Digital Marketing Tips for Analyzing a Split Test

By Ryan Deiss, Russ Henneberry

What do you do with the results of your split test to ensure success for your digital marketing campaign? After you run your split test, you either have a successful, a failed, or a null test result. After you’ve concluded the test, you can dig into the data to analyze what happened during the test period and determine your next steps. To analyze your split-testing data, follow these steps:

  1. Report all your findings.

    Collect and put your testing data into words. You can use a test report sheet or PowerPoint deck for this. Considering breaking your report into the following sections:

    • Slide 1: Test title, URL, timeline, and metric(s) measured
    • Slide 2: Hypothesis
    • Slide 3: All the variants you tested
    • Slide 4: In-depth results
    • Slide 5: Results showcasing the winning variant, conversion lift, and confidence rate
    • Slide 6: Analysis
    • Slide 7: Other observations
    • Slide 8: Recommendations
  2. Report your conversion range.

    The conversion range is the range between the lowest highest possible conversion rate. This range may be written in the form of a formula, as in 30% lift @@pm 3%, or you might say that you expect conversions to be between 27 and 33 percent. Be sure to report your conversion rate as a range. When you report a 40 percent conversion lift, but you really have a range of 35–43 percent, you’re doing yourself a disservice by not properly setting expectations for your results or your recommendations.

    Don’t let your boss or client think that the conversion rate is static. It isn’t. Set proper expectations by reporting on your conversion rate as a range. Tools such as Visual website Optimizer create this range for you.

  3. Look at each variant’s heat map.

    Observing each variant’s heat map helps you find new things to optimize and test. Place these finding in the “Other Observations” section of your report.

  4. Analyze key segments in Google Analytics.

    Here, you’re determining whether the test indicates a higher or lower conversion rate for certain types of visitors.

  5. Implement the successful variation.

    Ideally, thanks to the results of your split testing, you know what works. Now you can put that knowledge to work. Use your data to make educated decisions about what changes you should make on the page.

  6. If the result of the split test was null, pick your preferred variation.

    At this point, if your test has declared no winner from either variation, you can choose which one you’d like to implement. Use this data to develop a new hypothesis and create a new test.

  7. Use your findings to create new hypotheses and plan future tests.

    Optimization is a process. Your latest findings should feed into your future work. Here is where you can learn from segments, heat maps, or the test proper to develop your next iteration or fuel a test on a new page.

  8. Share your findings.

    At the very least, you should send your report over to your boss or client, and to your colleagues who have a stake in the test. If you want to go above and beyond, you could even publish your findings as your own primary research. Case studies are valuable resources that can establish you as an authority in the market and also generate leads within your market.