How Do Searchers Use Autocomplete? Finding Answers, Finding Products, and Researching Keywords

Searchers can select autocomplete suggestions as they type to complete the terms they intended to search for, but seeing a suggestion does not necessarily mean using it. General Google searches, product searches within e-commerce sites, and keyword research by marketers are three distinct uses of autocomplete and should not be treated as the same behavior. A 2020 study of Google search behavior recorded the use of suggestions in 23% of searches, but this does not establish that the same rate applies to all searchers today. Nor should the study be used to determine which types of queries use suggestions most often.

The study recorded a usage rate, but its accounts of which queries use suggestions most often conflict

A study published by Backlinko in 2020 analyzed 1,801 Google user sessions and reported that 23% of searches used autocomplete suggestions. That figure refers to searches—not to “23% of people habitually using autocomplete”—and is not a universal rate across search platforms.

However, the article offers two conflicting accounts of which queries use suggestions more often:

• The summary says suggestions are clicked more often in informational and local searches than in commercial queries.
• The main text says suggestions are used most often in commercial and transactional queries.

The overall usage rate from this historical sample can therefore be cited, but the article should not be used to rank query types by usage rate. The study’s author argues that autocomplete influences search behavior, but the percentage alone does not prove that it changed users’ original intentions.

With a clear target in mind, suggestions may not change what someone searches for

Warrior Forum contributor DABK explained that if he was not searching for a particular name, he would ignore Google’s suggestion of it; if he was already looking for it, he would search even without the suggestion. This offers one way to understand the behavior: whether a suggestion matches the original search goal may matter more than whether it appears at all.

However, this is a forum contributor’s self-report and judgment, not a survey of general search behavior. His account does not support his generalization that everyone behaves this way. Nor does it establish that a name appearing in suggestions will bring clicks or customers.

On-site product searches: completing names and discovering other options

A 2022 e-commerce article by awoo describes another scenario: customers may not remember a product’s full name, and autocomplete can provide related terms to help them find it. The article uses suggestions such as “laptop stand” and “laptop bag” for “laptop” to illustrate how suggestions can also present options beyond the original product.

This is an example of functionality and design, not a measurement of how often customers actually use suggestions. It illustrates how autocomplete can help complete search terms and explore products, but it does not establish that customers will necessarily buy additional items.

Keyword researchers are a different kind of user

A 2020 article on Neil Patel’s website recommends entering keywords related to a field, finding long-tail keywords in the predictions, and then using SEO tools to assess which are worth pursuing. These users are actively gathering content ideas, not necessarily carrying out everyday searches.

Observing how marketers experiment with search terms therefore does not directly reveal how ordinary searchers choose terms. Keyword research techniques are no substitute for observing actual search behavior.