A/B Testing for SEO vs. A/B Testing for CRO
Split testing is a popular strategy used in digital marketing.
Marketers use it to see what version of their ad copy gets the most clicks, which subject line results in the most email opens, and which form color results in the most contacts. This type of A/B testing is a great way to leverage data to make the most lucrative decisions for your business.
However, there’s another type of A/B testing that’s a little less known, but just as powerful as the conversion-focused A/B testing you may be used to.
SEO split testing.
A/B testing for SEO aims to answer the question, “What changes to my website will positively affect my rankings and traffic?”
It’s a great way to take the guesswork out of SEO, but because A/B testing is most often associated with A/B testing for conversions, it’s often misunderstood. So we’re here to break down the major differences between the two.
A/B Testing for Conversions: What everyone thinks of when you say “split testing”
When someone says “A/B testing,” the version that most likely comes to mind is split testing for conversions. In other words, you’re testing to see which version of a single page produces the most favorable result.
Using a tool like Google Optimize or Optimizely, you would set up an experiment on a single page. Say, for example, you chose to test your home page’s contact form messaging.
You hypothesize that by making the form’s messaging more benefit-driven, you can increase the number of people who visit your home page and fill out a form.
So you show half of your home page visitors the version with the test messaging, and you show the other half the original messaging. Once you have statistically significant data, you compare the conversion rates of Group A and Group B.
If your hypothesis was correct and the new version resulted in higher conversions, you can now apply that as a permanent change to your contact form, knowing that it will result in more conversions than your previous messaging.
SEO split testing is a bit different.
SEO Split Testing: A data-driven approach to SEO experiments
People have been running SEO experiments for about as long as search engines have existed. Since Google’s algorithm is shrouded in mystery, we’re not left with many other options for figuring out what works.
The problem is, SEO experiments have traditionally been very manual and difficult to measure due to variables like:
- Algorithm changes
- Competitor behavior
Enter SEO A/B testing. It takes a similar approach to running SEO experiments as A/B testing for conversions, except instead of splitting visitors into groups, it splits URLs into groups.
Say you managed a multi-location website with organic landing pages for every city the business has a presence in, like a franchise website.
For SEO purposes, each of those pages has unique content, but they’re all fairly similar in format. The layout is probably the same, the H1 structure is probably the same, and the title tags probably all follow the same general format.
To run an SEO split test, you split these similar pages in half, leaving one half the same (the control) and the other half would inherit the test (ex: new title tag format).
The goal is to see the performance difference between the two sets of pages. After some time, you’ll hope to notice one group outperforming the other (ex: the test group pages’ click-through rates are increasing while the controls’ are not).
This gives you a strong sense of certainty that rolling out that change to all your pages will have a positive uplift on the organic performance of the website.
Split testing for SEO vs. CRO: Which is a better investment?
Although SEO and conversion rate optimization (CRO) activities are often pitted against each other, they’re both incredibly valuable in their own right.
It’d be like comparing apples to oranges. You can’t really do it. They’re just different.
To decide which split testing method you might want to invest in, it can be helpful to think in terms of your sales funnel.
- At the top of your funnel you have traffic. This is where you acquire potential new customers.
- In the middle of your funnel are conversions. This is where you turn traffic into leads.
- At the bottom of your funnel is where you close the leads and turn them into customers.
SEO split testing and CRO split testing feed different parts of your funnel.
SEO split testing is a great option if you’re struggling with the top of your funnel and want to focus on growing traffic to your website. CRO is a great option if you’re already getting a ton of traffic to your website but you’re converting leads at a much lower rate than you’d like. They can even work together to grow the top and middle of your sales funnel at the same time.
How difficult is it to run A/B tests?
A/B split testing platforms like Google Optimize and Optimizely make it fairly easy to run CRO experiments on your site. SEO experiments, on the other hand, have traditionally been viewed as complicated and labor-intensive.
RankScience wanted to change that, which is why we built our platform to completely automate the SEO experimentation process.
By adding a single line of code to your website (no developer resources needed!), we can deploy SEO split tests to see exactly which changes will result in a positive ROI for your website. And because we can do this with a single line of code, we can roll out these changes at scale, making this a perfect solution for large websites that face the impossible task of rolling out SEO changes to hundreds, thousands, or even millions of pages.
We did this for Coderwall and grew their traffic by 57% with a single A/B test. You can read more about that in our case study or drop us a line if you’d like to speak with someone on our team directly about running SEO experiments on your own website.
Organic has the potential to be your business’s most lucrative digital channel. Don’t leave it up to guesswork!