According to IFABC Global Web Standards, a unique user (UU) is "An IP address plus a further identifier. The term "unique visitor" may be used instead of "unique user" but both terms have essentially the same meaning (see below). Sites may use User Agent, Cookie and/or Registration ID." Note that where users are not allocated IP addresses dynamically (for example by dial-up Internet service providers), this definition may overstate or understate the real number of individual users concerned.[1]
Unique users is a common measurement of the popularity of a website, often quoted to potential advertisers or investors, and measured over a standard period of time, typically a month. However, Greg Harmon of Belden Research says "may overstate" is a gross understatement. Remember, it's just an identifier of a computer, not a person. And usually, the computer is identified by a "cookie" which is most often specific to an individual browser on that computer. Since an increasing percentage of people in the United States (at least) now have access to a computer at home and at work or school, one may have to divide the reported total of unique users in half. Then another increasing fraction of people regularly delete cookies from their machines—presumably both at home and at work, and yet another large fraction use more than one browser on each of their machines. This means that for a typical news site, for example, which people might typically visit more than once a day to keep up with breaking news, the reported unique users might overstate the number of different people by a factor of four. On the plus side, for those wishing to impress advertisers or investors, the reported number of sessions or visits and pageviews are probably more accurate, so that smaller group of people visits much more often and looks at more pages than the raw numbers would suggest.[2]
Contents |
Similar to the TURF (Total Unduplicated Reach and Frequency) metric often used in television, radio and newspaper analyses, Unique Users is a measure of the distribution of content to a number of distinct consumers.
A common mistake in using Unique User numbers is not adding Unique User numbers across the same different dimensions. A Unique User metric is only valid for its given set of dimensions e.g. time, browsers. For example a website may have 100 unique users on each day (day being the dimension) of a particular week. With only this data, one cannot extrapolate the number of weekly Unique Users (only that the Unique User count for the week is between 100 and 700).
When calculating movement of unique users through conversion funnel the same time period must be used at every step.
Unique User counts for websites are typically counted by using cookies. When a browser visits a website, the website checks for the existence of a particular cookie. If the cookie is present, the cookie value is captured. If the cookie is not present, the website will create a cookie.
A unique visitor has the same general meaning as unique user and is simply an alternative title for the statistic.
When cookies are used to track Unique Users, then
1) Cookie Flushing will cause Unique User numbers to be inflated.
2) Browsers set to not accept cookies will be ignored or an IP address and/or User Agent String may be used as a proxy for a unique identifier placed in a cookie.