Ahead of our Nov 12 webinar, How to Succeed with Auction in the Header Bidding, I’ve put together this post to explain why header bidding is gaining a lot of interest among publishers. Hopefully it’ll whet your appetite until Tony Casson, one of our product managers, and our guest speaker Gabriele DeWitt, VP of Monetization at Curse, give you the full scoop on the webinar.
If you’ve not heard of it before, header bidding is something you should keep an eye on. It’s a new ad optimization technique that has the potential to turbo most your ad revenue, but that at the moment is only being used by larger publishers.
Header Bidding vs. Waterfall or Daisy Chain Optimization
In a nutshell, header bidding (which is also known as auction in the header) raises ad revenue by re-inventing how ad networks and exchanges bid for your inventory. It does this by placing code on the header portion of your site, which then requests and receives bid prices from all your ad partners at the same time every time someone visits your site.
To see how this benefits you, we need to consider how publishers traditionally optimize their ad operations. Typically, publishers prioritize their ad partners based on what they think each partner will pay for an impression. To do this, publishers create multiple line items in their ad server with each one representing a different floor price and partner. When the publisher’s site receives a visitor, the ad server offers the potential ad impression to each line item one at a time, starting with the line item containing the highest floor price. If a partner bids more than the floor price set against them, the ad server awards the impression to that bidder. This type of ad optimization technique is known as a waterfall or daisy chain set up.
So what’s wrong with this approach? Well, a (more than) couple of things:
- It’s based on estimates on which ad partners will pay the most for any impression. This is highly inaccurate as ad partners constantly change their buying strategy, which means publishers often temporarily prioritize the wrong partner.
- It leaves money on the table. Waterfall optimization doesn’t ask every partner in your ad stack what they’ll bid for the impression; it stops at the first partner who bids more than their floor price. Because of this, publishers may award the impression to one partner, when another bidder would have paid more.
- It can cause discrepancies. Waterfall optimization is a complex process that requires multiple calls between the publisher and its partners. This happens thousands of times, so discrepancies often occur between the publisher and advertiser on the number of ads that have been served. If the difference is big enough, then manual reconciliation, which takes time and money, needs to be performed.
- Delayed reporting. The publisher’s ad server only records the floor price of the winning bid, not the partner’s actual bid price, so the publisher doesn’t accurately know their actual revenue until the check comes in, which can take weeks.
- There’s a lot of manual work involved. All of the above means that a lot of manual work is needed to create and maintain the best possible waterfall set-up by creating complicated targeting rules, pass-backs, and line items, and even then you’re not getting the best results.
Header Bidding Benefits
By requesting bids from all partners simultaneously, header bidding solves many of these issues:
- Publishers no longer have to prioritize partners based on estimates. Bidding is based on real prices, so no need for guesswork.
- They maximize revenue for every impression. You now receive prices from EVERY partner at once, so the impression is always awarded to the highest bidder.
- Lower discrepancies. Less back and forth between the publisher and its advertisers means less likelihood for erroneous counting.
- Accurate and timely recording. If your ad server is set up in the right way, header bidding enables it to record the actual price paid for every impression.
- Less on-going manual work. header bidding requires some complex work up front, but once it’s done, the publisher doesn’t have to constantly tweak their ad stack to get the best results.
The downsides? Well, code has to be placed in the header portion of your site, which could impact how long it takes your page to load. This can negatively impact your visitor experience, so publishers need to engineer their site to guard against this and ensure their partner has the right technology and services to help them. In addition, because the technology needed for header bidding is new, it requires a fair bit of technical resource to implement. However, in the next few months, you’ll probably see this requirement reduced as vendors make the technology easier to implement.
Hopefully that should give you an idea of header bidding and its potential. If you’re interested in learning more, then make sure to book your place on the Nov 12 Auction in the Header Webinar. In it, Tony will go in-depth into header bidding, while Gabriel from Curse will share their experience in implementing it and the results they’ve seen. The webinar is definitely one you shouldn’t miss if you’re considering header bidding now or in the future.