you have similar pages that can be separated into a control group that stays the same, and a test group where you make changes - you can perform an SEO AB Split Test.

Here are a number of things to consider.

i, If you have made a change on a single page - use a Simple SEO Test

ii, If you made a change to the entire site eg improve page speed by removing some 3rd party javascript - use a Simple SEO Test

iii, SEO AB Split Tests are more complicated to setup as you need 2 groups of pages - test group and control group. But they do give the most definitive results.

iv, Simple SEO Tests are open to misinterpretation due to Google algorithm changes, seasonal fluctuations etc. You need to keep an eye on the whole site performance and compare it to your Simple Test results. 

v, If your site is eCommerce, SEO AB Split Tests are really good as you usually have 100's of product pages that are the same. If you can make a change to 10 of them, and keep 10 the same - this allows you to have a control and test group that AB Split Tests need.

vi, Page Titles and Meta Descriptions can usually be changed through a Content Management System. Having good Page Titles and Meta Descriptions can give a really good boost in Click Through Rates from the search results and are a good thing to test. Testing Page Titles and Meta Descriptions can be done with either test.