What's the difference between hits, page views and unique visitors?

If you use our Web statistics system to track visitors to your site, you'll see references to "hits", "pages viewed", "unique visitors", "authenticated visitors", "entry pages", "exit pages", "spiders" and "robots".

A hit is counted each time someone views a file on your website. A single Web page can be made up of many files. For example, if your main Web page is one HTML file, ten image files, a JavaScript file and a CSS stylesheet file, that would show as 13 hits when people view that page. If you're just interested in how many people are looking at your site, you probably don't care about hits. Each request counts as a "hit" regardless of whether it came from a person, a search engine, an RSS reader, or any other source.

Pages viewed is the number of HTML pages or scripts that your visitors have looked at. A "page" is a hit that isn't an image, JavaScript or CSS file and which wasn't loaded by a search engine robot.

A visit is one or more pages viewed by one person. If someone goes to your site and looks at five different pages, for example, that counts as a single visit. If they return the next day and view more pages, that’s a second visit.

Unique visitors is the approximate number of different people who visited your site. It includes human visitors, but usually excludes search engine robots and other automated systems. Visitors are tracked by the IP address of the computer the person is using. If the same IP address returns to view your site within the month, that will add to hits and pages, but won't increase the number of unique visitors. Note that the number of unique visitors is only a rough estimate: the IP address of some visitors could change between visits (depending on their type of network connection), and different visitors can sometimes appear to share the same IP address if they're behind a "proxy server" at a large company or ISP.

Authenticated visitors are people who visited a password protected directory on your website. If you don't have any password protected directories, you will have zero authenticated visitors.

A page is counted as an entry page if it's the first page viewed by a visitor. Similarly, an exit page is the last page viewed by that visitor. You can use this information to tell which pages people use to enter and leave your site.

Robots and Spiders are computers that examine the content of your website, rather than human viewers. For example, when Google examines your website to index the content, that will be shown as a robot or spider.

The countries shown by the statistics program are calculated by determining which ISP a visitor is using, then checking the country of that ISP.

The bandwidth listed is the amount of data transferred when visitors look at your site. Our page about bandwidth explains more.

The number of pages and hits in the Connect to site from section count links from other sites and exclude clicks on links within your own site. So if someone reaches your site as a result of a link on another site, then views two more pages on your site, that will show as one page in this section, not three.

How accurate are the statistics?

They're pretty good, but not perfect.

A visitor's IP address may change between visits, and some visitors go through "proxy servers" — computers at large ISPs such as AOL that can "cache" their own copy of your website files. For example, it's possible for two AOL users to view your website, but for AOL to show the second person a "cached" copy of what the first person saw without connecting to your site again. That would show as a single visit in the statistics.

In addition, the listings of what IP addresses belong to which ISP, and which country that ISP is in, can sometimes be inaccurate.

Because of potential problems like this, your website statistics (like all statistics) should be considered useful information that might not be accurate down to the last detail.

(We also have a page that describes why other statistics programs may show different results.)