Cpm

[ArtColombia]

Assume an adgroup working CPM on a number of different sites.
Assume that the ad quality will not improve overnight. :)

How many prints/impressions would be required on that site
before you would throw in the towel and cut that site from
the campaign?

[Biff_Tiberius_Farnsworth]

I calculate it using ROI. If I get one conversion out of 100 clicks and it pays me $10 then I usually wait to see if $20 worth of impressions get me the conversions (sometimes I don't get reports on conversions for 24 hours) So I would shut it down at $18 because it always takes some time to shut down and then when I finally get my reports for conversions, I decide whether or not to turn it back on. Worst case, I lose $20 but I learn what doesn't work and move on to my next test.

~B

[ArtColombia]

Thanks Biff:

Good piece of advice, and probably a better focus than thinking just CPM alone.

Let me see: if 1000 prints costs $1, then you´ll wait to get 20,000 prints.
On that basis you would hope to get 100 clicks and 1 conversion as a minimum.

100/20,000 x100% = 0.5% CTR which I believe is reasonable for many campaigns,
though there have been some products recently presented on the forum claiming
up to 50% CTR on CPM campaigns (wow).

Of course if the CPM is 25c minimum then more prints could occur without affecting
ROI of the process assuming just one sale is made.

Time for me to reconsider my strategies! :)

[Biff_Tiberius_Farnsworth]

I was making a general example but in this one, I should have been more clear. If one conversion gets me $10 and I spend $20, I am hoping for 5 or 6 conversions from approx 500 to 600 clicks from those 20,000 impressions.

Then I would have an ad pumping out $50 or $60 for $20 being put into it. Biff like.

:)

My experience has often been around $2 per thousand impressions (aka prints).

~B

[ArtColombia]

Thansk for clearing up that point...sure a 2.5% CTR is much better, especially if it brings in $50 profit !!!:)