Step by step process to adjust for ad network discrepancy in Prebid

Step by step process to adjust for ad network discrepancy in Prebid

If you are a publisher running Prebid, you need to look out for the discrepancy between your ad server and the ad network. These discrepancies can happen for a lot of reasons and will most likely lead to ad revenue loss. Regardless of the reason, publishers should address discrepancies as these are indicators that something might be wrong with the setup.  While the numbers may not 100% match most of the time, it is important to correct and adjust the discrepancy within the acceptable range (+/- 10%).

Start with a Discrepancy Report

Prepare your discrepancy report for you to monitor daily. The discrepancy report will help publishers compare the revenue and impressions from both parties and identify which bidders are showing huge discrepancies. Your discrepancy report should include:

  • The ad server revenue and impressions. In this example, we will be using Google Ad Manager (GAM). You can break down the bidders by adding your bidder’s key value (e.g., hb_bidder) in the report. Save the report so you can run it instantly next time instead of creating from the report builder all over again.
  • Get the revenue and impressions from your bidder’s dashboard. Some ad networks allow scheduled reports via email, too. Getting the report through email is easier instead of logging in to the bidders’ dashboards daily.
  • Take note that your currencies should match. Most ad networks will report in USD. If your GAM and ad network currencies are not the same, add another column for currency conversion.
  • Some bidders will show the gross amount in their dashboard. Make sure to adjust your discrepancy report also to show the net amount. Add a column to calculate the gross to net adjustment automatically.
  • Insert a set of columns to calculate the revenue and impression difference and discrepancy. To find the discrepancy percentage, get the difference between your ad network and GAM revenue, then divide the difference by your GAM revenue.
  • Finally, include a column for the discrepancy rate. The discrepancy rate will be used when calculating the adjusted bid for bid scaling. It is the sum or difference after we take the discrepancy % from 100. Simply put, you can also calculate it by using this formula:

prebid discrepancy 1

Discrepancy Rate = Ad Network Revenue ÷ GAM Revenue

Here’s an example of how the report and calculations are organized:

prebid discrepancy 2

Impression Discrepancy

The discrepancy report should now provide you with insights into where your discrepancy is coming from. Ideally, the impression discrepancy should be fixed first since it is most likely the reason why there is revenue discrepancy with your bidders. Impression discrepancy often serves as an indicator whenever there are problems with the configuration like incorrect placement/bidder IDs, line item errors, or even errors from the ad network’s backend. Focus on fixing the discrepancies on bidders with a high share of voice. If you’re seeing a 25% discrepancy on a bidder whose total impression in 1 day is 200, the data is too small to conclude from, and it is statistically insignificant.

Revenue Discrepancy

If you’re still getting over 10% revenue discrepancy even after fixing the impression discrepancy, the next steps are to:

  1. Identify if the discrepant bidder is sending gross or net bids. Similar to the revenue reporting discussed earlier, bids may be coming in where the ad network has not adjusted their revenue share from the bid and is sending gross bids. These bids need to be scaled down. Prebid’s guide about adjusting the net/gross bids can be found here.
  2. Apply bid scaling. Bid scaling enables a fair auction by adjusting the bids based on the bidder’s discrepancy. In the image below, let’s use January 1st as an example. January 1st has an 18% revenue discrepancy, so their bids also needed to be scaled down by 18%. If your existing adjusted bid is $1, multiple it by your discrepancy rate. So your new adjusted bid will be, 1 x 0.82 = 0.82. Use Prebid’s bidCpmAdjustment to adjust the bids in Prebid.

prebid discrepancy 3

Conclusion

Monitoring, fixing, and adjusting discrepancies is a tedious process. Aside from manually collecting the data, this also requires code adjustments and, if not done right, might result in problems with the ad serving. PubGuru offers a discrepancy report for all ad networks connected to our publisher’s account. Taking away the manual process of running Google Ad Manager reports and copy-pasting your ad network’s data.

discrepancy table

Our ad optimization team will monitor the discrepancy for our publishers.  Our setup includes scaling the revenue share if needed as well as optimization of the bids through bid scaling. Our technology eliminates all the hassle and manual work so publishers can focus on high-level optimization and strategies and creating engaging content.

Are you ready to finally get discrepancies under control? Sign up to MonetizeMore today!

Kean Graham

CEO and Founder at MonetizeMore

Kean is the resident expert in Ad Optimization covering areas like AdSense Optimization, DFP Management, and third-party ad network partnerships. Kean believes in the supremacy of direct publisher deals and holistic optimization as keys to effective and consistent ad revenue increases.

Get our latest ad optimization tips delivered to your inbox

Fill out my online form.

Submit a Comment

Your email address will not be published. Required fields are marked *