# Testimonial Carousel or All Quotes Visible A/B Test

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/testimonial-carousel-or-all-quotes-visible-ab-test

## Hypothesis

If testimonials help visitors decide, an open layout that shows them all without browsing may keep the primary action completion rate higher than a carousel. If the longer page pulls visitors away from the main action instead, the more compact carousel could come out ahead.

## When to run this test

Watch for later carousel cards that rarely enter the viewport, session recordings of people passing the slider without touching the arrows, or visitors who stop scrolling partway through a long testimonial block; each is a sign that homepage testimonials are not reaching visitors. The expected pattern is that the better format brings testimonials into view without pushing the next section aside, and carries visitors on to the main action. If your tracking counts a rendered testimonial as seen, set up viewport-entry measurement first.

## Primary metric

**Primary action completion rate**: Compare whether the presentation format changes completion of the target action.

## Other metrics

- **Testimonial visibility**: Measure testimonials entering the viewport, including content beyond the first card.
- **Testimonial interaction rate**: Measure comparable clicks or scroll behavior in both formats; slider use and visibility are different behaviors.
- **Visibility of the next page section**: Check whether the open layout pushes important content below the visible area.

## What to check before and during the test

- **Layout**: Use the same testimonial set and order in a slider and an open layout.
- **Visibility**: Measure viewport entry rather than treating rendered testimonials as seen.
- **Slider behavior**: Predefine navigation and autoplay; autoplay is a separate variable.
- **Testimonial count**: Keep the number and wording of testimonials unchanged.
- **Device**: Review touch and desktop behavior separately; make the slider keyboard-navigable and autoplay stoppable.

## What to watch when reading the result

- Let completed primary actions pick the winner; testimonial interaction rate is only a supporting signal. If testimonial visibility or visibility of the next page section fell in the winning format, find out why before accepting the result, and check whether important information became hard to reach.
- Keep a clear sign in the carousel arm that more testimonials can be browsed, and make transitions respect reduced-motion preferences. If either one is missing, the test measures a flawed setup rather than two presentation formats.
- Read touch and desktop results separately, since people interact with a carousel differently on each. The finding is tied to this set and order of testimonials; do not carry it straight over to a testimonial block on another page.
- If the difference in the main action is small and the visibility metrics point in opposite directions, report the result as inconclusive instead of picking a winner. Test whether chance could explain the gap with the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test).

## Frequently Asked Questions

### Does rendering a testimonial count as an impression?

No. Measure actual viewport exposure, including later carousel items. An expanded layout does not guarantee every testimonial is seen either.

### Is automatic carousel advance part of this comparison?

Playback behavior must be fixed beforehand. Comparing autoplay is a separate question. Preserve applicable pause controls, keyboard access and motion preferences.

### What if exposure falls while completed actions increase?

Use completed main actions and the predefined guardrails. Lower visibility alone is not failure. Check that important content remains accessible without claiming the behavioral mechanism is proven.