Forum Moderators: DixonJones
I've read about this type of site appearing in logs as referrals, but not as exit links.
Firstly, how are they doing it and how is it showing up in our stats as an exit link (for it to show up, surely the page would need to include our tracking code and their link)?
Secondly, is it likely that some of our customers and possibly search engine spiders are being taken off to these other sites (we dropped overnight on Google for all positions, but this may well be pure coincidence)?
/claus
If your redirect count is set up like this (example.com is your site, script language is cgi, but it can be php, asp, whatever):
<a href="http //www.example.com/redirect.cgi?www.the-site-we-redirect-to.com">blabla
Then that script might be called from another page outside your domain, just by modifying the part after the "?". So, i could have a link on my site, which used your redirecting script to send me to a pr0n page.
If "www.example.com" is really a third party provider, the url will probably have an ID of some sort as well (to count for your site), in which case i only need to change the ID.
/claus
Our tracking system runs from a javascript and image (the javascript file runs on our server and the image on the third party reporting site). I always thought this type of system ran by reporting back to the third party site rather than by means of a redirect.
>> reporting system lists visitor exit links
...i reckoned that it could not be such as system as the one you are describing now, as such a system has no way of identifying where a user goes to after he leaves your site, or by which means he chooses to do so. On the other hand, how the user enters your site can be identified. It is possible that a browser entering your site would state that it comes from some odd site not in your business at all, that's not uncommon. That could be the refferal/log spamming you've heard about - or a legitimate link from a site on another topic.
What i thought you were using was a click-tracking software - these are found as third party solutions as well. They operate like i described above, by setting up all your links to go through some redirect script.
I assume (now) that your links are ordinary links and that you do not use a special redirecting url for them. If they tell you that you have a link that is followed from your site to another website, they must also be able to tell you which page that link was supposed to be located on. If they can't or if it's not there, then their software is not performing as it should.
>> I always thought this type of system ran by reporting back to the third party site
They usually do. I've been working with such systems ("webbugs") for some years, and the image/javascript combination at the bottom of your pages is the preferred way to do it.
>> how are they doing it and how is it showing up in our stats as an exit link
That is the questions that your third party software provider should really answer you. How *exactly* are they tracking links out of the site? If you are very good at Javascript you will be able to find out by examining their script source code, but these scripts are often very long, as they tend to track all sorts of things.
Could you send me a "sticky" with the company name, as these matters are what i do, and i would not like to miss it if this 3. party company really has found the holy grail, and hence are able to identify outgoing links - even if it's just for some specific subset of the available browsers, under some given conditions. I have a few questions i'd like to ask them out of pure professional interest.
Assuming they really can do this; then they seem to be messing up some databases/customer accounts - and that's not really a good sign, especially as it seems that your legitimate business has just caught a glimpse of the traffic on a not-so legitimate site belonging to another customer of theirs.
/claus
>If they tell you that you have a link that is followed from your site to another website, they must also be able to tell you which page that link was supposed to be located on
Yes, they do.
>how are they doing it and how is it showing up in our stats as an exit link
By they, I mean the perpetrators, not the tracking company.
>if this 3. party company really has found the holy grail, and hence are able to identify outgoing links
They only identify outgoing links from our site, not where people go after they leave our site.
>> The links are made up of another domain name, but include one of our file names,
>> ie, www.anotherdomain.com/ourfilename.htm
This might sound a bit complicated, but anyway, it's perfectly possible, and it even makes sense with the drop in rankings and all:
It could be, that "anotherdomain" (being in the shady business obviously) is in fact mirroring your pages (mirror = to show your content on their domain). If this is true, they are probably so smart as to only show these (your) pages to search engine spiders and not to real users. (you mentioned that this link would only open up a lot of windows)
Still, nothing on the internet is completely waterproof, and a link like the one you mention above might actually have been found and clicked by an ordinary user on their domain. Perhaps even by one of the persons operating this shady domain, just to see if it worked. Well, it did.
So, it seems it's perhaps not your link tracker anyway. However, such issues are hard to track down sometimes. Try searching Google for a few words that are included on this page but not likely on a lot of other pages, and see if the other domain comes up in the results. Try a couple of combinations, it might not be visible the first time, but if you find the right words you will see it.
You will need to search Google with the "duplicate filter" turned off. The way to turn it off is to insert this string after the url of a search in the address bar of your browser:
&filter=0Just append this to the url, and hit "go" or "Enter" or whatever you usually do. Then the same results will show again, but results that were filtered away before will now be included.
/claus
