Problems with Sedo Link Tracking

[Whizzbang]

Hi all,
I just posted a document on a problem that Sedo is having with their click tracking system that may be of interest to those parking there.
http://whizzbang.typepad.com/whizzbangs_blog/2007/05/problems_with_s.html

I think that some of us could be losing a bit of money with this one.

 

[mvl]

Can you quantify the problem? Do you know how widely used those browserversions are? They look pretty mainstream to me.

 

[Whizzbang]

Yes, they are pretty mainstream. We managed to triple the revenue for some of our domains for item 2 that I spoke about in my blog and item 1 we are still quantifying the affects but it's looking to be about 20%.

 

[SedoCoUk]

Hi all,
I just posted a document on a problem that Sedo is having with their click tracking system that may be of interest to those parking there.
http://whizzbang.typepad.com/whizzbangs_blog/2007/05/problems_with_s.html

I think that some of us could be losing a bit of money with this one.

Hi Whizzbang.

I’ve just checked with our parking department. It looks like the reason the click was not forwarded to the advertiser is that the IP was temporarily blocked because there were too many clicks from that particular IP address. It looks like there were over 50 clicks from that IP.

In order to protect the advertisers, we temporarily block IPs from clicking. This is also the reason for the view not being counted in the example you provided in the screen shot. Temporarily blocking IPs is also in the interest of every domainer to ensure high domain traffic quality. I hope this clarifies. Also, you should have just received a detailed log file from Uli Priesner. Please be on the lookout for that in your email!

Thanks,

Monica

 

[AlienGG]

Yes, they are pretty mainstream. We managed to triple the revenue for some of our domains for item 2 that I spoke about in my blog and item 1 we are still quantifying the affects but it's looking to be about 20%.

You can't click on your own ads. That's click fraud, man. lol, I can't believe that you even complaint about it! Looking to have your account suspended?

 

[Whizzbang]

Interesting - we actually have permission from the founders of Sedo to be able to click on 20 of our advertisements per month for testing purposes. Just for the record we do NOT do click fraud.

The next thing is that as part of our testing regime we clicked on the advertisement from multiple IP addresses (as we know the blocking rules) and none of these have been tracked. Uli the CTO of Sedo sent me through the web logs and both Sedo and ourselves are investigating the reason for this.

In addition, it still does not explain why there were NO advertisements displayed at all versus not recording the clicks etc. Please view my blog (whizzbangsblog.com) to see the screen captures of what we experienced.

Also, the second issue that I reported with using an iFrame for calling the Sedo page has not been discussed. When we took the iFrame away we had an immediate jump in CTR of 5% or so which suggests that users were not seeing the advertisements and when they did, they clicked.

 

[AlienGG]

Also, the second issue that I reported with using an iFrame for calling the Sedo page has not been discussed. When we took the iFrame away we had an immediate jump in CTR of 5% or so which suggests that users were not seeing the advertisements and when they did, they clicked.

I had the same issue before. My server log recorded 4500 IPs a day, but Sedo only recorded 20 for the same duration. I thought some browsers did not support iFrame, and some, due to high security setting, were not allowed to display iFrames. However, after I switched to Parked, I saw more uniques in a day for the domain in the iFrame than what's reported by the Sedo Tracker (a counter to log traffic for websites) outside of the iFrame.

 

[Whizzbang]

We did not experience additional views when we turned off the iFrame with Sedo, we only experienced an increase in the CTR. The views were still being counted but the pages were not displaying any advertisements....hence our test to find out what was going on.

The reason that you saw an increase in the number of uniques at Parked is most likely due to their definition of a unique. Sedo defines a unique as a unique IP address every 18 hours while DS says every 24 hours. I'm not sure what Parked defines a unique. At parklogic (we're parking company agnostic) we're thinking of publishing an index of what is sent to a parking company and what they report....ranking the parking companies accordingly. I think that this would bring a level of transparency to the whole process.

 

[SedoCoUk]

Whizzbug,

We are evaluating the webserver log files for your first issue and are currently researching the second issue you mention regarding the iFrame.

Let me emphasize that we do not block any mainstream browsers.

We appreciate the constant feedback we receive from all of our customers as well as any potential bugs that we’ve been made aware of.

We are happy to work with you to resolve any issues and ultimately improve our parking program for everyone.

Thanks!

Monica

 

[Whizzbang]

Monica,
I realise that it's in no ones interest to block mainstream browsers but our data and your data is suggesting that there is definately an issue here that may have slipped into the Sedo system.

 

[netfounder]

as I know ,any parking provide company owned block system,
It is public secrets
dont care this issue.for protects himself,protects advertisers
if return 100% earning,every one going to die

 

[Whizzbang]

Hi all
I just posted an update in my blog on the Sedo Link Tracking issue - I hope that it helps. http://whizzbang.typepad.com/whizzbangs_blog/2007/05/sedo_link_track.html