Autocomplete is the “set and forget” dream of AB Split Testing. It:
Monitors your test until your experiment has enough data
“Completes the test”
Hides the losing variations
Alerts you to the test completion.
Changes the test status to “complete”
Less Work, Better Conversions
Autocomplete improves your conversions for ALL website visitors, as they will all see the better variation as soon as possible, rather than when you have a chance to complete the test manually. Its less work. We automatically hide the losing variations for you, but we dont delete the losers. In case you want to iterate, the other variations are still available to you when editing.
That there is a winning variation (over 95% confidence)
That the experiment has run for long enough.
That the variations have all received enough views.
What is 95% confidence mean?
We use Bayesian statistics is a mathematical approach we use to measure the likelihood of success for each version of your content.
As we collect data from your website visitors’ interactions, the Bayesian method continuously updates and refines its predictions. When we say we have “95% confidence,” it means that based on the data collected, we are 95% certain that one version will perform better than the other. This allows us to make reliable decisions about which version of your content is more likely to achieve the desired outcome.
Bit more nerdy than that please?
The Bayesian formula, also known as Bayes’ theorem, is expressed as:
P(A|B) = [P(B|A) * P(A)] / P(B)
In this formula:
P(A|B) is the posterior probability, the updated (posterior) probability of event A occurring given that event B has occurred.
P(B|A) is the likelihood, the probability of event B occurring given that event A has occurred.
P(A) is the prior probability, the initial (prior) probability of event A occurring.
P(B) is the evidence, the total probability of event B occurring.
Please note that while understanding the underlying mathematics can be useful, for most users of the AB Split Test plugin, the important thing is to understand that the plugin is using a robust, statistically sound method to analyze the A/B test results and provide reliable insights.
