# Menu Category Order A/B Test on Listing Pages

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/menu-category-order-ab-test-on-listing-pages

## Hypothesis

An order based on category-page visits may move the sections shoppers enter most to the top and make them quicker to reach, but it may also push down high-revenue categories that get fewer visits. Whether the visit-based order or the current one delivers the higher purchase conversion rate may depend on whether easier access outweighs that loss.

## When to run this test

Analytics showing heavy traffic to categories listed low in the menu, session recordings of visitors browsing the menu at length, or site searches for category names the menu already contains suggest that shoppers struggle to reach the category they want. If the visit-based order helps, shoppers reach their category with fewer interactions and that easier access shows up in purchases. If shoppers misread the category names themselves, review the names first, because reordering does not fix an unclear label.

## Primary metric

**Purchase conversion rate (CR)**: Compare purchases among all eligible visitors assigned to each menu order, not only those who open the menu or select a category.

## Other metrics

- **Category click-through rate**: Compare clicks on categories moved higher in the menu.
- **Menu interaction depth**: Count interactions needed to reach a category; this does not measure elapsed time.
- **Menu scroll depth**: Use scroll behavior to describe navigation, not as proof that content was read.
- **Exit rate**: Track whether visitors leave more often after seeing the menu.

## What to check before and during the test

- **Order**: Compare the current menu order with the same categories ordered by category-page visits over the 90 days before launch.
- **Keep constant**: Keep category names, icons and menu structure unchanged.
- **Navigation**: Measure category clicks and the path to the selected category.
- **Conversion**: Compare purchases across all eligible assigned visitors, not only category clickers.
- **Device**: Review mobile and desktop menus separately.

## What to watch when reading the result

- More clicks on a category moved to the top do not make that order the winner. Judge purchase conversion across all eligible assigned visitors, not only menu clickers; if high-revenue categories lose sales, an order built on visits alone should not go live.
- Make sure high-value categories stay findable in both orders. Read the mobile and desktop menus apart; if the categories moved up crowd the mobile menu or menu interaction depth rises, do not ship that order on mobile. The finding only speaks for this category set and an order built from this period's visit data.
- If the gap between the two orders stays small, first confirm that every visitor saw the menu order assigned to their arm for the whole test. Also check that each arm received the share of visitors you planned (sample ratio mismatch); if the split is off, do not read the difference. If the split is clean, see whether the purchase gap is significant with the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test); if it is not, record that the test could not tell the two orders apart.

## Frequently Asked Questions

### Are more clicks on top menu categories sufficient?

No. Use purchase conversion as the primary outcome. Include all assigned eligible visitors, not just menu clickers, and examine whether the ordering creates additional purchases.

### Does reduced click depth mean faster category discovery?

Click depth measures interaction count, not elapsed time. Measure time to the category separately using consistent start and destination events.

### What is the risk of ranking menu items only by visits?

Important revenue-generating categories may become harder to find. Preserve their accessibility and avoid crowding the mobile menu with promoted items.