avatar image

In-app bidding has automated most waterfall optimization, yet developers still manage multiple hybrid waterfalls, each with dozens of manual instances. Naturally, this can be timely and overwhelming to maintain, keeping you from optimizing to perfection and focusing on other opportunities to boost revenue.   

Rather than analyzing each individual network and checking if instances are available at each price point, breaking down your waterfall into different CPM ranges allows you to visualize the waterfall and easily identify the gaps. 

Here are some tips on how to use CPM buckets to better optimize your waterfall’s performance.

What are CPM buckets?

CPM buckets show you exactly how much revenue and how many impressions you’re getting from each CPM price range, giving you a more granular idea of how different networks are competing in the waterfall. CPM buckets are a feature of real time pivot reports, available on ironSource LevelPlay.

Identifying and closing the gaps

Typically in a waterfall, you can only see each ad network’s average CPM. But this keeps you from seeing ad network distribution across all price points and understanding exactly where ad networks are bidding. Bottom line - you don’t know where in the waterfall you should add a new instance. 

By separating CPM into buckets, (for example, seeing all the ad networks generating a CPM of $10-$20) you understand exactly which networks are driving impressions and revenue and which CPMs aren’t being filled 

Now how do you do it? As a LevelPlay client, simply use ironSource’s real time pivot reports - choose the CPM bucket filter option and sort by “average bid price.” From here, you’ll see how your revenue spreads out among CPM ranges and you’ll start to notice gaps in your bar graph. Every gap in revenue - where revenue is much lower than the neighboring CPM group - indicates an opportunity to optimize your monetization strategy. The buckets can range from small increments like $1 to larger increments like $10, so it’s important to compare CPM buckets of the same incremental value.

Pro tip: To best set up your waterfall, create one tab with the general waterfall (filter app, OS, Ad unit, geo/geos from a specific group) and make sure to look at Revenue and eCPM in the “measures” dropdown. In the “show” section, choose CPM buckets and sort by average bid price. From here, you can mark down any gaps.

But where do these gaps come from? Gaps in revenue are often due to friction in the waterfall, like not enough instances, instances that aren’t working, or a waterfall setup mistake. But gaps can also be adjusted and fixed. 

Once you’ve found a gap, you can look at the CPM buckets around it to better understand the context. Let’s say you see a strong instance generating significant revenue in the CPM bucket right below it, in the $70-80 group. This instance from this specific ad network has a lot of potential, so it’s worth trying to push it to a higher CPM bucket. 

In fact, when you look at higher CPM buckets, you don’t see this ad network anywhere else in the waterfall - what a missed opportunity! Try adding another instance of this network higher up in the waterfall. If you’re profiting well with a $70-80 CPM, imagine how much more revenue you could bring at a $150 CPM. 

Pro tip: Focusing on higher areas in the waterfall makes a larger financial impact, leading to bigger increases in ARPDAU.

Let’s say you decide to add 5 instances of that network to higher CPM buckets. You can use LevelPlay’s quick A/B test to understand if this adjustment boosts your revenue - not just for this gap, but for any and all that you find. Simply compare your existing waterfall against the new waterfall with these 5 higher instances - then implement the one that drives the highest instances.

Božo Janković, Head of Ad Monetization at GameBiz Consulting, uses CPM buckets "to understand at which CPMs the bidding networks are filling. From there, I can pinpoint exactly where in the waterfall to add more traditional instances - which creates more competition, especially for the bidding networks, and creates an opportunity for revenue growth."

Finding new insights 

You can dig even deeper into your data by filtering by ad source. Before CPM buckets, you were limited to seeing an average eCPM for each bidding network. Maybe you knew that one ad source had an average CPM of $50, but the distribution of impression across the waterfall was a black box. Now, we know exactly which CPMs the bidders are filling. 

I find ironSource CPM buckets feature very insightful and and use it daily. It’s an easy way to identify opportunities to optimize the waterfall and earn even more revenue"

Božo Janković, Head of Ad Monetization at GameBiz Consulting

Understanding your CPM distribution empowers you to not only identify your revenue sources, but also to promote revenue growth. Armed with the knowledge of which buckets some of their stronger bidding networking are performing in, some publishers actively add instances from traditional networks above those ranges. This creates better competition and also helps drive up the bids from the bidders

There’s no need for deep analysis - once you see the gaps, you can quickly understand who’s performing in the lower and higher buckets, and see exactly what’s missing. This way, you won’t miss out on any lost revenue.

Learn more about CPM buckets, available exclusively to ironSource LevelPlay here

Let's put these tips to good use

Grow your app business with ironSource