# Bundle Options on the Product Page: A/B Test Example

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/bundle-options-on-the-product-page-ab-test-example

## Hypothesis

If the product page offers two- and three-item bundles next to the single item, visitors who want more than one can choose that in a single step. Revenue per visitor (RPV) may climb as a result, though the gain may shrink if bundles put off shoppers who want just one.

## When to run this test

Frequent orders with several units of the same item, the same item reordered at short intervals, or support requests about buying in larger quantities show that visitors want more than one. If bundles work, average order value rises without a clear drop in purchase conversion, and revenue per visitor goes up. Bundle demand may stay weak for a product people neither rebuy nor use several of at once, so check the quantity mix in your orders first.

## Primary metric

**Revenue per visitor (RPV)**: Compare total revenue per assigned visitor across the package presentations.

## Other metrics

- **Average order value (AOV)**: Check whether package selection changes order value.
- **Purchase conversion rate (CR)**: Check whether steering visitors toward a package affects purchase rate.
- **Package selection mix**: Review which package visitors choose.
- **Return rate**: Track returned units and orders over the same follow-up period.

## What to check before and during the test

- **Options**: Add the predefined two-item and three-item packages.
- **Terms**: Keep the single-item price fixed and define each package's quantity and price in advance.
- **Placement**: Use one fixed selection area near the purchase flow.
- **Device**: Check usability on mobile and desktop.
- **Set**: Keep the number of package options unchanged during the comparison.

## What to watch when reading the result

- A higher average order value is not a win by itself, because bundles may put off single-item shoppers and pull purchase conversion down. Decide on revenue per visitor, and use the package selection mix to see which bundle carries the revenue.
- Wait until both arms have had an equal follow-up window, then compare returned units and orders; if extra units come back later, the revenue gain will not hold. It speaks only for this product and these bundle quantities and prices.
- Ahead of launch, check that the single item is not overshadowed by the bundles, that the price and quantity comparison is consistent for every bundle, and that bundle buttons do not overlap on mobile. If you change the CTA design in the same test, you cannot tell whether the bundles or the button made the difference.
- Revenue swings a lot from order to order, so a small gap in revenue per visitor can stay unclear for a long time; do not make the bundles permanent until it settles. You can check whether bundles pulled purchase conversion down by comparing the two arms with the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test), which tests the conversion gap, not the revenue gap.

## Frequently Asked Questions

### What happens to the single-item option when bundles are added?

Keep the single item and its price available and visible. Fix bundle quantities and prices before launch, and do not let bundles obscure the single-item choice.

### Does a higher average order value mean bundles won?

Not by itself. Bundles may discourage people who want one item. Use revenue per visitor as the primary outcome and track returned orders and quantities over equal follow-up periods.

### Where should the bundle selector appear?

Use one fixed selection area near the purchase flow. Check that mobile buttons do not overlap. Keep CTA design unchanged.