User signals are the behavioural cues search engines observe from how people interact with results and pages — clicks, dwell time, pogo-sticking, and return visits. Read in aggregate rather than one visit at a time, they hint at whether a result actually satisfied the person who searched.
Google is careful to say it doesn't use analytics-tool metrics like bounce rate directly, yet its 2023 antitrust testimony confirmed that aggregate click behaviour, via systems such as Navboost, does inform results. The nuance matters: treat these as symptoms to interpret, not a dial to game.
A SaaS help article ranked third for "cancel subscription" but saw heavy pogo-sticking — visitors landed, couldn't find the steps quickly, and returned to Google within eight seconds. The team moved the actual cancellation steps above the fold and added a short summary box. Over six weeks the return-to-SERP rate dropped noticeably and the page settled into position one. Nothing about the keyword changed; the experience did.
Because signals reflect satisfaction, the dependable way to improve them is to improve the page itself — the core aim of on-page SEO and good user experience.