A returning user is a visitor your analytics tool recognises from an earlier visit, because their browser still carries the identifier it was given last time. In Google Analytics 4, a returning user is one who had at least one previous session before the session being reported.
How a returning user is recognised
The first time a browser loads a page carrying the GA4 tag, GA4 records a first_visit event and stores a random client identifier in a first-party cookie called _ga. On later visits the tag reads that cookie, finds an identifier it has seen before and treats the visitor as returning rather than new.
You can see the split with the “New / returning” dimension in an exploration, or add the returning users metric to a report. Two details catch people out:
- The same person can be counted as both new and returning in one date range. Someone who first visits on Monday and comes back on Thursday is a new user on Monday and a returning user on Thursday, so the two rows often add up to more than the total.
- Returning means “this browser has been here before”, not “this customer has bought before”. A supplier who checks your stock page every week is one of your most loyal returning users.
Recognition only lasts as long as the identifier survives. Clearing cookies, moving from a work laptop to a personal phone, browsing in a private window, or using Safari, whose Intelligent Tracking Prevention shortens the life of cookies written by scripts, all make a returning visitor look new. If customers log in, sending a user ID lets GA4 join their visits across devices.
Why it matters
Few enquiries or purchases happen on a first visit. Someone looking for a kitchen fitter in Bristol or an accountant in Leeds usually compares several sites, leaves, and comes back when they are ready to act. The share of activity from returning users is a rough signal of whether your marketing brings people back, through email, remarketing, brand searches or simply being memorable.
It also changes how you judge channels. A paid social campaign may bring mostly new users who rarely convert on their first visit, while the sales arrive days later from the same people returning through a direct visit or a branded search. Judging channels only on the session that converted makes the first one look useless.
In the UK the figure is shaped by consent. Under PECR, analytics cookies such as _ga normally sit behind consent on a UK cookie banner, so someone who declines analytics cookies is not recognised at all, and someone who accepts on one visit and declines on the next breaks the chain. A falling returning-user share after a compliant cookie banner goes live is a change in measurement, not a loss of loyalty.
Common mistakes
- Treating returning users as returning customers. Match the figure against sales or CRM records before drawing conclusions about loyalty.
- Adding new and returning users together and expecting the total user count.
- Comparing the returning share of a Safari-heavy audience with a Chrome-heavy one as if both were measured equally.
- Celebrating a high returning share that comes from staff, suppliers or existing clients using a login area. Filter internal traffic and report logged-in areas separately.
- Sending an email address as the user ID. Google’s terms forbid personally identifiable information in Analytics, so use an internal reference instead.
How to act on it
Build a simple exploration with “New / returning” as the dimension and sessions, key events and session key event rate as the metrics. If returning visitors convert far better, the cheapest gains often come from bringing more people back: a newsletter sign-up, a saved basket, remarketing to recent visitors who consented, or a prompt follow-up after an enquiry.
Check your reporting identity setting, exclude internal traffic and annotate the date your cookie banner went live. Then judge the trend over months rather than individual weeks, because a single campaign can swing the mix for a short time.
If you want returning visits tied to real revenue rather than counted on their own, designing that measurement is part of the digital marketing strategy and consulting work I do with UK businesses.