Setting up A/B testing on a post-purchase offer
A/B testing lets you show different versions of an offer to different customers at checkout and see which one converts better. Cleverific tracks the accept rate for each variant and marks the one that's winning.
Note: A/B testing is available on multi-step and single-page funnels. It is not available on grid funnels.
Add a variant to an existing offer
- From the offer setup page, click Add a variant in the A/B Testing panel.

- The new variant is pre-filled with the same product as your existing offer. You can change anything you want to test: the discount, the copy, or swap in a different product entirely.

- Click Save in the top bar.
Once you have two or more variants, the experiment starts automatically when the funnel is live.
Traffic split
Traffic is split evenly across variants by default. An even split gives each variant the same number of impressions in the same time window, which makes the results comparable and helps you reach a conclusion faster. For example, if you're testing two variants and 200 customers check out, each variant is shown to 100 customers.
Reading the results
- In your Cleverific app, go to Analytics → Funnel Explorer

- Select your funnel in the dropdown.

- Scroll down to the A/B testing section at the bottom of the page.

Once each variant has at least 200 impressions, Cleverific runs a statistical analysis and shows one of three verdicts:
- X is winning: one variant has a 95%+ probability of being the best
- No clear winner yet: variants are close; keep the test running
- Not enough data yet: fewer than 200 impressions on one or more variants
Below the verdict, you'll see two charts: accept rate with 95% confidence intervals for each variant and revenue per impression by variant.
Pick a winner
When you're ready to commit to one variant, you can remove the others.

Note: Removing a variant on a running experiment permanently deletes its accumulated results and can't be undone.