Server-Side A/B Testing - Is it ok for SEO?

[michael-au]

Hi all

We are looking at running an upcoming server side experiment using Adobe Target's Node.js implementation and was wondering if it is safe to run server-side A/B testing and what considerations should be made?

For example, we have an upcoming redesign of our internal search result pages where we wish to use server side Target to show 50% of my traffic the old search design, and show 50% of our traffic a completely different search result page, are there any implications to SEO and organic search?

[michael-au]

Answering my own question but we came across this article that we have guided our experiment design on:

• https://www.abtasty.com/blog/ab-testing-seo/

In it, it mentions:

• not exclude Google’s bot from your A/B tests
• do not display pages that are too different from one another to your users. 

Specifically for redirect experiments it recommends:

• Never block Google’s bots via your site’s robots.txt file
• Place a canonical attribute on the variant page and set the value to the original page.
• Redirect visitors via a 302 or JavaScript redirection
• When a redirect test is completed, you must put into production the changes that have been shown to be useful