If your website’s getting plenty of traffic, but relatively few conversions, chances are are that you need to make some changes to your site. CRO allows you to identify what resonates with your visitors and influences them to take the actions you want them to
Conversion rate optimisation (CRO) testing provides you with the data you need to increase the number of website visitors that convert into leads, sales, subscribers (or whatever the goals of your website may be).
But, just as pure instinct isn’t enough to decide what aspects of a visitor’s experience on your site influences them to take action, you can’t rely on guesswork to identify what to test in the first place.
Here’s how to plan your CRO tests effectively to help you get the most value out of the process. This is the most important part of the process; investing extra time in the planning stage will pay off tenfold later on.
Before you start conversion rate testing…
- To be generating enough traffic to measure.
- To have correct and accurate tracking set up.
- A CRO testing tool, such as VWO or Optimizely.
Successful CRO planning – what to test?
- Get a small group together representing various teams within your business and brainstorm ideas for improvement from a range of perspectives, including a fresh pair of eyes where possible.
- Keep a running wish list of suggestions for changes and enhancements to your site.
- Use Google Analytics data – look at popular landing pages that have a high bounce rate and examine the common paths to conversion, reviewing any pages with a high exit rate.
- Usability testing – spend a small amount of money on getting some user feedback (typically $1 per question). You can show certain pages on your site and set users
tasks to complete and then ask for their feedback.
- Put yourself in the shoes of your customers or users: how can you show them what they want to see when they want to see it?
- Look to improve high traffic pages, even if they’re already performing well.
Setting KPIs for CRO
How will you measure success for each test: in CRO terms, what will the ‘conversion’ be for each test? The more conversions you get, the faster you will get a statistically valid result.
This might be when people take the next step in the customer journey and won’t necessarily be your conversion page. For example, if you’re testing a landing page, getting a customer to the basket page might be a sufficient KPI.
The ‘PIE’ framework
Ensure that you’re making changes to pages that can have a significant effect on your conversion rate and average order value (AOV); there’s no point wasting time testing a page that gets no traffic, or a change so minor it won’t make a difference.
The PIE analysis framework sets a score for the potential, importance and ease of each test and ranks them to create your test plan.
What impact do you think the change will have on conversion rate and AOV (average order value)? Major changes will typically have a greater effect on conversion rate than smaller changes. Replicating something that works on another part of your site, or another site, often has a greater chance for success than completely new ideas. Look at previous test data to help estimate improvement.
How important is the page or feature? What is the size and value of current traffic to the page? Remember to speak to other departments for their plans and priority areas and to check that the tests you’re planning won’t cause issues for other teams and/or your agency.
How difficult is it going to be to set up the test and permanently change your site if it is a success? There is little value in the short term of finding an improvement that you can’t make to your current site.
Score each factor out of 10; sum the three scores and rank to create your test roadmap. Here’s an example:
Want to learn more about CRO?
Our free eBook takes you step-by-step through a tried and tested process to help you maximise the benefits of CRO for your business. As well as planning your tests, this shows you how to: build and check your tests; do simultaneous testing; run and monitor you tests; and apply the results of your tests.