Measuring Conversion Rate Experts and Calling Winners

By this point you may have started implementing many of the tools we discussed in the last chapter and you’re working to optimize your landing pages and lower your bounce and exit rates. You might even be the king or queen of CRO jargon among your friends—using phrases like “conversion flow” and “user experience” around the water cooler.

You’ve got your most recent analytics reports and you’ve designed some simple A/B tests. But how do you know which tests are improvements and which ones aren’t moving the needle? And more importantly, what will you do once you’ve called the winners and optimized your first set of tests?

• It’s crucial you stay goal-focused. A lot of information is going to be available, so don’t get overwhelmed or bogged down and lose sight of what you’re trying to improve.
• Test everything—just not all at once. You might feel tempted to do a complete overhaul of your site if the first test ends unfavorably, but this will not result in actionable data. Test one specific change at a time so you’ll know what’s working and what isn’t.
• As we mentioned in Chapter 4, keep records of everything. Write down what you changed, why you changed it and what you expect to find. Take screenshots of all variations. You will need this data later.

Just like with all your efforts leading up to this point, data is key. Just because a test version of your site is prettier or cleaner doesn’t mean it’s the obvious winner. This is where the baseline we talked about in Chapter 4 comes in. This baseline is your site’s pre-optimization “average.” It consists of your typical conversion rate, average bounce rates for top landing pages, how much time the average visitor spends on your site and any other metric you think indicates engagement (which, as we’ve discussed repeatedly, typically leads to conversion).

We warned you in back Chapter 4 of the dangers of calling a test too early. Bear with us here, we’re going to discuss this stuff in a bit more detail.

In order for you to have confidence that you’ve in fact found a winner, your results need to be statistically significant. This means the margin of error, or likelihood that your results are merely chance, is low. In general, the larger the sample size, the smaller the margin of error.

If statistical significance, or significance level, is less than 5% probability, this means that your result is at least 95% likely to be accurate (or that it would be produced by chance no more than 5% of the time). Your testing software might use the terminology “95% chance of beating original” or “95% probability of statistical significance.”

[your testing software’s] significance calculation makes a critical assumption that you have probably violated without even realizing it: that the sample size was fixed in advance. If instead of deciding ahead of time, “this experiment will collect exactly 1,000 observations,” you say, “we’ll run it until we see a significant difference,” all the reported significance levels become meaningless.[1]

Yikes. In order to avoid what Miller refers to as repeated significance testing error, it’s critical to set a sample size and stick to it. Luckily, most testing software–like Optimizely, Visual Website Optimizer, and Google Content Experiments–includes this. For example, Optimizely recommends at least 100 people see both variants of an A/B test before results begin to become significant. If your software lacks this function, consider using Miller’s free Sample Size Calculator.

Something else to keep in mind is that tests may swing wildly over time. Just because one version of the test starts off strong, that doesn’t mean you can prematurely end the test. Stick to your predetermined sample size, and know that statistical fluctuations that will even out over the duration of the test.

For example, you can flip a coin 5 times, and even though the odds of it landing on heads or tails is 50%, it’s not unlikely to get 5 heads in a row. However, over 100 flips or 1,000 flips the averages quickly even out to 50/50.

For this reason, one sampling of 1,000 users is infinitely more valuable than several samplings of 10 –15 users each.

The winning version of a site is the one with consistently better metrics—it’s as simple as that. Remember, the metrics you’re looking at here are, of course, conversion rates, but also whatever other metrics you’ve determined affect conversion (engagement metrics in particular).

Big, clear cut wins are always great. But that’s not always how tests turn out. For example, a change that lowers bounce rates for the test site but doesn’t improve (though doesn’t worsen) conversion rates could still be considered a “winner”—an improvement worth keeping, because, after all, your goal is to improve your site piece by piece, one test at a time.

Recommended Read - A list of the Best A/B Testing tools to get desired results

Remember, you can learn just as much—or more—from a failed test as you can from a successful one, as long as you understand that the resulting data is an opportunity.

As we’ve mentioned nearly every chapter—optimization is not a one-and-done approach. There is no ultimate optimization goal, no perfect version of your website. Once you’ve declared a winner you should do one of two things: either you ask yourself how you can further optimize this particular element of your site, or ask yourself what else about your site can now be optimized. The best optimization efforts are cyclical and continuous.