300% higher ARPU and 3x more installs - both results from game developers that used ironSource’s ROAS optimizer to scale their UA. That’s because this tool automates and simplifies bid optimization for advertisers so they can save time, bid as accurately as possible, drive more installs, and maximize return on ad spend (ROAS) - all while getting full visualization into performance.
The ROAS optimizer can benefit advertising campaigns for IAP and ad-based games alike, and there are a few tips and tricks that can help you scale and reach revenue goals more effectively.
1. Select the right optimizer for your game
There are three types of ROAS optimizers that can help you reach these goals: impression level revenue (ILR), in-app purchase (IAP), and combo (a mix of ILR and IAP).
Each uses different types of revenue data, needs different durations for optimization, and has unique defining features. Games with a short LTV, lower bids, and high IPM include genres like hyper-casual and arcade, which are also the games that usually rely heavily on ads for revenue. The ILR optimizer shares these defining features.
Meanwhile, the IAP optimizer shares its defining features - long LTV, high bids, and low IPM - with games that rely more on in-app purchases for revenue, like those in RPG and simulation genres.
To know which is the right choice for you, start by considering your revenue breakdown and the optimizer’s prerequisites. The ILR optimizer is best if 90% or more of your app’s revenue comes from ads. Note, though, that this optimizer does require you to be on LevelPlay mediation.
Choose the IAP optimizer if more than 90% of your ad revenue comes from in-app purchases. While the IAP optimizer doesn’t require you to be on LevelPlay mediation, it’s necessary to share post-install events (PIE).
If your app’s revenue doesn’t meet the 90% threshold for either IAPs or ILR, go for the combo optimizer. Since the combo tool combines the features of the two other optimizers, it needs both PIEs and information from LevelPlay mediation to have all of the relevant data and optimize your campaign to the fullest extent.
Neon Play shows the benefits of different types of optimizers with their game Idle Army - they switched from the IAP optimizer to Combo after two months to build upon their success. As a result of the switch and with ironSource’s guidance, their scale grew 350% and ARPU on D10 increased 300%. Mark Allen, Neon Play’s Director of Games, describes how ironSource helped them achieve this scale: “They asked us about our k-factors, the ROAS goals, and then with that they set up the optimizer on our behalf and we kicked it off. And to be honest it's been working really well since."
2. Pay attention to the right metrics
Looking at the right KPIs is important for measuring the success of the ROAS optimizer. As the name implies, ROAS is the most important metric for determining performance - it’s not CPI or retention like you might think, which are the traditional ways for advertisers to measure how effective their bids and ad creatives are. ROAS is the key metric for the optimizer because it closes the monetization loop - when profit increases, you can scale more revenue quickly and easily. However, you can also measure KPIs like scale and purchase rate, as these are often boosted by the tool as well.
When choosing a ROAS goal, it’s necessary to consider a holistic set of data, including LTV, organic uplift, and ARPU curve. Then determine your margin, which ideally results in more profit at higher scale - increasing margins can lower CPI but also inhibit scale, while lowering margins can increase costs but lead to greater scale. It’s important to also factor in granular data, like geos and device type which can affect LTV.
The team behind a popular simulation game initially focused on scale before using the ILR optimizer. When they decided that ROAS was a better metric for measuring revenue, they looked to it as their primary KPI and worked with the ironSource team to set up the ROAS optimizer. Using the tool, they exceeded ROAS goals, doubled scale - reaching as high as 3.5x more installs than before they began using the ILR optimizer - and the team was able to buy the best-quality traffic compared to any other SDK networks they were advertising with.
3. Set up and analyze the right data
Before launching the ROAS optimizer, it’s important to have enough data to let the tool do its job. The optimizer performs best when it has historical data to analyze so it can optimize bids more quickly and react faster when you add new sources or make campaign adjustments.
After turning on the ROAS optimizer, looking at only matured and aggregated data is a must for measuring performance. Giving data time to mature avoids incorrect conclusions about the optimizer tool’s performance and ROAS goals - it could be 3-7 days until data has matured and results show, depending on the type of optimizer. For example, if you set a ROAS goal for D7, it doesn’t make sense to look at the data of a user who downloaded the game 5 days ago. That user can still impact total ROAS within the next 2 days of using the app. Therefore, you should look only at the data of users who have engaged with the app long enough - at least equal to or more than the optimization age.
And regarding aggregated data, it’s not helpful to look at a day-to-day view of performance because ARPU will vary greatly even in the course of one day. Instead, looking on a week-by-week basis gives a more stable, accurate view of performance and the effect of the optimizer.
Maintaining ROAS optimizer performance
Once you’ve set up your optimizer - whether it’s the impression level revenue, IAP, or Combo - you can keep it performing at its best by adjusting ROAS goals and re-optimizing. Holidays, changes in ARPU trends or whale behavior, and game fatigue can all impact campaign performance and should be accounted for when setting new ROAS goals. We suggest updating your KPIs once every two weeks so you have at least 7 days of matured data to use as a benchmark.