Google’s SEO Guide On A/B & Multivariate Testing

Google posted some guidelines on how you can conduct A/B or multivariate testing and stay clear of any issues with being listed in its search engine, such as avoiding penalties.

No Cloaking

Google says you shouldn’t cloak, show its crawlers something that humans wouldn’t see. From its post:

Make sure that you’re not deciding whether to serve the test, or which content variant to serve, based on user-agent. An example of this would be always serving the original content when you see the user-agent “Googlebot.” Remember that infringing our Guidelines can get your site demoted or removed from Google search results—probably not the desired outcome of your test.

Use rel=”canonical”

Google says publishers should make use of the rel=canonical method to ensure that any alternative pages reference what should be the main one:
We recommend using rel=”canonical” rather than a noindex meta tag because it more closely matches your intent in this situation. Let’s say you were testing variations of your homepage; you don’t want search engines to not index your homepage, you just want them to understand that all the test URLs are close duplicates or variations on the original URL and should be grouped as such, with the original URL as the canonical. Using noindex rather than rel=”canonical” in such a situation can sometimes have unexpected effects.

Use 302s, Not 301s

Google recommends using the temporary direction method, a 302, over the permanent 301 redirect:

This tells search engines that this redirect is temporary—it will only be in place as long as you’re running the experiment—and that they should keep the original URL in their index rather than replacing it with the target of the redirect (the test page). JavaScript-based redirects are also fine.

Don’t Run Experiments Longer Than Necessary

If you’ve been running an experiment longer than Google expects one should run, it warns that you could face penalty. How long is too long isn’t said. Google just says:

Once you’ve concluded the test, you should update your site with the desired content variation(s) and remove all elements of the test as soon as possible, such as alternate URLs or testing scripts and markup. If we discover a site running an experiment for an unnecessarily long time, we may interpret this as an attempt to deceive search engines and take action accordingly. This is especially true if you’re serving one content variant to a large percentage of your users.

If you follow these guidelines, Google does not promise there will be no impact in your search results but does say there will be “little or no impact on your site in search results.”

Related Topics: Channel: SEO | Google: SEO | Top News


About The Author: is Search Engine Land's News Editor and owns RustyBrick, a NY based web consulting firm. He also runs Search Engine Roundtable, a popular search blog on very advanced SEM topics. Barry's personal blog is named Cartoon Barry and he can be followed on Twitter here. For more background information on Barry, see his full bio over here.

Connect with the author via: Email | Twitter | Google+ | LinkedIn


Get all the top search stories emailed daily!  


Other ways to share:

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • SpecificityPronk10
  • Authority Buzz

    Helpful information Barry as A/B split testing is definitely necessary. Head’s up: conducting in the first sentence should be conduct. And Mot in the headline should be Not.

  • Jennifer Lewis

    Thanks, Barry! Question – does this apply when you’re using Google Analytics to conduct the A/B test? 

  • Neeshu Srivastava

    can i use rel=”canonical” on my wordpress page numbers will it crawl the pages if so how to achieve it

  • Dan Carlyon

    I would say yes, as i’m sure people would still think of using analytics to deceive the search bots. I would follow the methods above no matter which way you decide to do your A/B testing

  • Dan Carlyon

    have you tried the all in one seo plugin? I think that should be able to add the tags for both posts and pages.

  • Experiments

    Super important pose from Google. Wehad Matt Cutts saying this already some months ago but sweet to see this on the official blog as well.

    Since we at are an A/B and multivariate testing tool, we get of penalty panic visitors that keep asking this, great to be able to refer them to the official Google blog.

  • Ivy Ken

    sir i want know . How to see any website daily visitor and daily earning and monthly earning and when sell site how many dollars selling site. please and me any one. i m weating for ans ,

  • Matt Coffy

    Great highlights for making sure that Google will not penalize you. I’m particularly interested about the 4th point for experimenting. Of course it’s always a goal to run tests to see what works, but yes, we have to set time and follow through with what’s necessary so that we won’t appear manipulative. Thanks, Barry.

  • Ken Villegas

    I thought 301 redirect is recommended? What if I have an old site and I want it to transfer to a new site, which is a wordpress site, then I will use 302 redirect for each page?

  • Jason Losover

    I’m sorry, but this is ridiculous. People are just OK with all of this extra work to run a simple A/B test, when it used to be so easy to do in Website Optimizer? Now you have to manually create an entirely new page (which isn’t necessarily easy to do in some CMS) and then worry about Google penalizing you for duplicate content. 

  • Danny Sullivan

    That’s recommended for permanent changes. If you’re doing testing, the pages in testing are really supposed to be temporary, thus the 302 recommendation.

  • Danny Sullivan

    Website Optimizer doesn’t get some type of pass on the last item. If you’re keeping it running constantly, despite it being a Google product, you could be prone to a penalty.

  • Carlos del Rio

    The last item is horrendous. How does Google define how long is necessary for me to get good results from a page.


    It’s only a matter of time for someone to actually publish the Bible Guide to Google Seo! I think someone had prior to Google Panda and Google Penguin but everything has changed dramatically and for the better. The Variant testing is a powerful tool for analysis if used in the right manner and for short term analysis. Content will ALWAYS be king but many factors such as age of domain, Google Authority and Google Knowledge is what everyone is monitoring more so post updates.

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest


Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States


Australia & China

Learn more about: SMX | MarTech

Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!



Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide