Optimization is a continuous process in which each experiment leads to the next experiment. In fact, there are often so many ideas that the backlog must be prioritized. But what if you run out of ideas?

A question often asked by our clients or during our trainings Is: when am I done optimizing? Our answer? Never! By continuously experimenting, you learn more and more about your customers which not only makes you better at predicting behavior, but also allows you to recognize earlier breaks in trends where behavior changes (this happens regularly, think of seasonal influences or changes in the economy).

You're never done optimizing

Of course, you may be running out of test ideas because you've been blinded by your own website for a while. High time to get your organization involved! 

As a CRO team, we are busy gathering as much information about our customers as possible. Sometimes we forget a little bit that there are many other teams in our company, for example, customer service, data scientists, marketing and perhaps the employees in the physical stores. These people can also know a lot about the purchase process or behavior on the website.

By including these colleagues in your optimization program, you suddenly have much more knowledge and experience at your disposal, as well as many more test ideas. Another advantage is that you can make people enthusiastic about optimization, testing and validation, which increases the chance that your optimization program will be a success. So my advice is above all not to wait until you run out of ideas, but to involve your colleagues in your experimentation program as early as possible.

Once you get people excited about optimizing, they are bound to start bringing in test ideas. Very nice! The challenge is to stimulate the flow of ideas but also check for quality. 

To avoid having all kinds of tests on the backlog that are not going to produce anything, you want to pick out the good ideas in advance. You can do this by having your colleagues answer a number of questions when they submit a test idea. 

These are the questions:

1. What do you want to test?
2. Can you test on this page (enough traffic and conversions)?
3. Why do you want to test this?
4. Do you already have proof that your idea is a good one (previous experiment, data, science)?
5. What is your hypothesis?

The questions we use for this are also the questions on which we prioritize. By showing that the questions affect prioritization, you also motivate your colleagues to substantiate their ideas. Because otherwise the idea will end up low on the prioritization and less likely to be tested than an idea that is well-substantiated.

So if you've gotten your colleagues excited about optimization, this is how you can encourage your colleagues to contribute qualitative testing ideas. And that way, you'll never run out of test ideas again!