If you want to improve your testing program, there are a lot of process refinements that can lead to huge increases in efficiency and effectiveness. You can work to increase test velocity, focus on lowering your bust rate, cultivate a culture of testing and data driven decision-making, tackle analytics challenges or data discrepancies and a whole lot more. The goal is to create more tests that win—and the bigger the lift the better. But lift alone can’t describe the impact a test will have on your business. To make testing a strategic driver companywide it must make a significant contribution to growth. This is not always easy to achieve but fortunately, there’s one simple action that can help.
Download Now: Test Prioritization Score Sheet
The technique is this: When it comes time to develop a test idea, always focus on the one that presents the greatest potential impact. That’s it. And though it sounds obvious, actually identifying such ideas—and ensuring they find their way to the top of the queue—can be a challenge.
How do you determine potential impact?
Predictions, of course, are difficult. For the purposes of planning, however, a very simple calculation can be used to develop a basis for comparison. The first step in determining a test cell’s potential impact score is calculating the conversion value. To do this, you can use the following equation:
Traffic Volume X Conversion Rate = Conversion Value
Once the conversion value has been determined, a hypothetical lift rate—say 10 percent—can be used to estimate the potential impact of a winning test. The variation with the largest potential impact (conversion value X .10, in this case) should take precedence over other ideas.
When should this calculation be used?
The impact score can be used at two different stages of test ideation. First, the score can help direct testing roadmaps and focus testing on the areas of a site that represent the greatest opportunity for improvement. The impact score can also be used to prioritize test ideas once strategies have been developed.
Though simple, this relatively objective number can be helpful—especially when internal politics, personal preferences, and other factors have a tendency to dictate the focus of testing.
What about learning?
Impact scoring creates a clear focus on generating tests that lead to meaningful wins—and in most organizations, this is the primary goal. Actually achieving a win, however, requires an understanding of the customer and the target segment, something that can only be achieved through iterations of a test-and-learn process. And for some organizations, generating new learning is the main focus of testing.
By using a slight variation of impact scoring, it’s possible to arrive at an estimate of a test cell’s learning factor. Rating a test’s possible contribution to understanding customer behavior, it’s applicability to multiple channels, and it’s potential for driving ongoing strategy, will make comparisons a bit easier. To get a better sense of how such a rating system could work—and how to create compound scores—download our test prioritization score sheet.
Impact scoring can’t tell you which test idea is best—or guarantee a big win every time—but it does provide a reference point that can be used to compare ideas, helping you prioritize items on your roadmap or in your test queue.