Zero-Result Searches: What to Show When Nothing Matches
One in ten searches on a store ends with nothing. Most stores show a dead end and a shrug. The good ones show the shopper a way forward.
About 10.5 percent of ecommerce searches end in zero results, and 68 percent of stores treat that page as a dead end. Show the query, offer a spelling correction and alternate searches, and surface related categories and popular products, and the dead end becomes a second chance.
How often the search comes up empty
About 10.5 percent of searches on an ecommerce store end with no results. That is roughly one shopper in ten, and they are not casual browsers. A shopper who types a query is a shopper with intent, and the zero-result page is where that intent goes to die. Baymard's research across 344 ecommerce sites found 68 percent treat the no-results page as a dead end, and 56 percent of benchmarked sites fail to adequately support search needs at all.
The scale is what makes it expensive. A store doing 100,000 searches a month has 10,500 zero-result searches, and most of those shoppers leave. The fix is not better search alone, because some queries will always miss: the product is out of stock, the name is misspelled, the shopper used a term the catalog does not use. The fix is what the page does when the search misses.
Why most zero-result pages fail
The standard zero-result page says no results found and stops. No spelling suggestion, no alternate search, no related category, no popular products, no recently viewed items. The shopper has to start over from scratch, and most of them do not. They go back to Google, or to a competitor, and the store never learns what they wanted.
The failure is a data failure as much as a design failure. The store is not logging which queries return nothing, so nobody knows the top ten dead queries, and nobody fixes them. Salesforce's guidance is to analyze the top ten no-result queries first, because those ten queries are the cheapest wins on the whole site: fix the spelling, add the synonym, stock the product, and the page stops being empty.
The recovery pattern that works
A good zero-result page does five things, in order. It shows the query, so the shopper can see what the store understood. It offers a spelling correction, because a typo is the most common cause of a miss. It suggests alternate searches, using synonyms and related terms from the catalog. It shows related categories, so the shopper can browse instead of searching. And it shows recently viewed and popular products, so the page is never actually empty.
The pattern is not clever, it is thorough. Each element is a second chance, and a shopper who takes any one of them stays on the site. The worst zero-result page is an empty one. The best is a page that treats the miss as a conversation: we do not have that, here is what we do have, and here is how to find it.
What to measure after the fix
Measure the zero-result rate first, then the recovery rate: what share of zero-result sessions end in a product view, an add to cart, or a purchase. Track the top queries weekly, not monthly, because a new product launch or a seasonal term can create a dead query overnight. And track the queries that stay dead after the fix, because those are the products the store does not carry, and that is a merchandising decision, not a search problem.
The goal is not zero zero-result searches. Some queries will always miss, and a store that returns results for everything has a relevance problem in the other direction. The goal is a zero-result page that keeps the shopper, and a log that tells you which misses to fix first.
Keep reading
Related guides on this site: the three search analytics reports shoppers actually want how zero-results pages can rescue relevance triaging the zero-results log More from Product recommendations shoppers actually want.
Frequently asked questions
What is a zero-result search?
A search on a store that returns no products. It happens when the product is out of stock, the query is misspelled, or the shopper used a term the catalog does not use. About 10.5 percent of ecommerce searches end this way.
How common are zero-result searches?
About 10.5 percent of ecommerce searches end in zero results. On a store doing 100,000 searches a month, that is roughly 10,500 dead searches, and Baymard found 68 percent of stores treat the page as a dead end.
What should a no-results page show?
The query itself, a spelling correction, alternate searches, related categories, recently viewed items, and popular products. Each element is a second chance for the shopper to stay on the site.
How do I find my worst zero-result queries?
Log every search that returns nothing, and review the top ten queries weekly. Those ten are the cheapest wins on the site: fix the spelling, add the synonym, or stock the product, and the page stops being empty.