How to Create a CRO Testing Plan
No marketing campaign is complete without a plan, and that includes CRO testing. It’s easy to think that A/B testing can be sprinkled on top of your existing marketing plan, but it’s more than that. Conversion Rate Optimization is a valuable tool that can inform the design of your digital marketing campaigns in addition to increasing your revenue, and enhancing the user experience on your website. In that sense, CRO is your recipe for success. To make sure you are getting the best out of your digital marketing, Here’s how to create a CRO testing plan.
Step 1. Website Audit
A solid CRO testing plan begins with an audit. A good CRO website audit has 2 steps; the first is to look at your website analytics to find underperforming pages and navigation drop offs. The second step is to look at those pages on the live version of your site to look for testing opportunities. Here’s what you should be looking for.
Website Analytics:
- Look for pages with high traffic
- Look for pages that are key to the customer journey
- Look for pages with a bounce rate or exit rate
- Look for pages that are already being targeted for ongoing marketing campaigns.
Once you have determined which page(s) you want to test, you’ll move on to the live site audit.
Site Audit:
- List every piece of content on the page. This includes banners, copy, images, etc.
- List every call-to-action on the page. This includes buttons, links, and written CTAs
- List every potential distraction or anxiety a customer might have on this page.
- List the ways the mobile website differs from the desktop version of the site
- Look at the navigation menu. How is it arranged, and is it easy to use?
Both the website analysis and live site audit can be broken down into small, more specific steps. However, for the sake of space and clarity, that will have to wait for another article. Instead, we are moving on to step 2.
Step 2. Prioritize your Opportunities
Once you have completed your website audit, you’ll have at least 5 potential areas to test: content, CTAs, distractions/anxiety, mobile optimization, and sitewide navigation. For each of these potential testing areas, you should craft a specific, open-ended business question you want an answer to. Examples of these questions are: “How much content do users want on this page?”, “Which CTAs do users want on this page?”, “How do users want the navigation menu to look?”. Each of these questions will serve as a testing purpose.
Now you have to prioritize which ones of these business questions you want to focus on first. That’s where the 3 E’s come into play.
The 3 E’s:
- Exist- What exists on the page?
- Exploit- How can we make what’s on the page better?
- Explore- What can we add to the page?
With CRO testing, we recommend starting large and working your way down to the smaller details. To do this, we highly recommend following the 3 E’s in order when you create a CRO testing plan. This will help you keep track of what you have already tested, and what you can test next.
This all sounds very confusing, but here’s how the 3 E’s Look in practice:
- Remove content from the pages to see if users react positively or negatively to its removal. If the conversion rate increases when the content is removed, that means users didn’t like that content.
- You can remove the content users didn’t like, or perform further tests to learn what users don’t like about it.
- Find out where users want each piece of content on the page.
- Find out how users want each section of content to look. These tests may include testing the size, color, layout, images, and font of the section.
- Test the addition of new content on the page, if desired.
- Test to see where users want the new content on the page etc. etc.
We always recommend starting with removal tests because they get you the most information with the least amount of effort. Not only are these tests easy to build, but they are the quickest way to identify areas on that page that need improvement.
With each round of testing, your changes should be getting increasingly more specific. First start with the page as a whole, then section by section, content by content, then part by part, until the whole page is done.
Step 3. Draft your roadmap
You have to have a testing roadmap. That is a non-negotiable. This roadmap will outline your testing strategy and keep you on track.
Once you know what your testing opportunities are, and you’ve prioritize them, it’s time to put them on your roadmap. Your roadmap should outline how many tests you are running per month, what pages you are testing, whether the tests are running on desktop and/or mobile, what variations you are creating (see next section), and what your success metrics will be for each test.
Keep in mind that your roadmap needs to be flexible. The results of one test will inevitably impact your future test ideas, and that’s okay! CRO is all about following the data. This means re-visiting your roadmap each month and updating it accordingly. Once you find a vein of gold, you want to keep digging until you’ve mined it all. Making sure you have a detailed excavation plan will ensure you don’t miss anything along the way.
To make sure you are planning ahead, but still keeping things flexible, we recommend planning for 3 months at a time. You can always add a “future test ideas” section to keep track of ideas you want to circle back to.
Step 4. Create your Variations
Creating your variations may be the most important step to creating a CRO testing plan because the variations are the test. I mean that literally. They are the hypotheses that will answer your business question.
Because your variations are hypotheses to the business question, they should all be directly related to the test’s purpose. Each variation should be a potential answer to the question. In order to get the most out of your test, each variation should be as different as possible, while still being relevant to the business question and being related to each other.
Remember to check your biases. Throw in at least 1 variation is the opposite of what you think will win, and always test your variations against the original. Sometimes the answer is the one your least expect! Here’s how this might look in practice.
Testing Purpose: “How much copy do users want on this page?”
Original Variation – 150 words of copy
Variation 1 – 250 words of copy
Variation 2 – 500 words of copy
Variation 3- No copy
Notice how all of these variations answer the specific question we asked, but differ in their execution. In this case the “negative” variation is to remove all the copy from the page. In this test, we believe that adding more copy will increase the conversion rate, but we also have to keep in mind that users may like the amount of copy on the page, or don’t want any copy at all.
It can be challenging to write good variations for a test. It takes a lot of practice to get it right. The best advice we can give in this endeavor is to be brutally honest with yourself when examining your variations. While writing your variations, ask yourself the following questions:
- What will I learn from these variations?
- What will I learn if 1 of these variations beat the original?
- What will I learn if half the variations beat the original?
- What will I learn if all of the variations beat the original?
- What will I learn if none of the variations beat the original?
If you can’t see how the potential results will serve your testing purpose you have two options.
1- Change your variations
2- Change the question you are asking.
Once you have determined your testing variations, plug them into your roadmap. Now you can start running your tests and getting answers!
Summary: It can be daunting to plan a CRO testing plan from scratch, but the information you can glean from well-planned testing can be invaluable to your business. Following these steps will help you craft a solid CRO testing plan for your website, but for best results you should work with a Conversion Rate Optimization specialist to get our campaign rolling today!