Is Header Bidding really a Win for Publishers?



Time to start the journey I referenced in the blog post and discuss what winning is for publishers. This is not only the first topic (I’ll be posting other questions) but also just the first part of this thread as I think for something like header bidding there are many aspects to explore. I don’t pretend to know all the answers and I’m happy to be schooled if I have this all wrong. I just want to hear some new perspectives.

Header bidding has been a hit for publishers. I don’t know that I can recall someone every saying that they yielded less revenue after successfully implementing header bidding (note the highlight on successfully). Competition drives up prices and for publishers that’s not a good thing, it’s a great thing.

So yeah, header bidding has been a win and would love to hear from anyone that disagrees on that point alone. Anyone think it’s flat out a bad thing?

But is the win short lived? Some of the bump in revenue from header bidding was gaming SSPs to fight amongst themselves. At some point, some of those SSPs are simply going to go away and there will be less competition. If/when this happens, does revenue go back closer to original levels before header bidding or will CPMs stay up?

In other words, has the move to header bidding resulted in longer term higher CPMs for publishers or are these gains short lived?


Well on your last point, the future of tech is always difficult to predict. What I love about it is that you wake up in the morning, and who the fuck knows what is going to have happened?

That being said, I truly feel like header bidding and all the stuff happening in Ad Tech is really great for publishers right now. I got my career started in this space as a direct sales guy for a niche site, and we crushed it. I have spent the last 13 years of my life closing direct deals, and funding wonderful media sites in the process. That feeling of being the guy that enables the content crew to do what they do is extremely rewarding to me.

The problem is, more and more clients are giving the excuse of “Oh well sorry bud, I put half my budget with google this year.”. In the past as a direct guy, you’d have looked at this like a disaster… but with the power of header bidding - where you can truly expose your Tier 1 inventory in a fair price auction - those spends are coming back to you. Sure, Google or whoever takes a cut… but to be honest once the tech matures more… its probably a more efficient system than me calling the marketing guy and cutting a deal with a silly piece of paper sent through a fax machine.

I do think you’re right that there is some parts of the system which are just giving a temporary boost, and there’s going to be a ton of changes which happen as this moves forwards…

But the bottom line is that with (properly) implemented Header Bidding, you actually can expose your best inventory for a fair price. On the publisher’s side, it changes the game a bit where you want to focus on aggregating a valuable audience, but that is a strategical re-alignment vs. a complete destruction of the business.

I also want to say that what I think is amazing about this is that from the people I talk to on the buy side of programmatic - they want quality inventory. They really want ads to be displayed around good content, and they’re willing to pay premium CPMs for it. They don’t want to be in the arbitrage or total click bait game. They want the good stuff.

My dream for the world is that this header bidding revolution is the start of great publishers actually making the money they need to fund quality content. There’s a huge problem with it admittedly where its also what has funded fake news/content farms/everything bad with social media… but I don’t believe that is the future.

TL;DR - Header bidding rules.


What a great answer to the question and I think I agree with everything you said down the line. You hit on a few things I’ll want to circle back to on other threads, especially the funding of great quality content.

So as someone with direct sales experience, it sounds like you’re all in on programmatic. I’m curious if you feel like you’re losing something in the process. Yup, I agree the fax machines should be put out to pasture, but do you really want an intermediary between you and your clients? Don’t you give up a lot by making all your inventory visible to all of these buyers?



I’d completely agree with the “header bidding rules” evaluation. We implemented HB about a year ago. We took the plunge and went 100% header bidding, no more waterfall… the result was a 75% advertising revenue increase in less than a year. So at Ancient History Encyclopedia, we’re all in for header bidding.

The benefits:

  1. Ad networks / SSPs don’t always all have great demand ready for your site. Sometimes “A” has more, sometimes “B” has more demand, depending on the seasons or which advertisers they happen to have. By plugging in a variety of demand sources, they balance each other automatically, and you no longer have to adjust your waterfall on a regular basis.
  2. It’s really easy to do multi-size bidding, meaning that a single ad unit can receive bids for different ad sizes. For example, a 300x600 ad unit can also display 300x250 and 160x600 ads. Win.
  3. Testing out new demand sources / networks / SSPs no longer has the risk of losing you revenue. We used to try out new partners in our waterfall, only to find out they were under-performing, which had cost us some money. If you add a new demand partner and they don’t deliver, they simply don’t make money. You don’t lose out any revenue because the other demand partners will have filled those impressions.
  4. More revenue! That’s the bottom line every publisher wants to hear, and header bidding does deliver. Once you hit five or more demand sources plugged into your header bidding, you’ll see a significant uplift.

The downsides:

  1. Ad loading delay. Depending on your timeout, it can take longer for your ads to load. You need to think differently about ad placement: If you have an ad at the very top of the page, it’s possible that the users will have scrolled past it by the time it loads, and CPMs will suffer. We’ve found that sometimes ads just a bit below the fold do really well.
  2. Mobile page load times. Header bidding adds a lot of requests to a page load, which on mobile can take a toll. Ideally, you implement it in such a way that you send fewer requests for mobile.

For the future, I wan curious to see how the different ecosystems of header bidding evolve. There’s Prebid.js, and also Pubfood, and a few more proprietary ones. Then there’s server-to-server growing rapidly. It’s the wild west once more, but things will standardize over the coming years.


@RobBeeler - The part about direct sales vs. programmatic is a deep philosophical conversation… as what you stated is very true. Having people in between your sales team and the clients is an enormous and very scary risk. Its driven CPMs down in Direct deals, it has turned clients onto the fact that they can reach their audience in other places and you lose that strangle hold you used to have on reaching your audience.

That being said, its important with tech to run with trends. Header bidding has actually brought CPMs far enough up that they are starting to make real money. You also get a slice of the huge campaigns from major corporate advertisers, which in the past was only reserved for publishers with enormous scope. You can make real money there if you aren’t saving all of your best inventory for only direct clients.

I think a lot of digital publishers are worried about this thing, but its where the advertisers want to go. If you embrace it, I can absolutely envision a future where direct teams only sell super premium, custom campaigns. The regular bulk of your advertising will be traded. Sales teams might be dramatically reduced as a side effect, but that is just the ever-more automated world we live in.


Great stuff @Mr.Bishop and @jvdc! I’d like to keep this thread open and to hear from others. I may also have some more questions on this front as well.

Is Viewability a win for Publishers?

I would think that a look back at the history of AdSense might give an idea of what the future path of header bidding could look like.

I’ve been running AdSense for the past 13 years. Many pubs saw great eCPM rate in in the mid 2000’s. Peaks and valleys similar to the broader economy and then a steady trend lower over the past 5 years. I’ve recently started to see an increase – that corresponds with my implementation of header bidding.

I suspect that there will continue to be an uptick in eCPM rates, fill rate, etc until header bidding become more mainstream. At that time there will be a consolidation down to a few key players. Rates will then stabilize and move in correlation with the broader economy.

In the long term I think that any move that puts upward pressure on ad inventory values is a good thing for publishers and will benefit those that take advantage of the opportunities.


Thanks @Chrispcri. You hit on an interesting point when you talk about consolidation down to a key players. I think there are some pubs that would miss playing around with 14 header bidding partners. :slight_smile: My first thought about fewer players was that this would cause CPMs to go down since there would be less competition, but realized that if all the buyers worked with few players, the competition would be happening in their systems before they bid (I hope that makes sense). What I do think is a win is regardless with what happens with s2s, consolidation, etc. if you find a partner with unique demand, you have a method to plug it in and test it out.


A new angle on this discussion. Header Bidding has been a win for publishers. Does a move to server to server make it “less” of a win? Does anyone feel that in that move, publishers will be handing the keys over and lose something in the exchange? I’ve heard a few people rail against S2S, but not sure if that’s a small group that feels that way. Maybe some of you see S2S as an even bigger win. What do you think?