Quick answer
A unique visitor is one identifier, such as a browser ID, device ID or login, that visited a site at least once in a period and is counted once however often it returns. It is not always one person: one person on a phone and a laptop is usually two.
Formula: Unique visitors = count of distinct visitor IDs with at least 1 pageview in the period
What is a unique visitor?
A unique visitor is one identifier that visited your site at least once in a chosen period, counted once however many times it came back. The word "unique" refers to the identifier, not to the person: a tool counts a browser, a device or a logged-in account, and calls it a visitor.
That has two consequences. The same person on a phone and a laptop is usually two unique visitors. And the number depends on the period: a visitor who returns every day for a week is one unique visitor for the week and seven for the sum of the seven daily figures. Every number on this page is a question of "unique by what, over how long".
Google describes a close cousin in its own reports. Its "total users" metric is the number of unique users who triggered any event in the date range, and "active users" counts those who engaged with the site or app. Other tools use "visitors" or "users" for the same idea. The names differ; the work is in how the identifier is made.
How does a site decide two visits are the same visitor?
There are four common ways to tie visits together. Most analytics tools use one, or fall back from one to the next.
| Method | How it works | What breaks it |
|---|---|---|
| First-party cookie or local storage | The script stores a random ID in the browser and sends it with every visit | Clearing storage, private windows, and browser limits on script-set storage |
| Device ID | An ID stored once per app install or device | Reinstalling the app or a new device |
| Login ID | The site tells the tool who the person is after sign-in | Visits before login stay anonymous; people who never log in have none |
| Server-side hash, no storage | The server hashes network address and browser details with a secret salt; the result is the visitor ID | A new network or browser changes the ID; a daily salt makes the ID change every day |
Browser limits matter for the first row. WebKit, the engine behind Safari, states that its Intelligent Tracking Prevention deletes all of a site's script-writable storage after seven days of Safari use without interaction on that site, and that a February 2019 update limited cookies set through JavaScript to seven days. A visitor who comes back every 10 days can therefore be counted as new each time in a cookie-based tool.
Identifiers that are not stored in the browser avoid that limit but have their own. A hash of network and browser cannot tell two colleagues on one office network with identical browsers apart, and it counts one person on home Wi-Fi and mobile data as two.
Worked example: three people in one week
Sample data: three people visit a site over four days. Ana uses a phone on Monday and Wednesday and a laptop on Tuesday. Ben uses a laptop every day from Monday to Thursday. Chen uses a phone on Thursday. Together they make 10 sessions.
| Day | Visitor IDs seen | Daily unique visitors |
|---|---|---|
| Monday | Ana's phone, Ben's laptop | 2 |
| Tuesday | Ana's laptop, Ben's laptop | 2 |
| Wednesday | Ana's phone, Ben's laptop | 2 |
| Thursday | Ben's laptop, Chen's phone | 2 |
Now count the same four days three ways:
| Counting rule | Result | Why |
|---|---|---|
| Sum of the four daily figures | 8 | 2 + 2 + 2 + 2; Ben is counted four times |
| Unique across the four days, one ID per browser | 4 | Ana's phone, Ana's laptop, Ben's laptop, Chen's phone |
| Unique people, with login or one identity | 3 | Ana, Ben, Chen |
| Sessions | 10 | Visits cut by an idle timeout, not people |
Four correct numbers for one site and one week. If the daily salt rotates, the second row also becomes 8, because yesterday's Ben and today's Ben are different IDs. If Ana's laptop and Ben's laptop sit behind one office network with identical browsers, a hash of network and browser could merge them, and the second row would drop to 3 while the true number is 4.
Unique visitors vs users vs sessions vs pageviews
| Term | Counts | Goes up when | Example in the week above |
|---|---|---|---|
| Unique visitors | Distinct identifiers in the period | A new identifier appears | 4 |
| Users (people) | Distinct logged-in people | A new person signs in | 3 |
| Sessions | Visits, split by idle time | Anyone comes back after the idle limit | 10 |
| Pageviews | Pages loaded | Any page loads | counts every page |
For the difference between a visit and a pageview see page view vs visit, and for why monthly figures behave strangely see unique monthly visitors.
Why two tools report different unique visitors
Four causes explain most gaps, and you can check them in order.
- Different identifiers. One tool counts browsers, another counts logged-in people. A site where many visitors log in will show fewer "users" than "visitors".
- Summing days. A monthly card that adds up daily visitors counts a returning person many times. A range-wide figure counts them once. Compare like with like.
- Storage limits. Cookie-based tools lose returning visitors after short browser expiry windows; they report more "new" visitors than there are.
- Bots. Unfiltered crawler traffic adds identifiers. Tools differ in whether they remove it before counting.
Google also notes that user metrics are estimated and that thresholding can make total, active and new users differ from each other, so even one tool will not always reconcile with itself.
Common mistakes with unique visitors
- Adding monthly figures to get a year. A person who visits every month is 12 monthly unique visitors and one annual one. Count the year directly.
- Reading "unique" as people. It is people only if every person has one device and logs in. State what you count when you publish the number.
- Comparing a daily-salt count with a persistent-ID count. The first inflates repeat visitors over any range longer than a day.
- Treating a drop after a privacy change as lost traffic. A shorter-lived identifier produces more new visitors, not fewer visits.
How TapCub counts unique visitors
TapCub's website script writes neither a cookie nor local storage. The visitor ID is an HMAC of the visitor's network address and browser details, made on the server. IPv6 uses only the first 64 bits and the browser's version numbers are dropped, so a browser update does not create a new visitor. A new network, a new browser, or another device does.
The privacy preset decides whether the salt survives the day. The default, balanced, keeps a site-level salt, so the same visitor on the next day has the same ID and new versus returning can be compared. Strict rotates the salt every day, so visitors cannot be tied across days and new versus returning is unavailable.
In the reports, Web analytics › Overview counts visitors on the visitor ID. The visitor total adds each day's visitors, so someone who came on three days counts three times; for ranges of 93 days or less the data report also shows visitors deduplicated across the whole range, which is the second row in the table above. After login, the person ID can merge a person's phone and laptop, so user mode usually shows fewer than visitor mode. Compare visitor, device and user totals in All-platform analytics › Users › Identity map.
Sources
- Google Analytics Help, Total users, active users and new users: definitions of total, active and new users. Checked 2026-11-21.
- Google Analytics Help, Sessions: sessions end after 30 minutes of inactivity by default; session and user metrics are estimated. Checked 2026-11-21.
- WebKit, Full Third-Party Cookie Blocking and More: seven-day deletion of script-writable storage and the seven-day cap on JavaScript-set cookies. Checked 2026-11-21.
- TapCub documentation, counting modes and web metric definitions (docs.tapcub.com). Checked 2026-11-21.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Product Experience Specialist
Writes about product experience, practical workflows, and growth, turning complex processes into clear steps for everyday decisions.