How exactly does in-app bidding work? Why has it become such a big deal? Is it better than the “old methods” of ad monetization?
Table of Contents
I have the answers to all these questions and more.
In this guide, I will take a deep dive into this method of programmatic advertising. I’ll go over everything you need to know about in-app bidding – its benefits, best practices, how it compares to the waterfall model, the best solutions on the market, etc.
Let’s get started.
Programmatic Advertising and In-App Bidding
Over the last couple of years, the world of mobile app monetization and advertising has changed quite a bit.
One of the biggest tech accomplishments of the app&game industry is programmatic advertising.
Simply said, programmatic advertising is an automated way to sell and buy ad inventory.
This is actually an umbrella term.
It includes the two most popular methods for buying and selling ads – the waterfall model and in-app bidding.
How popular is programmatic advertising?
According to PubMatic, today, 66% of in-app budgets go into programmatic or open exchange, while only 34% go into direct buying.
What does this mean for app publishers? On top of other things – more opportunities to capture some of all that money going into in-app advertising.
What Is In-App Bidding?
Every good guide begins with a definition, so let’s start this one by defining in-app bidding.
In-app bidding is a programmatic advertising model that includes two sides: app publishers and advertisers.
In it, app publishers sell their ad inventory in a unified auction, and advertisers bid against each other in real-time.
From the publisher’s perspective, what makes this model special is that they can offer inventory to all demand sources at once.
From the advertisers’ perspective, they get a fair chance to compete on ad inventory at the same time. With some other methods, they would only gain access to a limited percentage of inventory.
The History of In-App Bidding
If you really want to understand in-app bidding, you need to understand its origins.
You’ve probably heard about the term “in-app header bidding”.
If this is the case, you’re probably wondering – what is the difference between this term and in-app bidding?
There is no difference.
It’s just a matter of wording and, well, history.
Header Bidding on Desktop
The phrase “in-app header bidding” comes from the initial idea behind in-app bidding.
The thing is, before this type of bidding came to the world of mobile, a similar model existed on the web.
On desktops, header bidding came to life in 2015. As soon as it appeared, it rocked the existing desktop advertising world.
The thing is, at the time, advertisers got tired of waterfall bidding and Google’s auction rules. For this reason, they started looking for alternatives.
However, header bidding turned out to be much more than just an alternative.
Here’s how it works.
In this model, the web publishers auction inventory in their web page headers in real-time.
For this purpose, they add a wrapper to their websites’ headers.
This allows advertisers to simultaneously compete for a place at a header of a web page. Whoever wins the auction, their code gets incorporated into the header.
As a result of this method, publishers started getting the highest possible price for each impression. In fact, according to Fyber, thanks to this model, web publishers have seen up to 40% higher eCPMs.
On the other hand, the benefit for advertisers was pretty obvious – they got to the top of the page.
In-App Header Bidding
While all this was happening on the web, the app world stuck with the “good old” waterfall bidding model.
Nevertheless, app publishers and advertisers also started craving something like header bidding.
The problem? There are no headers in mobile apps.
For this reason, it was more challenging to develop solutions for-in app header bidding. Luckily, in the last couple of years, fully functional in-app bidding solutions appeared.
Instead of relying on headers, in-app header bidding works with an SDK integration that generates ad requests for an app.
The ad requests participate in an auction in the cloud with server-side demand partners. Once there is a winning bid, the in-app bidding platform sends out a response to the client SDK.
As you can see, this process is everything but simple. No wonder it was so difficult to develop.
In-App Bidding vs. Waterfall Bidding
There is no talking about in-app bidding without talking about waterfall bidding.
We can look at it as a new vs. traditional relationship.
In this relationship, waterfall bidding is the older, traditional way, while in-app bidding is new and improved.
The waterfall model has been the standard of ad monetization for the past decade.
Essentially, waterfall bidding works just like water flowing down a cliff.
In the waterfall, the ad server calls ad networks in a predetermined order, one by one. This order is based on historical prices and performance. There is also a price floor – the minimum price that will be accepted for the placement.
Then, the requests for bids go down the “cliff” of bidders until one of them meets the price floor.
In this process, it’s not important how much the advertisers are willing to pay for an impression. Instead, historical data has an advantage. As a result, the mediation platform will give an advantage to the ad networks it has previously worked with.
Sounds kind of unfair?
That’s exactly what publishers and advertisers don’t like about it.
In the waterfall model, publishers are potentially leaving money on the table. The thing is, they are relying on the mercy of the first bidder in line to bid for ad space. Even if someone at the bottom of the list would bid more, the publisher never gets to find this out.
A lot of advertisers find this model discriminatory. The way the waterfall works, not everyone gets a fair shot to compete for ad space.
Improved (In-App Bidding)
In-app bidding was created to resolve the issues surrounding waterfall bidding.
This bidding model allows publishers to offer inventory to all potential bidders. This also includes the ones that wouldn’t rank high in the waterfall.
While the waterfall model prioritizes the first bidder, in-app bidding gives the advantage to the highest bidder.
Here, historical data is completely irrelevant.
In the in-app bidding model, it’s important what happens in real-time, not what happened before. To make this happen, it works on the concept of a unified auction.
In the unified auction, multiple demand sources gain access to the same ad inventory. They bid for it in real-time, and the higher bidder wins. This way, the prices can go up quite a bit, resulting in higher eCPMs for publishers.
A true unified auction should have these three characteristics:
- Being inclusive for all demand sources
- Simultaneously distributing ad requests
- Allowing everyone to bid in real-time
At heart, unified auctions aren’t much different from real-life auctions.
Imagine a bunch of millionaires auctioning for something in a crowded room. In this case, nobody cares if someone is a new, self-made millionaire or they come from an “old money” family.
Whoever bids most at the moment wins the auction, and that’s all the wisdom to it.
In-App Bidding + Waterfall Bidding
Despite their differences, there doesn’t have to be a versus in this relationship.
In-app bidding and waterfall bidding can work hand in hand. This mix is called a hybrid setup, and it’s prevalent among app and game publishers. According to Business of Apps, hybrid setups are dominating the market in 2021.
Now, let’s go over how hybrid setups work.
This simplified, 5-step process should give you a general idea:
- If an app requests an ad, the request first goes through the process of in-app bidding.
- The server determines the winner of the auction (the highest eCPM).
- The ad mediation network quickly runs a check of ad networks in the waterfall model to see if there is a better offer (higher eCPM).
- If there is no higher bid, the winner ad runs instantly.
- If the waterfall contains a higher bid, this ad will run first.
Sounds great, right?
This kind of solution allows publishers to get the best out of both worlds. With this setup, they should achieve the highest revenues regardless of where the winning ad comes from.
The only problem is, from the tech side, the hybrid setup is difficult to implement. In it, it’s most important to achieve that both of the systems fight for an impression simultaneously, which is challenging.
Since most of the solutions that support hybrid setups have appeared over the past year, they are still relatively new. For this reason, it’s still early to tell how well they work and how effective they are.
Benefits of In-App Bidding
In-app bidding brings great benefits for both parties in the ad exchange – publishers and advertisers. Even app users can feel its advantages.
First, let’s take at what publishers gain from in-app bidding.
Higher Demand (and Revenues)
In an in-app auction, multiple demand sources bid in open access, real-time auction. For this reason, publishers can expect more demand sources bidding for impressions.
In other words, more competition among advertisers.
This should also bring publishers more high-paying bidders.
As the demand and bid density increase, this should result in more revenue for the publisher. Looking at metrics, publishers should see increased eCPMs and ARPDAUs.
Easier to Maintain
Generally, waterfall bidding requires more maintenance than in-app bidding. Some publishers even manually optimize their waterfalls based on running reports. They analyze historical data and adjust how demand sources rank in the waterfall.
Unlike waterfall bidding, in-app bidding isn’t nearly as demanding. What matters here is real-time pricing, not historical data.
In this model, the buying/selling process is completely automated. This gives publishers more time to focus on other important things concerning their app.
Increased Inventory Value
In the in-app bidding model, premium inventory is sold early on to the highest bidders.
Consequently, this makes the remaining inventory more desirable. As a result, the prices of the unsold inventory should increase.
Increased Fill Rates
It’s important to fill up as much ad space as possible.
The most obvious way to increase fill rates is to use multiple ad networks. Whenever one ad network fails to fill in ad space, another one probably will.
However, in the waterfall model, there is always a chance that no ad network will meet the set price floor.
In this case, ad impressions are left unfulfilled.
With in-app bidding, the chances of this happening are much lower.
Since it involves a bigger competition, there is a better chance a publisher will have bids for each ad impression.
Better UX for Players
Another issue that comes with waterfalls is latency. And, of course, a poor user experience that comes along with it.
This can be a big turn-off for app users.
When users reach a video ad placement that doesn’t load quickly, they may get impatient and exit the app. If this happens, the ad request doesn’t get filled, and the publisher earns zero revenue.
In-app bidding helps publishers with this as well. The thing is, in comparison to the waterfall, there are fewer steps in the process of ad serving. Therefore, ads can load faster. With a decreased latency, app users can enjoy a seamless ad experience.
More curious about the positives for the advertisers? Coming right up.
Equal Access to Premium Inventory
Every advertiser dreams of the same things – access to quality inventory and quality users at scale.
These are also their biggest challenges.
In the waterfall model, some advertisers will always lack access to inventory and userbases they would like to see more of. Because of historical data, advertisers are treated unequally by default.
On the other hand, in-app bidding gives everyone an opportunity to reach the audience they want to reach.
In this model, all advertisers get a chance to respond with a bid for any ad request. As a result, they gain access to all of the publishers’ inventory. Hence, they can reach all of the users that consume it.
Of course, as long as they are willing to pay for it.
In-app advertisers want to have a clear understanding of their current and future possibilities. They want to be able to estimate, predict, and scale their user reach.
With waterfall bidding, this is quite difficult. This model is very variable, which makes it hard to make precise predictions.
In-app bidding, on the other hand, is much more transparent.
Since all advertisers get equal access to inventory, they also get higher visibility of available inventory and audiences. According to their campaign goals, they can then decide to invest more in something they believe will work for them.
In this case, their decisions aren’t random and based on guesswork.
Instead, they can make informed, data-driven decisions.