Visits and unique visitors
A unique visitor is a web analytics metric that counts how many different people or devices visited a website in a given period. Each visitor is counted only once – no matter how often they return during this period or how many pages they view. In Google Analytics 4, this metric is called “users”.
Visits (in GA4: sessions), on the other hand, count every single visit. A unique visitor who visits a website ten times over a month therefore generates ten visits.
The two metrics complement each other. The number of unique visitors shows how many people a website reaches, the number of visits how often it is used overall. Within each visit, page views (page impressions) are generated. An example: a customer visits a shop on Monday and Thursday and looks at five pages each time. That makes one unique visitor, two visits and ten page views. A high proportion of returning visitors indicates customer loyalty, many new visitors a growing reach.
How analytics tools recognise visitors
The metrics are measured with web analytics tools such as Google Analytics 4. To recognise returning visitors, these tools store a random identifier in a cookie or use a device or app ID. If the same browser calls up the website again later, the tool recognises the identifier and counts another visit by the same user. The IP address is not used for this, as it changes frequently and can be used by many people at the same time. If the user is logged into a customer account, they can also be recognised across devices, provided a user ID is transmitted.
Why the numbers are never completely accurate
Visits and unique visitors are always approximations. If a visitor declines analytics in the consent banner, their browser blocks tracking or they regularly delete cookies, they aren't counted or are counted anew with each visit. If a person uses several devices or browsers – such as smartphone and laptop – they appear as several visitors; if several people share one device, they are counted as one. These inaccuracies can't be eliminated entirely. They can be reduced, however, for example with server-side tracking, Google Consent Mode, which partially models missing data, and a user ID for logged-in customers. In practice: trends and comparisons over time matter more than absolute numbers.