top of page
Writer's pictureSerena Andrioli

Fundamentals of Test Planning

How do you plan your conversion rate optimization tests? We collected 4 principles that might help you with test planning.

Having a robust testing methodology results to be almost equally important than the frequency in which you conduct experiments on your website and/or on your marketing landing pages.

Let’s go together through the 4 principles of test planning.


1. “You have to test everything” 🚫

This statement is simply rubbish, physically you can’t test everything because you don’t have enough traffic to do so, the amount of users you have on a specific page is sometimes too low to allow you to test everything. You got to test important things!


There are things that don’t need to be tested: for example don’t waste time to test if you need to align the text in the middle of a CTA. Just do it! There are individual fixes that don’t require testing but can be implemented after a conscious users’ research. Major changes like a brand new layout, new prices structure require a proper testing.


2. Build good hypothesis ✍️

It is fundamental to have a test documentation sheet in which you will write down the logic of your test hypothesis. When you design an A/B test, you must have a hypothesis for each potential idea. The way you structure the hypothesis might influence the test itself.

Often hypothesis are wrongly formulated. Have a look at this one: If I do this, our main key metric will grow by x%. Please don't do that! I highly recommend to follow this approach: IF...THEN...BECAUSE.


3. Define goals 🎯

It’s important to distinguish between primary, financial and tertiary goals. The primary goal is connected to the page interaction itself: the primary goal is tied to the function of the page in which the test will be running. The financial goal is usually the business revenue and/or the net conversion rate uplift expected from the test. Note that increasing revenue might not be the only goal of your test. At the end of the day, of course money talks but you might end up to run tests for which there isn’t a direct contribution to revenue growth. The tertiary goal refers to the impact of your test on other metrics.


4. Establish prioritization framework 💯

You will find various prioritization frameworks online: almost all of them are valuable and well structured. Those frameworks help you to score your ideas, to basically assign to each idea a numerical value that enables you to choose which idea to test first. Now once you select your prioritization framework, you need to adapt it to your company’s needs. I would like to give you a glimpse of the parameters you should consider for your scoring system:

  • Is the change noticeable in 5 seconds?

  • Is the change about adding and/or removing an element on the page?

  • What is the average number of monthly visitors on the page?

  • Which is the estimated conversion rate uplift you expect?

  • Which is the starting CVR?

  • How long do you expect to run the test?

  • Is the idea coming from users’ research?

  • Is the idea confirmed by eyes tracking?

  • How easy is it to implement the idea? (How many hours of working for the implementation? Do you need tech to support you?)



コメント


bottom of page