Question

Closed. This question is opinion-based. It is not currently accepting answers.

Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.

Closed 9 years ago.

Improve this question

I want to implement a simple in-house table that tracks user page views on my website. Without targeting some specific hypothesis, what is useful data to store? Eventually I ll use it to build graphs or decision trees to better learn about our user base. This is static (no javascript).

Things I can think of:

URL accessed
HTTP refer[r]er
HTTP Accept Language
Browser-agent
Session id
User id (if logged in)
Time visited

Answer 1

It depends on how public your site is. If your site requires authentication you can have more controlled statistics because you can trace the user (visitors) history. In the case the user does not require authentication you are limited to the information provided by the SERVER VARIABLES: HTTP_USER_AGENT; REMOTE_USER; REMOTE_ADDR; REMOTE_HOST; REMOTE_PORT; HTTP_COOKIE; HTTP_USER_AGENT.

I have implemented something like this for some non-public site each time the user logs on to the site, the information I m storing looks like:

User Key
Remote host IP
Date Logon
Last Request Datetime
Total time connected (minutes)
Last Request Minutes
Event/Action performed

Answer 2

Sounds like a good start,

I d be inclined to store visitor IP address, and derived from that via a geo ip lookup the location of the visitor.

Also you could consider reverse dns ing the IP to get an idea of the isp you re user is on, you might never use it but then again it could be useful if you have a report of downstream caching causing problems.

友情链接