A/B Testing and SEO: What You Need to Know to Succeed

Being indecisive is something that companies often struggle with when it comes to decisions about a website, but ironically enough a lot of this indecisiveness comes from the indecisiveness of consumers. You want one thing one minute, and then something turns you off the next. This is why A/B testing is so important for small businesses. It ensures that you website is satisfying the majority of consumers, and it allows for you to modify this as consumers modify their own purchasing decisions.

This then brings up the idea of SEO. If you’re constantly changing your website, how can you be sure that your SEO is going to remain in tact? Furthermore, is it all worth it in the end?

The SEO Repercussions of A/B Testing

It’s first important to understand that A/B testing doesn’t necessarily mean you’re changing your site entirely. That would be a redesign. A/B testing is just creating different versions of usually one webpage and then testing the different versions to see which has the best results. Companies typically use A/B testing software to complete a test. According to the company adogy.com, they got creative with their “work” page all thanks to A/B testing software, which you can check out by visiting the link. Each webpage has its own URL, and when someone visits that page the software will redirect some people to one page and some to another.

Despite different URLs, however, there are a few ways you can make sure that you have even higher visibility because your SEO remains intact:

• Use 302 redirects. These are only temporary redirects.

It’s important to remember that you don’t want to use a 301 redirect. 302 redirects are temporary, which gives the search engine bots a heads up that the page they have found might not be there for long. This will then send them back to your usual URL and keep that URL indexed; not the temporary one.

According to Adam Heitzman, Managing Partner of SEO Company HigherVisibility, “this also helps if you are planning to re-use these temporary pages in the future for other tests or if you have any users who might have bookmarked one of your testing pages or linked to it from another website. This is a common occurrence, and a 302 redirect will make sure that the link juice isn’t lost when the page is gone.”

• No cloaking. Do now show users and Google a different webpage.

This is an SEO black hat tactic that is against the Google Webmaster Guidelines. You cannot trick search engine bots into following one webpage but show another to users. This could accidentally happen when performing an A/B test, so keep that in mind to avoid being penalized (using a 302 redirect will help make sure this doesn’t happen by accident).

• Rel=”canonical” link attribute. This tag lets Google know a page is a test.

You want to use this tag on all of the webpages that you are testing so that Google knows they are only tests. Usually duplicate versions of each page will be created when testing, and although it’s very rare for Google to index one of your test pages, it can happen, especially if the test has been running for a while. The canonical tag on those duplicated URLs will make sure that they are not indexed.

However, you should not use this tag on your control page. You can watch this video to learn more specific information. Extra Tip: You can also choose to block the bots from your test pages by using robots.txt disallow command or a robots meta tag with noindex, both of which you can learn more about here.

• Timing. Make sure you aren’t running your tests too long.

You don’t want to be running a test for more than a few (2-3) months. If you run the test too long, Google might notice and then ask you to delete all of your other webpages. In other words, you want to make sure that you’ve made your decision in a good amount of time so that you have control over what stays, not Google.

The Takeaway

Google unfortunately hasn’t brought up the subject of A/B testing officially since 2009, but back in 2009 they did announce that they endorse A/B testing and actually want marketers to engage in these tests. It’s a great way to offer readers the best possible information in the best way, so not only is it something that won’t hurt your SEO, it’s something that will hopefully help it.

What has been your biggest problem with A/B testing? Do you have a story about losing your SEO value in the process? Let us know your experience and tell us your thoughts in the comments below.

Photo Credit: customersure.com

Amanda DiSilvestro gives small business and entrepreneurs SEO advice ranging from Google Analytics How-To’s to AdWords best practices. She works as a ghost writer and freelancer at amandadisilvestro.com and Coastal SEO Consulting, and works as the Marketing Manager for a Travel Company, Discover Corps, full-time. Visit her website to learn more!