Advertisers with limited stores of advertising are naturally interested in getting every cent of out their material. That’s why so many new technologies and advertising techniques have continued to roll out and evolve over the years.
One of the most popular strategies is defined as waterfalling. If you’re unfamiliar with this term, then you could be missing out on an opportunity that could significant boost your overall return.
The idea behind waterfalling is simple enough. In the advertising arena, you can think of impressions as a form of currency: Just as a frugal budgeter labors to make the most of every dollar spent, publishers want to get the most value out of every impression they serve.
Waterfalling is an algorithmic tactic aimed at earning as much value as possible for every impression. Through waterfalling (also known as daisy chaining), a publisher initially works with the network of record which offers the highest possible rate.
If that network is not available, the system moves on to the next-highest-valuable publisher, and thus it goes until all possible resources are exhausted.
Whilst there are many advantages to waterfalling, like anything in life, these advantages are balanced by cons:
- It’s highly efficient. The entire premise of waterfalling is to monetize every possible impression—and get the most value out of each of them as well.
- It’s programmatic. Waterfalling is a programmatic implementation, which means that everything happens automatically, behind the scenes. As you’ll see, programmatic implementation poses some key challenges, but once it’s initiated, it’s a hands-off technique.
- It can work with SSPs. Many publishers have started to move on from traditional ad networks, and are favoring supply-side platforms (SSPs) like Google AdX instead. In fact, the transition is relatively simple: Rather than moving through various networks, you’ll waterfall down through SSPs, and work with alternative providers to monetize every impression.
- It’s not an easy system to develop. Getting started with waterfalling isn’t as simple as flipping a switch. Though you can investigate to learn what other people have done in the past, waterfalling demands some unique development—and that inevitably requires time and money. Even then, there’s no guarantee you’ll come up with a perfect solution, because there are key programming challenges you’ll need to overcome.
- Inefficient waterfalling can cost you money. You can’t improve your rates of monetization with just any waterfalling solution. If you end up adopting a system that doesn’t work efficiently, it could cost you additional, unnecessary funds. You could miss out on opportunities in your scramble to find the best, or you could end up erroneously serving less valuable impressions.
The biggest challenge with waterfalling is managing to implement it in an effective way. So, why are there so many issues complicating this process?
- Programming is hard. First, if you’ve never tried your hand at developing ad software, it’s difficult. That’s why developers demand high salaries, and why it often takes months, even, years to develop a solution for a hard technical problem.
- Redundancy is a potential issue. Developers also need to bear in mind the fact that multiple buyers will customarily bid on multiple types of inventory and on multiple platforms at the same time. If you aren’t careful, this could lead to some major redundancy issues.
- Testing and retesting. Thanks to the inherent challenges of the process, there’s no guarantee that even a logical, on-paper solution will work in a practical environment. If you want to make sure your solution works, you’ll need to insert it into a live environment and test it multiple times over, which takes time and costs money … not to mention slowing your potential rate of growth.
The complicated process of waterfalling is likely to become far more easy to implement over time and no doubt, eventually, developers will find more intuitive, effective ways to tie complex systems together. Then, there will be semi-standard protocols for implementation (if the SSPs don’t jump ahead and complicate things even further).
The quintessential situation would be for some forward-thinking momentum to sweep the industry and the world becomes a witness to SSPs learning to work together and following similar protocols. In this kind of world, all buyers would see and bid on the same massive inventory at the same time, and developers’ jobs would get a whole lot easier. This may be a pipe dream, but it still exists as a possible path, paving our way to a day we see simpler waterfalling development.
(As published on The Drum)