What are unique visitors in web analytics?
A unique visitor is a single person counted once within a given time period, regardless of how many pages they view or how many times they return to your site. If the same person loads ten pages in one sitting and comes back again the next day, they count as one unique visitor for the week.
The purpose is to measure audience size rather than raw activity. Page views tell you how much content is being consumed; unique visitors tell you how many distinct people showed up. Both matter, but they answer different questions. A site with 1,000 page views from 1,000 different visitors has broad reach. The same page view count from 50 highly engaged visitors tells a completely different story.
The challenge is identification: how do you reliably recognise when two separate visits belong to the same person versus two different people? On the open web, with no login required, that is a harder problem than it sounds. The most common answer has been cookies.
How does Google Analytics 4 identify unique visitors?
When someone visits a site using Google Analytics 4 for the first time, GA4 writes a first-party cookie called _ga to their browser. This cookie holds a randomly generated client ID, a string like GA1.1.1234567890.1718400000, that has no inherent meaning but acts as a persistent label for that browser. The default expiry is two years.
On every subsequent visit, GA4 reads the existing cookie and recognises the browser as a returning visitor rather than a new one. That is the entire mechanism: a cookie written on arrival, read on every return. The concept of "unique visitor" in GA4 is really "unique cookie."
This distinction matters because cookies are fragile. They live in a specific browser on a specific device. They get deleted. They get blocked. And in Europe, they require consent before they can be set. If the cookie is never written because a visitor declined your consent banner, GA4 has no way to identify that person at all. They are invisible to your data.
What changed from Universal Analytics to GA4?
Universal Analytics counted unique visitors per session period and surfaced them simply as "Users." GA4 introduced a more granular split. It now distinguishes between "Total Users" (anyone who triggered at least one event, including a basic page view) and "Active Users" (those with at least one engaged session, defined as a session lasting 10 or more seconds, including a conversion event, or containing two or more page views).
"Active Users" is GA4's default metric in most reports. Because it sets a higher bar than UA's user count, direct comparisons between the two tools tend to show GA4 reporting fewer users. The label changed; the underlying concept of cookie-based identification did not.
GA4 also introduced "New Users" as a separate metric and added Blended ID mode, which combines cookie data with Google's own signed-in user graph. If a visitor is logged into a Google account on both their phone and laptop, GA4 can sometimes link those sessions. Without a Google login, cross-device visits still count as two separate users.
Why is your unique visitor count probably wrong?
Three issues combine to make GA4's user count systematically lower than your real audience size.
The first is consent rejection. Studies from EU privacy regulators and analytics vendors consistently show that 20-40% of European visitors do not accept analytics cookies. When someone hits "reject" on your consent banner, or closes it without choosing, GA4 never writes the _ga cookie. That visit generates no data at all. Your traffic from France, Germany, or any other market where consent rejection is high is structurally undercounted.
The second is cookie deletion. Users who clear their browser history, use private browsing, or switch browsers appear as brand new users on every visit. If a loyal reader comes back weekly but regularly clears their cookies, GA4 counts them as a new user every single time. Your new user count is inflated; your returning user count is understated.
The third is cross-device browsing. The same person on their phone and laptop counts as two separate users in GA4 unless they happen to be signed into a Google account on both. For sites without a login, this is unresolvable. A visitor who checks your site on mobile during their commute and returns on desktop in the evening adds two users to your count.
None of these are edge cases. They are built into how cookie-based analytics works.
Where do you find unique visitors in GA4?
GA4 does not use the phrase "unique visitors" anywhere in its interface. The equivalent metric is "Users," with the caveat that which definition of "Users" you see depends on where you look.
In Reports > Acquisition > Traffic Acquisition, the default table shows "Users" (which maps to Active Users) and "New Users" side by side, broken down by traffic source. This is the most commonly used view for understanding where visitors come from.
In Reports > Engagement > Overview, the headline figure is also "Active Users" by default, shown over the selected date range. You can switch to "Total Users" using the metric selector if you want the broader count.
The Realtime report shows users active in the last 30 minutes - useful for watching live traffic but not for period-level analysis. For custom breakdowns combining user counts with dimensions like device type, geography, or traffic source, the Explore section lets you build free-form reports with more flexibility than the standard views.
Is there a more accurate way to count unique visitors?
Cookie-free analytics counts every visitor without relying on browser storage. There is no consent banner, which means there is no rejection gap. Every person who loads a page is counted, regardless of their privacy preferences or browser settings. The total visitor number is more complete because no one is silently excluded.
The trade-off is lower precision on the new-versus-returning split. Without a persistent identifier, it is harder to say with certainty whether a visit on Tuesday and a visit on Friday came from the same person. Cookieless tools use aggregated signals to estimate this, but they cannot match the session-linking accuracy of a tool that has a two-year cookie in the browser.
For most sites, the completeness matters more than the precision. Knowing that 800 distinct people visited this week is more actionable than knowing that 560 people visited (with 240 missing because they rejected consent) but being able to say with confidence that 120 of the 560 had visited before. TrackTrendy is built on this approach - no cookies set, no consent required, no visitors lost to the gap. The total count reflects what actually happened on your site.