How to Prioritise Your Tests

By Neil McKay


  • You can’t test everything at once.
  • Prioritising your tests will help you focus on what really matters.
  • Prioritise based upon importance, potential and ease of implementation.

how to prioritise tests

OK, you’ve done your conversion optimisation research for your website and it’s brought up hundreds of insights on conversion blockages. What next?

Do you dive in and begin testing what is at the top of the pile in the hope it’ll deliver a big win? No; instead, you should continue your systematic optimisation process by prioritising your tests.

Just think of all the research you’ve done so far. It’s been a complex operation and its provided a broad insight into your website and more importantly it’s allowed you to discover what actually matters.

You can’t test everything

Prioritise your A/B tests based on your research. This reinforces everyone’s understanding that conversion optimisation is a thorough process that gains positive outcomes. No gut feelings, nothing is left to chance.

We’ll follow a prioritisation framework to define the order of the tests. The intention is to move forward as quickly and smoothly as possible to run tests that have the best chance of making a sustained positive difference to your business’s conversion, income, and profit.

Create a bucket list

One of the best ways to streamline your prioritising is the bucket method (first shown to me by Peep Laja). It’s a really nice way to prioritise because it brings clarity to the process. Here’s how you categorise your buckets and what you put into each one.

how to prioritise tests

Slice the PIE

Once everything is placed in its respective bucket, the next step is to assign scores to the individual items in the buckets. You can do this by using the PIE Framework™ (WiderFunnel).

PIE stands for Potential (how much improvement can you make), Importance (how valuable is the traffic to the webpage—most important pages are the ones with the highest volume and the costliest traffic) and Ease (of implementing the test on the webpage).

PIE framework to prioritise experiments

In the above example, you score potential, importance, and ease out of 10 and then divide by three to give you a final PIE score.

The above framework is particularly useful when you have similar final PIE scores because you can quickly look across the columns to find the importance to the business, for example. And, that prevents debates about which test should be done before the other when the PIE scores are similar.

Or, you may prefer Bryan Eisenberg’s Time, Impact and Resources (TIR) framework. This is another very logical, step-wise approach.

Score out of 5

  • Time: 1 to 5 (5 being quickest)
  • Impact: 1 to 5 (5 being highest)
  • Resources: 1 to 5 (5 being least)

You multiply the scores together, and in projects with a large number of tasks, you can use a 10-point scale.

Create your own framework

When analysing customer’s websites, I combine the “bucket” and PIE approaches and then create a simple table to track everything logically. Here’s an example.

example of how to prioritise tests

Amalgamating the above frameworks helps me to focus on what’s important and what’s not. The more information that I list, the clearer it is for me to understand what I need to test and why. And, it is much simpler to communicate my findings with the website owner and the project team.

Whichever framework you prefer, the outcomes for using a scoring methodology amount to much of the same: focusing on what matters. And, that gets people focused on high priority tests that will deliver the most to the business as soon as possible.

Furthermore, the framework will help you to avoid costly and resource-hungry tests that may not amount to much in the long run.

Make prioritising your priority

Finally, you must always bear in mind that you don’t have an infinite amount of time for testing everything you would like to.

The more you prioritise your testing so that can you test the right things, and the more tests you do, the better your chances are of seeing sustained positive uplifts throughout the year.

You need to be aware that there is no such thing as a standard for selecting which pages to prioritise for testing—every website is different and that’s why you need to drill down into the data and use scoring and tabulation to make sense of it.

And, that’s why you work through the prioritisation framework systematically, while factoring in the things that make your business unique.

The framework is the key to keeping on track and helps you to decide on the next three to six months of optimisation work. Importantly, you won’t be left standing with your arms full of data and wondering where you begin to test. Prioritising will help you to get organised and it will show you which tests need doing first.

Neil McKay
Hi, I’m Neil and I’m passionate about helping organisations grow their businesses online by applying data, science and psychology. Our company Endless Gain is central to this passion. For more than 15 years, I’ve been helping businesses here in the UK, Europe and USA grow their online profits. Today’s businesses must be more scientific in how they approach business growth and that’s where we can help. In addition, our approach to conversion optimisation enables companies to maximise their marketing channel investments.

Our Clients