1. Define the goal of the test
Before running an A/B test, decide which metric you want to improve:
-
conversion to application,
-
conversion to order,
-
number of payments, and so on.
Example: The goal is to increase the number of course purchases after watching a free introductory lesson.
2. Formulate a hypothesis
Define what change may lead to an improvement in the chosen metric.
Example hypothesis: The headline “Watch a free lesson and get a bonus — checklist” will result in more purchases than the headline “Free lesson to get to know the course.”
3. Decide what exactly you are testing
Change only one element at a time (headline, image, button, etc.) so you can clearly identify what influenced the result.
In this test: the headline of the free lesson page.
4. Prepare the variants
Create two (or more) options:
- Variant A: “Watch a free lesson and get a bonus — checklist”
- Variant B: “Free lesson to get to know the course”
5. Split the audience
Randomly segment subscribers:
-
50% see Variant A,
-
50% see Variant B.
6. Define the metrics
Decide which indicators you will use to evaluate the result:
- lesson views,
- clicks to the payment page,
- number of payments.
7. Set the test duration
Don’t end the test too early.
-
Minimum: 7–10 days,
-
or until statistical significance is reached (sufficient number of impressions and actions).
8. Analyze the results
Use BotHelp analytics and your CRM to compare performance across the variants.
9. Choose the winner
Implement the best-performing option into your main funnel and use it going forward.
If you haven’t found the answer to your question, ask us in the chat inside your account, write to us in Telegram at @BotHelpSupportBot, or email us at hello@bothelp.io.
Get 14 days of full access to BotHelp — a platform for creating newsletters, automated funnels, and chatbots.