# What to Put on a Zero-Results Search Page: A/B Test

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/what-to-put-on-a-zero-results-search-page-ab-test

## Hypothesis

If a failed search shows a spelling suggestion, popular categories and similar products instead of an empty screen, visitors can find another path to a product without giving up. That gives the zero-results recovery rate room to rise.

## When to run this test

Search logs make the case for this test when a noticeable share of queries return nothing and most of those visitors end their session on the empty screen; misspelled queries point the same way. When the recovery page does its job, visitors move on to a suggested category or product, and repeat searches and session continuation rise. If the real problem is that terms the catalog does carry fail to match, fix the matching first, start the test after that and leave the matching rules alone while it runs.

## Primary metric

**Zero-results recovery rate**: Among eligible visitors assigned on the zero-results page, measure the share who reach and view a product page from search results or a relevant category within the follow-up window.

## Other metrics

- **Search again rate**: Compare new queries among visitors assigned to the zero-results page.
- **Session continuation rate**: Compare visitors who continue interacting with the site in the same session window.
- **Purchase after zero results**: Count purchases among all eligible assigned zero-result visitors during the follow-up period.
- **Page load time**: Measure page speed after the recommendation block appears under the same device and network conditions.

## What to check before and during the test

- **Recovery experience**: Add one predefined recovery experience combining spelling suggestions, relevant categories and suitable products.
- **Content rules**: Keep recommendation sources and eligibility fixed.
- **Search field**: Keep the query editing field available and consistent in both arms.
- **Order**: Keep block order fixed; testing each block separately requires separate experiments.
- **Device**: Review mobile and desktop separately.

## What to watch when reading the result

- Recovery rate counts visits to a product page; check whether those visits turn into sales with purchase after zero results, counted over an equal follow-up window. If page load time shows the blocks slowing the page down, the gain may exist only on paper.
- Because the variant shows all three blocks together, the result belongs to the recovery page as a whole, and this test cannot tell you which block did the work. Split the read by device as well, and leave block-by-block comparisons to later tests.
- Check that recommended products actually relate to the query and that the page has not turned into a campaign showcase; irrelevant suggestions may draw clicks but they mislead visitors. Offer the spelling suggestion as a choice and never swap the visitor's query for it automatically.
- Visitors who hit zero results may be a limited group, so the difference can stay unclear for a long time. Feed the current recovery rate and the minimum lift worth acting on into the [sample size calculator](https://alidemirbas.com.tr/en/calculators/sample-size-calculator) to find the visitors needed per arm. Divide that by the zero-result traffic each arm gets per day to fix the run length before launch, and do not call the test early.

## Frequently Asked Questions

### Can this test identify which recovery block works?

No. Spelling suggestions, category links and product recommendations form one recovery experience. The result applies to the whole page. Keep block order fixed and test individual blocks later. Use product-view recovery after zero results as the primary outcome.

### How should recovery rate be defined?

Among eligible visitors assigned at the zero-results page, count those reaching and viewing a product through search or a relevant category within the follow-up window. Track repeat searches and continued sessions diagnostically, plus purchases across all assigned visitors.

### How should recommendations be selected?

Keep them relevant and avoid turning recovery into a campaign page. Fix recommendation sources and eligibility rules. Automatically applying spelling corrections may surprise users.

### What if recovery blocks slow down the page?

Measure performance under comparable device and connection conditions. Slowdowns can offset recommendation benefits. Preserve a usable query-editing field in both groups and keep search-matching rules fixed.