By definition :
- A visit is "identified" by the session ID.
- A visitor is "identified" by the IP address.
The validity/accuracy of a "repeated visitors" report is somehow jeopardized by some event out of control of any tracking program.
1) For any website hosted by IIS, the sessionID's are NOT unique, which mean that for a given day IIS may assign a sessionID of 123456 to a user, and few days later it may re-use the same sessionID, making the "visits" concept inaccurate.
2) IP addresses are assigned dynamically by most of the ISPs, however it appears that people have a cable connection are assigned a longer lease period for their IP address assigned by the DHCP server of the ISP, making it look as if it would be a static address, but this is not the way most ISP are working.
The majority of the ISP are assigning an IP address to the client for the time of the connection, but as always, there are execptions to the rule. AOL, with their silly proxy configuration, changes the user IP address on the fly, and the user is not aware of it. This type of ISP behavior, leads to a situation for which you will end up having multiple IP addresses "attached" to the same sessionID.
So, to make the report somehow accurate/reliable, you will have to consider the SessionID-IPAddress pairs and determine if the user is/was connected using AOL, which would somehow make the overall logic/process a tad painfull.
If you don't care about the accurary being impacted by the AOL users, you can limit the validation to the SessionID-IPAddress pairs approach, but in such case, the accurary will decrease as far as the visits count is concerned, since you may end up with the SAME sessionID assigned to different IP addresses, thus leading you in a situatin that would make you believe you're dealing with and AOL user, but you may not. You'll know only if the UserAgent string is analyzed again, to determine if this is in fact an AOL user.
Anyhow, "#'s, who (IP address), any other information" is a bit evasive as far as I'm concerned. Just for the sake of it, provide a more accurate description of the pieces of information you would like to see on that report.
Without having to know all the code logic of the tool, the help file is providing some leading tips as to determine what some counters are based on and so on. This may help you at formulating a "clearer" request.
Finally, it is also important to mention the version of MetaTraffic you're actually using. Reason for that is really simple. If you're still using version 0.993 for instance, new reports were addded and the database structure was altered to accomodate new needs that were implemented in newer releases, which at times may even have what you're looking for.
For instance the "Visits By IP Address" report might the report you're looking for, which is available in v1.2 and up if I recall well.
Information on repeat visitors
Good day,
By definition :
- A visit is "identified" by the session ID.
- A visitor is "identified" by the IP address.
The validity/accuracy of a "repeated visitors" report is somehow jeopardized by some event out of control of any tracking program.
1) For any website hosted by IIS, the sessionID's are NOT unique, which mean that for a given day IIS may assign a sessionID of 123456 to a user, and few days later it may re-use the same sessionID, making the "visits" concept inaccurate.
2) IP addresses are assigned dynamically by most of the ISPs, however it appears that people have a cable connection are assigned a longer lease period for their IP address assigned by the DHCP server of the ISP, making it look as if it would be a static address, but this is not the way most ISP are working.
The majority of the ISP are assigning an IP address to the client for the time of the connection, but as always, there are execptions to the rule. AOL, with their silly proxy configuration, changes the user IP address on the fly, and the user is not aware of it. This type of ISP behavior, leads to a situation for which you will end up having multiple IP addresses "attached" to the same sessionID.
So, to make the report somehow accurate/reliable, you will have to consider the SessionID-IPAddress pairs and determine if the user is/was connected using AOL, which would somehow make the overall logic/process a tad painfull.
If you don't care about the accurary being impacted by the AOL users, you can limit the validation to the SessionID-IPAddress pairs approach, but in such case, the accurary will decrease as far as the visits count is concerned, since you may end up with the SAME sessionID assigned to different IP addresses, thus leading you in a situatin that would make you believe you're dealing with and AOL user, but you may not. You'll know only if the UserAgent string is analyzed again, to determine if this is in fact an AOL user.
Anyhow, "#'s, who (IP address), any other information" is a bit evasive as far as I'm concerned. Just for the sake of it, provide a more accurate description of the pieces of information you would like to see on that report.
Without having to know all the code logic of the tool, the help file is providing some leading tips as to determine what some counters are based on and so on. This may help you at formulating a "clearer" request.
Finally, it is also important to mention the version of MetaTraffic you're actually using. Reason for that is really simple. If you're still using version 0.993 for instance, new reports were addded and the database structure was altered to accomodate new needs that were implemented in newer releases, which at times may even have what you're looking for.
For instance the "Visits By IP Address" report might the report you're looking for, which is available in v1.2 and up if I recall well.
Regards
Information on repeat visitors
I would be great to implement a cookie based feature then which would seem to bridge the gap between the IP and SessionID issues you refer to.
I can't tell but I would like to be able to see information on repeat visitors (#'s, who (IP address), any other information)