A guide to effective SEO A/B testing on a budget
So, what’s SEO A/B testing, and why do you even need to think about it when you’re running your new venture? The answer is sort of simple.
SEO A/B testing for converting SEO strategy
Your site needs SEO to be visible to the search engines, and the way it works is a bit of a black box even to the guys in charge of running the engines. Over the years, the algorithms have become so complex that “what works best for SEO” is continuously changing. And this is where SEO A/B testing comes in.
For the SEO-responsible in your company SEO A/B testing also allows you to build concrete data that shows how much traffic you’re attracting with your SEO efforts and what is (and what is not) converting best.
SEO A/B testing also means that it will take longer for you to implement the changes on your pages, but it will also allow you to make significant changes proven to drive results.
Joy Brandon, Associate Director, SEO at Nebo, has covered the topic of Bootstrap A/B testing for SEO beautifully during INBOUND2020, and we’re here to reflect on some of the highlights.
Getting started with SEO A/B testing
Many testing tools allow you to run your SEO A/B tests, and many of them involve a considerable investment on your part. Tools like Google Optimize are great (and free!) for SEO split testing for your landing pages but don’t provide an insight into the performance of URLs structures, meta titles, and descriptions. And these elements get evaluated by the SEO algorithms first.
So, when you’re just getting started, you can create a simple test variation using Google Data Studio.
1. Find a hypothesis to test
Define what you’re trying to prove here. What is it that you’re testing? Ideally, you’d already have a set of pages with performing content. Find a hypothesis, write it out, and then find the pages you’d like to test it on. Check out our blog posts to make the pages rank higher and create content that sells for inspiration to your experimentation.
2. Find your A & B
Find the pages you want to test. Remember, you don’t want to compare apples to oranges. Choose pages that already have a similar amount of traffic and performance. Otherwise, your results will not be consistent.
3. Write a detailed brief
Create a detailed brief for your test: what do you want to achieve with this test, how often are you going to check the data, do you need additional development. When you define your test, you’ll see clearly if any assistance is required in implementation.
4. Create a dashboard
This is probably the most satisfying but also the most tricky step. Joy Brandon suggests using Google Data Studio to set up two identical dashboards: one for the original page and another for the test page. If you don’t have experience with Google Data Studio, there’s a useful guide for it. You can also use one of the Data Studio templates and follow the steps for connecting your data.
The essential recommended elements of the SEO A/B test dashboard are:
- Overview. Write a short experiment overview at the top of your Dashboard. It’s helpful to add the Dashboard in a report or show it to someone outside your team.
- Date. It’s quite self-explanatory: the date will help you filter page performance over time.
- KPI Scorecards. Your KPIs are the most crucial performance indicators. Put them at the top.
- KPI Time Series. These are useful to get a visual overview of performance over time at a glance. It’s important to note that it’s helpful to set up these dashboards a couple of weeks before the experiment starts, so there’s an apparent visual change in this indicator.
- Page breakthroughs. This element adds another granular filter for your page performance.
- Non-KPI data. Like revenue or Google Search console data. Any additional information that you find helpful or interesting to monitor for this test.
Use Google Data Studio filters to sort data. This way, you’ll avoid creating multiple segments in Google Analytics.
5. Monitor & reflect
Run tests and monitor the changes over time. We should mention that you should use the brief to outline every possible outcome. What if the test results come out negative? What if there are no changes? Be ready to accept and iterate on every output.
It would be best if you were running SEO A/B tests all the time. As we have mentioned, the algorithms keep on changing, and it’s essential to keep your ear to the ground and follow search engine optimization trends.
It’s important to perform SEO A/B tests to gather data and compare your SEO efforts’ performance. The search engine success logic keeps on changing all the time, and it’s a good practice to keep up and test the new hypotheses.