The Motive Web Design Glossary
hits (website statistics)
A hit is a request for a file made by a user-agent. User-agents include web browsers and search engine indexing programs, or spiders.
Each time a webpage is viewed a user-agent requests the individual files that make up the page from the computer where the website is stored (web server). A record of the hits received is automatically created and saved as part of monitoring the web server performance. This record is called a web server log.
A web server log can be processed and the resulting statistics interpreted to determine how people are using a website.
One webpage, many hits
Hits are commonly misinterpreted as a metric for website success, however the number of hits rarely translates to the number of people visiting a website. Nor are the number of hits the same as the number of webpages viewed. (Individual webpages accessed, or ‘page views’, are a more meaningful metric.)
A webpage is typically made up of a number of individual files. When a webpage is viewed, each of these files is requested from the web server, and each file request increases the hit-count for the website.
For example, if a homepage comprises:
- an HTML document;
- two stylesheets;
- ten images and;
- a JavaScript file;
then 14 hits will be added to the hit-count each time the homepage is viewed.
Hit-and-cache
Repeat-users generate fewer hits.
Determining website-use on the basis of hit-count is further complicated by the browser storing webpage files locally (caching). This is done to make using a website ‘quicker’. Once in-cache, a file is not requested from the server when the webpage is viewed. Repeat-users generate fewer hits.
Search engines are user-agents too
Although polite spiders often identify themselves to the server (making it easier to filter search engine ‘noise’ out of the web server log), each time a website is indexed, the hit-count is inflated by the number of webpages indexed.
Based on the fine-tuning of the Motive glossary (Nov-Dec 2004), the ratio is approximately 3:1; three search engine spider visits to each ‘human’ visit.
cache, crawler/robot/spider, HTTP, IP address, log file analysis, referrer logs, web server, traffic
References and further reading
- Page views are obsolete (Evan Williams)
25 Aug 06: Taking into account poor page construction and the increasing use of RSS to distribute content, should ‘time-spent’ be the preferred metric?
Glossary A-Z