How much traffic should I have to consider HB setup?


And what’s sovrn requirement for that?


@polapi header bidding will help to optimize yield for any amount of traffic, and we can provide header bidding ad tags for any account that has an approved site.

500,000 requests per day is the maximum amount of traffic that is required in order to receive a dedicated support engineer for an implementation, but it can vary greatly depending on your situation. Talking to your Publisher Advocate or Account Manager from Sovrn is the best way to find out if this is an option for you.

Thanks for reaching out, and feel free to post any other questions or concerns!


Something I’ll chime in here on is that as @Ted_Rand said - under 500k and you don’t get a dedicated engineer. We’ve worked with Ted on some implementations, and he’s absolutely fantastic. However, without him - installin a header setup yourself definitely takes work. Capable developer will have no problem, but it will take time. So if you are a coder, or you have one on staff then its something you can consider. If you’re outsourcing development he costs are going to add up.


In my case my AM refused multiple times to enable HB for my page which has about 100k page views daily claiming I am too small and that’s sovrn policy. Anyway, thanks for letting me know that if I were more persistent I might go for such setup.


Hi @polapi!

Due to the high demand for header bidding, we are limited in terms of our ability to provide one on one engineering support for every publisher. As a result, our development team has worked tirelessly to build out our header suite products as well as user friendly, self-service implementation guides. Our goal is to provide this solution to a wider range of publishers. If you would like, feel free to reach out to us directly at with your sovrn username. We would love to continue your header bidding discussion and look at the different options available to you.


You don’t have to use Sovrn’s wrapper. If you have some development skills, and you have a DFP account, you can get set up through Prebid.js all on your own. It’s not very complicated. You can integrate that with Sovrn and many other ad networks. We increased our revenue by 75% when we switched 100% of our ad impressions to Prebid.js.

I would say that it’s worth it starting at 500,000 monthly visits (as per Google Analytics). Below that and the development effort won’t be worth it for you.


Here’s a question - Without using Header bidding what do you think is the next best strategy?

Personally, I’m still stuck on daisy chains with some mixed price priority if I can’t go to header bidding. I’m not sure I see a better way… though after learning more about the mechanics of the header bidding auction its really difficult to be re-organizing networks and feeling like I’m not wasting my life away.

Users like @polapi - what do they do instead of Header Bidding? Sure you can hard code adsense, but the return on that is garbage. At least in a daisy chain you can manipulate price floors…


Thanks, Jan!

I think you nailed it on the head.


I’d say: “Go with header bidding!” And if your traffic is too low, then work on improving your traffic numbers. :slight_smile:

Also, Sovrn’s OnScroll and VET tags are great additive revenue sources, too. They work with both header bidding and waterfall / daisy chains.


I would personally avoid daisy chains.

Here’s what I’ve done in DFP for non-header bidding networks:

  1. Enable them and all other networks for price priority
  2. Have all networks compete at a rounded increment eCPMs. (For US traffic I use 0.25 increments)
  3. Create AdSense passbacks for with custom tracking for each ad network and have them use those for passbacks.
  4. All ad placements compete against AdSense in DFP.

I use a spreadsheet set of formulas to determine the following:

  1. Network’s actually eCPM. (Network Revenue + AdSense Passback Revenue) / DFP reported impressions
  2. Floor eCPM set to AdSense Passback Revenue / Network Reported Impressions Passed.

What I’ve found over many years is that different networks “lose” impressions at varying rates and “lose” passback impressions at varying rates as well. This is where not running your own reports and just relying on network reports and daisy chains will kill your earnings.

Keep in mind that you may need to create different line items for different geo traffic if you get a lot of non-us traffic.