You Can't Get There From Here

Impossible paths through a site, or are the stats lying?

[ken_b]

I understand that tracking a visitors path through a site can be difficult, so I haven't spent much time or given much credit to this data, but I'd like to get a better handle on what I see.

When looking at the visitor paths through a site, I commonly see some that simply don't make sense.

You can't get from page A to page B without going through at least one page in between.

Anyone have an idea of what to make of this?

[chadmg]

Say you can't go from Page A to Page C without going through Page B. Perhaps they are hitting Page B, then going on to Pages X, Y, Z, and A, then hitting the back button to get to B and then going to C? Can you check to see if they have ever hit that in between page?

[ken_b]

The paths that I'm most curious about are only two pages long.

Page A > Page C

But there is no link on page A that will take you to page C.

[webnerd]

Hi Ken_b
To the right of the GET for A and C in their log entries there should be a refering link.

cachea.a.a.com - - [05/Nov/2004:05:29:35 -0800] "GET /anydomain/C.htm HTTP/1.1" 200 979 "http://?.?.htm" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0)"

It will be always be there unless the user is direct linking to you (ie) bookmark, a webpage stored in his computer etc

[nancyb]

if there is no referer, then it is probably a bot and not a user.

[ken_b]

The bookmark thing is possible, I appear to have a lot of folks bookmarking the site.

But I'm not sure how likely it is that they would bookmark two pages, although I've done that myself a number of times.

I'm as confident as I can be that these are not bot hits, but never say never.

[bose]

Ken,

Could this be someone who has archived your pages off-line (locally on his harddrive) doing a "resync"?

In that case, pages could be refreshed out of order. I suspect it may be a home-grown (or a badly written) offline browser that may be sending you incorrect referrers.

Just a thought...

[McElvoy]

If both A and B can be reached from the home page, it's possible that this was a 4 page visit in reality: home (cached, for example by AOL), A, back to home (still cached), B.

What are the referrers of A and B?

[ken_b]

The cached page explains it nicely, thanks.

[Matt Probert]

I understand that tracking a visitors path through a site can be difficult

Try impossible. With caching along the chain (the local PC and intermediate computers), dynamic IP addresses, AOL issuing new IP addresses for every http request, it's a non-starter.

Save yourself some stress, don't worry about trying to track visitor paths.

Matt