At ironSource’s “Setting Up, Analyzing, and Optimizing Your Ad Monetization” webinar, our growth strategy team leader, Elyse Krumholz, guides you through how to best analyze and optimize your monetization strategy. Let’s take a look at what she discussed.
First, she noted that ad monetization is made up of two parts: your ad implementation strategy and ad network strategy. Ad Implementation strategy is all about deciding how to incorporate ads into our game, while ad network strategy decides which ad networks to work with and how to best utilize them. The best way to optimize our strategy? A/B testing.
In the webinar, Elyse wants to check out whether a new rewarded video placement is generating incremental ARPDAU, without sacrificing retention - so she runs an A/B test and sets two groups: one group with the new rewarded video placement and one group without it.
Following the A/B test, here are five reports you’d need to analyze performance and choose the test winner.
Performance reports
To start, Elyse guides us through performance reports, which show us KPIs like revenue, eCPM, fill rate, and impressions. We can filter here by metrics, or, in this case, by groups A or B to check out results of our A/B test.
Elyse filters by rewarded video (this is the ad unit we care about) and then breaks up A vs. B. Look at that – revenue and impressions are higher for group B (our test group with the new rewarded video placement). Great! But, according to Elyse, that’s not the whole story. What does this mean in the context of ARPDAU?
User activity reports
Next up: user activity reports. Here we can understand ARPDAU, DAU, DEU, engagement rate, impressions per DEU (usage rate), impressions per DAU, and sessions per DAU.
Using this report, Elyse compares the ARPDAU of A vs. B groups - and finds that ARPDAU for group B is higher by 25%. But why? Perhaps our game is engaging more users with ads than before, or maybe users are watching rewarded videos now. To find out, let’s check how the engagement and usage rate (number of impressions/DEU) differ by A vs. B.
As it turns out, the engagement rate didn’t change – in other words, no new users are engaging with rewarded video ads. However, usage rate in group B is 5, compared to 4 in the A (control) group. Basically, engaged users in your game are now watching 5 rewarded videos per day instead of 4. Finally, we see what leads to the increase in ARPDAU - more rewarded videos watched per session. But we’re still trying to figure out what increases ARPDAU without hurting retention. So let’s investigate further: how is this new rewarded video placement affecting our retention and LTV?
Cohort reports
To understand the effect on retention and LTV, we need to move to cohort reports. This way, you can group users who started the game on the same day (a cohort) and measure specific KPIs for these groups over different timeframes. Breaking down our results by A vs. B, we can understand if users who are now watching more rewarded video ads because of the new placement are also still sticking around to play our game.
Going to the cohort report, we filter for the time period of the test, break by A/B group, and see the results – a feature exclusive to LevelPlay. We find that a higher percentage of users in the test group also had higher D7 retention, compared to the retention rate of the control group. We’re getting higher ARPDAU due to higher impressions and higher usage rate, and those users tend to play our game for more days.
Real time pivot reports
Finally, we can dive even deeper into our data with real time pivot reports - which let us analyze nearly all the KPIs we just walked through, all in real time. So, from the minute we start the A/B test, we can examine changes in user behavior and ad network performance. Most importantly, we can react quickly to these results.
Beyond the A/B test, we can also optimize our game according to real time data, track live performance after updating our mediation stack, know about network drops right away, and monitor the stability of new app versions. We also have a compare mode, where we can compare KPIs side-by-side on one screen and easily compare KPIs all at once. We can also understand changes between different time periods very easily.
Ultimately, we can see how performance reports, user activity reports, cohort reports, and real time pivot reports work together to enhance our monetization optimization strategy. Using all of these reports together, we can dive more deeply into the data, to not only understand “how,” but also “why” – and in doing so improve our monetization strategy. To learn more, watch the full webinar below.
