Jump to content
Sign in to follow this  
WTravisH

Please Help: Openx 2.8.7 Consistently Over-Delivering Ads

Recommended Posts

We're running OpenX 2.8.7 and have recently started using "Contract" campaigns. We set the start and end dates, set the number of impressions, then let the Maintenance Priority Engine manage the delivery. All campaigns are set to Priority=5, so there shouldn't be any preferential treatment for one over another.

 

We have many different domains, which we sell in all kinds of different combinations, so we use targeting conditions on the banners to ensure that each campaign delivers to the correct domains only. Some campaigns deliver to only one domain, others to 30+ domains. We expected the MPE to find the correct compensated priorities to deliver everything on schedule.

 

After about 2 months of running contract campaigns, we're noticing that nearly every campaign is delivering well ahead of schedule. Most campaigns are year-long, and we're seeing up to 40% of the impressions served in the first two months. Our worst case campaigns are delivering about 250% or more of what they should be, but almost every campaign is at least 125% of where it should be.

 

At first, I thought the issue may have been an issue with the ad-server warming up and finding the right priority compensations to run correctly. But looking at campaigns delivery per day, the problem doesn't seem to be getting any better over time. In some cases, it's getting worse. At this rate, many of our advertisers will have served the majority their year-long campaigns in just a few months.

 

What could be causing this problem? Where might we look for a solution? Please help.

Share this post


Link to post
Share on other sites

Resolved!

 

I believe we have found a solution. It seems that the problem was with our cron job for the maintenance script. Our operation intervals were configured in OpenX as 60 minutes, but the cron job was running every 20 minutes. In the worst case scenarios, this caused a ~300% over-delivery. In the best case scenarios, there were too many banners running for everyone to get 300%, so everyone over-delivered as much as they could.

 

Correcting the cron job to run every 60 minutes seems to have fixed the problem, though we only have one day of data. I'll post further updates if we find otherwise.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  



×
×
  • Create New...