View Full Version : Default Ad Display Discrepancies?

Micha Dugga
08-25-2007, 09:04 AM
I'm new to the forums here, but have been reading as a guest for a couple of months. I've taken a lot of information from the forum here and applied it to our sites for generating revenue.

I thought I'd pose a question to you guys to see if you could offer some insight regarding something I'm observing between my Tier 1 and Tier 2 ad companies.

Before this gets too confusing, I'll try to simplify my question with a little bit of background.

On our company's flagship site (http://scaredmonkeys.com), We're running Tribal Fusion leaderboards and skyscrapers. On the same page, we're running Casale rectangles and popunders. 3 visible ads and 1 popunder from 2 ad companies.

The reason we're mixing tier 1 and tier 2 companies this way is because we just last week got signed up with Casale (so far, we LOVE them), but our representative there requested that we take it slow by only initially running a couple of spots on Casale before jumping in with both feet.

So, here's my question...

Where Casale is my Tier 1 (rectangles, specifically), I show the number of default ads that are running to be far more than what Tribal Fusion (rectangles Tier 2) reports having run. In Theory, if Casale says "we served 10,000 default ads", and Tribal Fusion is set up as my tier 2, then shouldn't Tribal Fusion's TOTAL ad count show 10,000 ads as well? It doesn't. It shows far, far less. In some cases, only 25% of what you'd expect.

I could write this off as a Tribal Fusion accounting problem, but it's happening in the other direction as well.

Where Tribal Fusion is my Tier 1 (leaderboards, skyscrapers), they might report serving 20,000 defaults. Then my Tier 2 (in this case, Casale) shows a number significantly smaller than what TF is reporting. Casale should say they attempted to serve 20,000 ads, but my reports show a much smaller figure.

Keep in mind, I'm looking at overall numbers in my tier 2. I'm not counting any tier 2 fill rates that might default to a tier 3 provider (which we're currently not even using).

So, to make a long story short, why is it that two very strong, reputable companies like Tribal Fusion and Casale seem to be letting so many ads slip thru the cracks?

I wonder if I'd be better off by pointing every single default campaign to a PHP script on my server so that I could track exact numbers from apache logs?

Any suggestions or insight would be greatly appreciated.


08-25-2007, 10:25 AM
There are many issues here that cause this.

1. load times, a user could navigate away from the page before the second tier ad has time to load, but the first tier spot loaded, so one place counts it but the other doesn't.

2. The same crappy traffic tier 1 doesn't want to display an ad for, tier 2 doesn't want either, so they cycle that back again to the defaults.

3. Possible older/less technical browsers or crawlers that can load the JS for tier 1 fine, but the layered js of tier 1 redirecting to tier 2 messes them up.

In anycase, it is completely normal and happens with everyone, just the nature of things.

Micha Dugga
08-25-2007, 02:45 PM
Thanks for the great info, Chris.

So, since this seems to be the "nature of the beast", is there a general concensus as to what the best way to fill all inventory might be? From my math, it looks like a minimum of 20% of our ads are leaking thru the holes because of this.

I suppose I could set up a script on the server side that evenly switched between the two ad companies ads on the same page. One page load it'd load one ad company's ads, and on another load it would load the other company's ads. In theory, if both of these had at least a 50% fill rate at "full throttle", then alternating them randomly would fill 100% of our inventory. Another benefit would be that we'd get top CPM from both networks (assuming they fill highest paying ads first).

It's a bit of a math game, isn't it? It would cut our overall traffic to both tier-1 providers in half. I wonder if Casale or TF would get ticked off about us trying that?

08-25-2007, 08:25 PM
I just checked out my stats and for leaderboards yesterday, which I default TF > Valueclickmedia. TF shows 52k defaults and VCM shows 47k total leaderboards (filled ads + defaults). This seems normal to me at about ~10% discrepancy due to the reasons Chris stated above.

If you are seeing a drop of 80%+ then you are most likely only looking at filled ads, which can be quite normal since most the traffic passed on by your first tier is usually poor/intl.
Creampie porn (http://www.****tube.com/categories/16/creampie/videos/1)

08-26-2007, 04:19 PM
The problem with that approach Micha is that the ad networks do not work like that. You can't think that if they're only filling 50% of your traffic that you'll only send them 50% and they'll fill 100% of what you send them. In reality they'll keep filling just 50% of what you send (so 25% total).

There isn't really a good fix for the problem other than to keep trying to get better ad rates & fill percentages.