# Testing Category Card Images on a Listing Page

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/testing-category-card-images-on-a-listing-page

## Hypothesis

If each category name sits next to an image that represents the category, visitors may recognize the one they want without reading every name. Among visitors who see the category cards, category click-through rate could go up as a result.

## When to run this test

When visitors skip the category cards for site search, enter a category and come straight back, or read the names one by one in session recordings, the right category is probably hard to find, and an image test is worth a look. If the image does its job, visitors recognize the category from it and move into it with less hesitation. Choose images only from products the category actually carries, and if the real problem is unclear category names, fix the names first, because an image does not rescue a vague label.

## Primary metric

**Category click-through rate**: Among visitors who see category cards, compare how many open the relevant category; assess purchase impact separately with conversion.

## Other metrics

- **Purchase conversion rate (CR)**: Check whether better category navigation carries through to purchases.
- **Category return rate**: Review returns alongside subsequent browsing; a return alone does not prove the category was wrong.
- **Page load time**: Make sure the images do not slow the page.
- **Categories seen**: Make sure visitors continue to see the same number of categories.

## What to check before and during the test

- **Images**: Add predefined representative images next to category names.
- **Keep constant**: Keep names, order, category count and destination pages unchanged.
- **Representation**: Do not make an image suggest a narrower range than the category actually contains.
- **Readability**: Keep category names easy to read.
- **Separate test**: Comparing photos with icons is a separate test; track category visibility on mobile in this one.

## What to watch when reading the result

- If category click-through rate rises but purchase conversion rate (CR) stays the same, the image may be getting visitors into a category without getting them to the right one. Do not ship a click lift that comes with lower conversion or fewer categories seen on mobile, and read category return rate together with the browsing that follows.
- What you learn here covers these category cards and this image set only. Comparing photos with icons is a separate test, so do not assume the winner carries over to a different image style.
- When the gap between versions is small, first check that visitors who saw the cards split across versions as planned (sample ratio mismatch). If heavy images make some visitors leave before the cards load, that version gets undercounted. With a clean split, judge whether the category click-through gap goes beyond chance with the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test); if it does not, do not declare a winner, and record it as a difference this sample could not resolve.
- If page load time differs between versions, the result carries the effect of a slower page as well as the image. Optimize the images before launch, and if a speed gap remains, close it and run the test again rather than reading the result as an image effect.

## Frequently Asked Questions

### How should category images be chosen?

Select representative images in advance. Do not imply a narrower range or show a product unavailable in the category.

### Does a higher category click-through rate mean more purchases?

No. Category click-through is the primary outcome here, while purchase conversion measures downstream impact. More category visits do not inherently create more sales.

### What if category images slow the page?

Optimize the images and monitor performance. Report loading costs as part of the tested experience, and check that visitors still see an acceptable range of categories.