# Swapping a featured homepage product: +5.6% conversion, early signal

- Outcome: Won
- Page type: Homepage (Ecommerce)
- Elements: Product cards & quick add
- Test run by: [Clean Commit](https://liftarchive.com/practitioners/clean-commit/)
- Commentary by: LiftArchive (our analysis, not theirs)
- Original source: https://cleancommit.io/ab-tests/replacing-recommended-product-in-the-buy-now-section/
- Page: https://liftarchive.com/test/cleancommit-replacing-recommended-product-in-the-buy-now-section/

Replacing one product card in the homepage Buy Now section with a product the store wanted to promote raised conversion rate by 5.6% and revenue per visitor by 8.0%. Clean Commit ran this test for an anonymized Shopify store and published it in its A/B test library. It is an early signal: the test ran 11 days and had a 71.1% probability of beating the original.

## Result

**+5.6% conversion, early signal**

- Conversion rate: +5.6%, with a 71.1% probability of beating the original. Not significant.
- Revenue per visitor: +8.0%.
- Add-to-cart rate: +7.1% and orders +4.0%.
- Click-through rate: -0.2%. Browsing barely changed.

## Control

The homepage Buy Now section showed the store's original set of product cards.

## Variant

One card was swapped for a product the store wanted to push. Its image, name, price and Buy Now button matched the styling of the other cards.

## Why it likely won (our reading)

- The promoted product may simply suit more homepage visitors than the one it replaced.
- Clicks stayed flat while purchases rose, which suggests the new product was more likely to be bought once seen.
- Desktop gains looked larger but came from only 1,555 visitors, so they say little.

## Caveats

38,732 visitors over 11 days, with a 71.1% probability of beating the original. Treat it as an early signal, not proof.

## How to run this test

- Hypothesis: If we feature a different product in the homepage Buy Now section, then conversion rate will rise because more visitors see a product they are likely to buy.
- Primary metric: Conversion rate
- Guardrail: Revenue per visitor
- Guardrail: Sales of the product removed
- Who should test it: Stores with a homepage product section and a candidate product that sells well once seen. Run it for at least two full weeks, since product swaps can be swayed by short-term demand.

---

Credit: Clean Commit ran this test and owns the result. Summary and commentary by LiftArchive (https://liftarchive.com/test/cleancommit-replacing-recommended-product-in-the-buy-now-section/).
Usage: Free to use, quote and build on, with credit and a link: name the practitioner or organization that ran the test and link to its LiftArchive page. The results belong to the practitioners who ran and shared them; the summaries and commentary are LiftArchive's.