A/B testing on WhatsApp
An A/B test on WhatsApp is the controlled comparison between two variants of a message or campaign, sent to comparable contact segments to find the more effective one. Variants can differ in text, image, button, timing or sender. The winning version, chosen on reads, replies or conversions, is then rolled out to the rest of the list.
The test only has value if it isolates one variable at a time: two versions identical except for the thing being changed, sent at the same moment to groups drawn at random from the same list. Changing several things at once makes it impossible to know what worked. The metric must also be chosen upfront: a test on button copy is judged on replies or clicks, one on timing is judged on reads.
The operational process: form a hypothesis, for example that a shorter message generates more replies; split a sample into two comparable groups and send the variants; wait an equal observation window for both; compare the chosen metric accounting for sample size, because on small numbers differences can be pure chance. The winning variant then goes to the rest of the list, and the documented result feeds the next tests.
With SendApp you can duplicate a campaign, vary the text or template, target comparable tagged segments and compare deliveries, reads and replies from the statistics.
Frequently asked questions
From theory to practice
Everything you read in this glossary can be done with SendApp: try it free, no credit card required.