Sunday, August 15, 2010

CMF ADS: Infected by a superbug!

CMF Ads - Own House Ad
One month ago i started again to buy for several dollars campaign ads in CMF Ads Network instead of buying one of their low-cost Network ads (which by the way exploit just the CMF ADS publishers!).
As there are not many campaign ads running anymore in CMF Ads because they missed out that low-cost Network ads will harm their monetary system and as we can see leads to a severe recession with deflationary tendencies in their CMF ADS economy (a blog post about the devastating economical situation in the CMF ADS Network is prepared for Monday. There i will bring some light into their newly announced changes and "improvements").

When Network Ads (NA) got introduced, they've been running equally side by side with Campaign ads (CA).
This meant when you had SINGLE Ad widget and 2 CA, each CA got 50% of the available page views.

From end of January 2010 on around 8 - 12 NA have been running additionally each month which reduced the visibility for CA, especially on single ad widgets.
ONE CA would get now only -> 100% page views/(10-14 Ads) = 7 - 10 % of the available page views.
For a blog with 10.000 page views, your ad views would have been reduced from 5.000 to just 700 - 1.000 ad views. Most likely you paid up to 1 $ for that ad spot! Now the return in ad views would have been just like a low traffic blog on which you could advertise for 0.25 $ or less!

One week after the introduction of their Network Ads CMF ADS announced that the equal sharing will be changed and NA will show up only 30% of the time when they are running. This should mean CA could get 70% of the available page views - theoretically!

In the last weeks i had some extreme examples in which this was and still is completely NOT THE CASE. There have been some high-traffic blogs which get several thousand page views and my Campaign ad should get at least 1.000 page views after 10 days but instead of getting 100 ad views/day, the ad showed up just 35 - 45 times daily (in other cases up to 75% less views).

The behavior of the Campaign ad is just like before their announced change to a 30% reservation for Network ads. It had to share the page views EQUALLY with the Network Ads and didn't get 70% of the available page views!

Taking as example Ben Barden's Blog (CMF Ads Admin). This blog gets roughly 1.400 page views/month.
CMF Ads Page view Statistics
30% would be around 400 page views reserved for all running Network Ads, so a Campaign Ad should be exposed 1.000 times/month.

CMF Ads Campaign Ad Statistics
In my case i got just 170 ad views within 20 days instead of 500 ad views (excluding the surge in the first 5 days)!

In an email conversation in which i told Ben about this matter with the two "high-traffic" blogs, he replied me as if there is NO WEIGHTING and that the page views i received are the correct numbers even when Network ads should just get 30% of the available page views.
For a better view of the e-mail click on the screenshot!
CMF ADS email about Campaign Ads
Hopefully this superbug in their code could be changed to an algorithm in which Campaign ads are favored with 70% as they give paying advertisers a bigger portion of exposure when you want to buy an ad individually and publishers can earn again more money because they deliver the ad views an advertiser deserves!

I'm sorry for the members which ads got cancelled in the last days and weeks as their CPM was -due to this superbug - much higher than the 10 cents which i'm usually calculating to pay for my ad views.


  1. Hi Martin,

    To be honest, I am a little surprised by this post because my email should not be presenting you with any new information, I was simply answering your questions.

    Here is the CMF Ads blog post where we originally announced the weighting:

    In that post, I state the following:

    "Every ad slot on your widget will now have a 30% chance of displaying a network ad, and a 70% chance of displaying a non-network ad. This basically means that the non-network ads placed on your site will take precedence over the network ads.

    However, it is very important to note that this weighting only applies to network ads that you approve from this moment on. Any ads that are already running on your widget will not be weighted in this way."

    30% weighting has never ever been an absolute 100% guarantee of any NUMBER of pageviews. My blog is a good example because if you look at the graph, you will see a few spikes in traffic. Blog traffic changes all the time - 70% of traffic on one day is not going to be the same as 70% of traffic on the next day. Not only that, but as I said in the email, the number of ads is likely to change from the time you place your ad to the time it ends.

    Blog traffic changes and ads start and end all the time - these are very important points, but I couldn't find any mention of these in your post. I'm not sure why...


  2. I know the content of the CMF Blog post but if my ad gets only a 15% chance instead of 70% to be displayed, there is somewhere a superbug in the system because shouldn't be out of 10 page views 7 display a Campaign ad, when i'm the only CA?

    An algorithm usually doesn't care at all about different daily traffic, it just cares about the 30%/70% each time an ad has to be displayed. If there are 100 ad views to display, there should be roughly 70 times the CA and 30 times ONE of the NA.

    When NA's should be displayed only 30% of the time, then it doesn't matter if there is 1 or 10 of them running on a widget throughout the month. With 10 NA's each of them should get 3% of the ad views and 70% should be given to the Campaign Ads. With 1 NA this will get 30% of the ad views and 70% will be going again to the Campaign Ads.

  3. You may be right in saying that the number of ads running shouldn't affect the weighting, however, varying stats make a huge difference.

    What you see from the graph on my blog is that it received 1,377 pageviews for the last 30 days. (at the time of the screenshot)

    This does not mean we "reserve" a percentage of 1,377 pageviews for network ads for the next 30 days. If we did that it would only work if my blog received 1,377 pageviews for the next 30 days. If my blog gets twice as many pageviews in the next 30 days, we'll have reserved pageviews based on the wrong values.

    We cannot know a blog's stats ahead of time. All that happens is that every time an ad is selected for display, there is a 30% chance of a network ad displaying. Again, this isn't new information and we did say that in the blog post.

    Perhaps you're hoping that we can guarantee X pageviews for your ads. The only way to do that would be to sell pageviews. As we've said, we are indeed looking to add that to our options in the near future, though for the moment it will not replace our existing options.

  4. The problem is i didn't get 500 ad views on your blog just 170 even when there would have been 2 CA's, i still should have received that amount.

    You got from Day 10 until Day 30 on which my ad was running around 850 page views, so my ad was displayed just 20% of the time while Network ads have been displayed 80% of the time??

  5. You post is very informative. My blog is young and has about 2,600 page views this month. So I'm one of those publishers who is using campaign ads, rather than network ads. I look forward to reading your future posts on this topic.

  6. Hi Martin,

    Actually, if there is a problem here, I think you would have received more pageviews than the stats are showing.

    I am just about to push some changes out that may affect advert statistics over the next few weeks. Blog statistics should stay the same.

    It would be very useful if you could test this by purchasing a new campaign ad on my blog. I will be happy to give you the funds to do this, so you will get a free ad no matter what. If you're interested, let me know via email. Please don't buy the ad until you see a post on the CMF Ads Blog announcing the change to advert stats or you won't see much of a change.

    One thing to note, this change may only make a noticeable difference to the larger blogs in the network. With that in mind, if you do wish to do some testing for us, it might be better to buy an ad on my Top Ten Blog Tips blog as that has over 5,000 pageviews.


  7. I left CMF simply because I think that the admins are a bunch of douche bags. Really. I'm better off with just Entrecard and Adgitize.


Blog Widget by LinkWithin