Recent question we received:
We are running a banner ad on a magazine's website.
They are tracking the impresssions and hits our banner gets with Mambo.
We are tracking the hits using a specific url with Opentracker.
The number of hits do not match.
The differences are major - 3700 on their end about 10 on ours.
How do we reconcile this?
I have checked that the url they have embedded in the banner is the one we
It turned out that in this case we cannot track the traffic because we are unable to track this link. The link /banner has been designed to resemble a re-direct. We can see that because the link says bannerid / destination=
technically, this is not a referrer?, because there is a script in between that removes the origin.
the problem is the same for log file analysis - so that your log files will also not reveal the origin.
this redirecting process is done by PPC companies in order to prevent the information from being tracked by anyone but themselves.
basically, the banner is engineered to have the same appearance as though somebody had typed the url in.
We are sorry that we cannot help in this case. The option we see is to request that Mambo / the company sending you the traffic, add something like: ?ref=mambo or something like this at the end of the url, which you will then be able to easily identify in Opentracker.
Presumably, however, they "cloak" this on purpose in order to avoid accountability? We cannot say for sure.
If we were able to see / track the banners, we would offer this advice:
To set up your comparison, and be able to look at both sets of data and draw realistic / accurate conclusions, we suggest the following:
start by making two lists of exactly what is being defined. For example, we can tell you that Opentracker tracks people only, humans viewing your banner, and we can tell you how many of them click through to your site, etc.
Important Note: The process for implementing the code is a question of inserting the code = single line of html, into the page, or in this case the advertisement in question. This is only possible if the page / banner / email is written in html. If the banner is a .gif it is not possible to insert the code.
We measure unique visitors & pageviews
Find out what the other system tracks / defines & see where valid comparisons can be made. How/ what does Mambo measure? Are impressions or click-through measured? Are people who click twice measured twice? Are search engine? spiders & robots measured, etc. Do they provide you with any guidelines in terms of what/how they measure?
The second exercise would be to test somehow, for example, click through the banner & observe the process in Opentracker. This will give you an idea of what type of numbers we are generating.