New vs returning visitors: what they mean and how to find them

New vs returning visitors, defined

A new visitor is a browser your analytics tool has no earlier record of. A returning visitor is a browser it has seen before. To tell them apart, the tool stores a random identifier in the browser on the first visit and looks for it on the next one. No identifier found means new. Identifier found means returning. A tool given a signed-in user ID can also recognise a person across devices (see the approach table below).

That one detail explains most of the confusion. Without a signed-in user ID, someone who visits on their phone in the morning and their laptop at night is two new visitors. Someone who clears cookies or uses a private window can get a fresh identifier on every visit. Someone who blocks the tracker completely is not recorded at all. And nobody is "returning" in a tool that keeps no lasting identifier at all (more on that below).

Why bother with the split? New visitors tell you how well you reach people. Returning visitors tell you whether what you made was worth a second visit. A site can grow on either, and the two need different fixes.

New vs returning users in Google Analytics and GA4

GA4 says "users" where older Google Analytics said "visitors", and it defines the terms precisely. These are the definitions from Google's own help page:

Three things follow from those definitions.

• "New" is tied to the date range you pick. A user is new if their first visit event falls inside the range. Widen the range and the new count changes without any visitor behaving differently.
• A new user is not the same as a first-ever visit to your whole site. It means the first visit GA4 recorded for that browser. If someone cleared their cookies since their last visit, GA4 starts again.
• The identifier is a cookie. On websites GA4 keeps a client ID in a first-party cookie called _ga to tell users and sessions apart, so the same cookie rules apply as everywhere else.

Google's own wording is a good guard against over-reading the numbers. It states that new users can exceed active users, and that new and active users won't exceed total users. When your new and returning counts do not add up to what you expect, check the date range and the metric names first.

How to find returning visitors in Google Analytics

Google moves report names around, so search for the metric names rather than following a menu path from a screenshot.

1. Pick the date range first. Everything below is relative to it. Use at least 28 days if you want returning visitors to have a fair chance to appear.
2. Add the metrics. In a report you can customise, or in an exploration, add New users, Returning users and Total users.
3. Split them by something useful. Page, channel or source is where the answers are. A single sitewide ratio hides the pattern.
4. Try the New / established dimension in an exploration when you want a time-based cut instead of the metric.
5. Work out the share yourself. Returning users divided by total users, for the same range and the same filter. Write down the range with the number.

If you still see numbers that look wrong, our guides on how to check website traffic and how to see who visits your website cover what each report is really counting.

What is a good new vs returning visitor ratio?

There is no good ratio for every site. Compare your own results using the same tool, comparable date ranges and the same traffic channels.

The ratio depends on things you can reason about yourself:

So treat your own ratio as the benchmark. Record it monthly with the date range, split by channel, and watch the direction. A falling returning share on a page you just rewrote is a finding. A sitewide ratio that differs from some other site's is not.

Reading the shift also needs care: more returning visitors is not always good. If a customer has to keep coming back because they could not finish the job the first time, that looks healthy in the report and is not. Pair the split with bounce rate and average session duration, and check whether returning visitors do the things you want, like buying or signing up.

How to increase returning visitors

People return when they have a reason and an easy way back. None of the following is a guarantee, but each one targets a real cause of one-and-done visits.

1. Find out who already returns. Look at which landing pages and channels bring back the largest share. Make more of that, rather than guessing.
2. Give people a way to be reminded. A newsletter, an account, a saved item or a feed turns a one-off visit into something with a next step.
3. Link to the next thing. A reader who finishes one page with nowhere to go leaves. Related links inside the text show them the next page is worth reading.
4. Publish or update on a pattern people can notice. A site that changes on a rhythm gives people a reason to look again.
5. Tag the links you send out. Put UTM tags on newsletter and social links so you can see which message actually brings people back. Our UTM parameters guide and the UTM builder show how.
6. Fix the first visit. Returning starts with a good first impression: pages that load, say what they offer, and answer the question that brought the person in.

Compare returning visitor counts and share over comparable date ranges, and check whether those visitors buy or sign up. If customers buy more than once, the customer lifetime value calculator turns order value, how often they come back and how long they stay into what one customer is worth, so you can see what a returning customer is worth against what it cost to win them.

Why cookieless tools count returning visitors differently

Recognising a returning visitor needs something that survives between visits. Tools make one of three choices:

This is why two tools on the same site report different new vs returning splits, and why neither is "wrong". They are measuring different things: one measures browsers it can recognise, the other measures nothing it can recognise. If a cookieless tool tells you everyone is new, it is being honest about what it can know.

Blockers and consent choices push the numbers the same way. A visitor who declines or blocks a tracking script is invisible to the tool, so they show up in neither column. Our Google Analytics and GDPR guide covers the consent side.

How Stickybeak counts returning visitors

Stickybeak is live analytics for every site you own, and it is in early access. How it handles this split follows from how it is installed, and the details are on its docs pages.

• Standard mode sets first-party cookies on your own domain: a visitor ID that lasts 1 year and is renewed on each page load, a session ID that lasts 30 minutes, and a count of how many sessions that browser has had. The session count is what tells new visitors from returning ones. Details are on what Stickybeak collects.
• Privacy mode sets no cookies, and the docs are plain about the trade: every visitor looks new, and there are no returning-visitor numbers. See privacy mode.
• Bots are labelled and kept out of visitor counts, so they do not pad the new side.

The same two limits apply as anywhere else. A new browser, a new device or cleared cookies means a person is counted as new, and a blocked script means the visit is not recorded. Where first-party serving is available to your workspace, the script comes from your own domain and there is less for a blocklist to match; see serving Stickybeak from your domain. The hosted script is third-party, so blockers can detect it more easily. We do not claim blockers never catch either.

Our own numbers

Stickybeak has only been tracking usestickybeak.com since 6 October 2026, so we have no new vs returning split worth publishing yet. When we do, it will come with the date range, the visitor total and what the count includes.

Sources, checked 17 October 2026: Google Analytics Help, Understand user metrics (total, new and returning users), Dimensions and metrics (New / established) and Data collection (the _ga client ID cookie). Stickybeak docs: What Stickybeak collects and Privacy mode.