Autocomplete: The Search Box That Sells Before the Shopper Types

The shopper typed three letters and stopped, because the search box already showed them the product. That is autocomplete doing its job, and most stores get the details wrong in ways that cost the sale.

Autocomplete is the dropdown that appears as a shopper types in the search box, suggesting queries and products before the search is submitted. It is the highest-leverage element on the store, because it shapes the query before the shopper finishes it. Baymard Institute's 2022 research found autocomplete on about 80 percent of sites, but only about 19 percent get all the details right, and 58 percent of sites fail to copy the active suggestion into the search field when the shopper hits enter.

Eighty percent have it. Nineteen percent get it right.

Baymard Institute's 2022 research put autocomplete on roughly 80 percent of ecommerce sites, and only about 19 percent get all the details right. The 2026 benchmark is no better: 56 percent of 170-plus sites fail to support search needs adequately, and autocomplete is one of the most common failures. The feature is everywhere and the execution is poor, which is good news for the store that fixes it, because the fix is a short list of details.

The most common failure is the simplest one. Fifty-eight percent of sites fail to copy the active suggestion into the search field when the shopper presses enter, so the search runs on the partial query the shopper typed instead of the suggestion they highlighted. The shopper sees the right product in the dropdown, hits enter, and gets a results page that does not contain it. That is a sale lost to a bug, not to a competitor.

Related reading on this site: RelevanceRail and Query Understanding: When Search Has to Guess What the Shopper Means cover the neighbouring problems in more depth.

The details that decide whether it works

Speed is the first detail. The suggestions should appear in under 200 milliseconds, fast enough that the shopper never waits for them. A dropdown that arrives after the shopper has already finished typing is a decoration. Five to seven suggestions is the right count: enough to cover the likely intents, few enough to scan. More than that and the dropdown becomes a second search results page.

Images and prices are the second detail. A suggestion with a thumbnail and a price lets the shopper judge whether it is what they want without submitting the search. A text-only suggestion forces a click to find out, and every click is a chance to leave. Typo tolerance is the third: more than 30 percent of searches include a typo, so a search box that returns nothing for a misspelled product name is a search box that fails a third of its users.

Autocomplete shapes the query, not just the results

The deeper point is that autocomplete does not just help the shopper find a product. It shapes what the shopper asks for. A suggestion for a category the shopper had not considered, or a product in a price band they had not thought of, changes the search before it happens. That is why the suggestions should be ranked by business value, not just by popularity: the product that converts best for a query should appear first, even if it is not the most searched.

The failure mode to watch is the suggestion that sells the wrong thing. A dropdown that pushes the highest-margin product for every query trains shoppers to ignore it. A dropdown that shows the best match for the query trains shoppers to use it. The difference is visible in the numbers: the click-through rate on the first suggestion, and the conversion rate for searches that started with a suggestion versus searches that did not.

Frequently asked questions

What is search autocomplete?

It is the dropdown that appears as a shopper types in the search box, suggesting queries and products before the search is submitted. It shapes the query before the shopper finishes it, which makes it the highest-leverage element on the store.

How common is autocomplete on ecommerce sites?

Baymard Institute's 2022 research found it on about 80 percent of sites, but only about 19 percent get all the details right. The 2026 benchmark found 56 percent of 170-plus sites fail to support search needs adequately.

What is the most common autocomplete bug?

Fifty-eight percent of sites fail to copy the active suggestion into the search field when the shopper presses enter, so the search runs on the partial query instead of the suggestion they highlighted. The shopper sees the right product, hits enter, and gets a results page without it.

How many suggestions should autocomplete show?

Five to seven, with images and prices, appearing in under 200 milliseconds. More than that and the dropdown becomes a second search results page. Typo tolerance matters too: more than 30 percent of searches include a typo.