To learn more about Mapendo CPI optimization strategies, don't miss our latest user acquisition case study!
Cost per install has always been a solid pillar of the mobile advertising ecosystem, together with technology. Frequently abbreviated as CPI,cost per install reflects the cost side of user acquisition campaigns for mobile apps and games. We've setup a cost per install calculator in a different article. It tells advertisers how much they’re paying to acquire a single new user. As a consequence, it becomes a relevant benchmark to measure the effectiveness of different user acquisition channels and platforms.
Therefore, everyone in the industry aims at lowering cost per install as much as possible, by using AI, testing different channels and constantly working out new optimization strategies. In order to better understand how CPI optimization works we’ve pulled some data out of user acquisition campaigns we are running for a mobile game.
What we are going to do in this article is to share some data about our user acquisition campaign we’ve been running for a mobile action game. It’s the flagship title of a popular game developer, with a well-established user base. Therefore, the advertiser focuses its UA efforts on acquiring high-LTV users at a lower cost, thus reducing CPI.
We’ve been talking about CPI for a long time, now it’s time to get down to the nitty-gritty!
- What factors affect cost per install rates?
- Mobile games cost per install by country
- Cost per install by app category
- Cost per install by platform: iOS vs Android
- CPI vs ROAS
- Final thoughts
As mentioned above, cost per install is the cost for app developers to acquire a single new user. But how is this cost worked out and what factors can affect it ?
First of all, the cost per install, or CPI, is easily calculated by dividing the total cost of a user acquisition campaign by the number of new users acquired. Advertisers can break CPI down by country, app category, platform, user acquisition channel, etc.
Secondly, all these dimensions affect cost per install in their own way. For example, iOS users tend to have a higher CPI than android users, because of their supposedly higher purchasing power, which makes them high-LTV users. For the same reason, users from tier-1 countries lead to higher CPIs compared to users from tier-2 countries and so on.
Last but not least, there are external forces which affect cost per install too. Competition, economic downturns or growth, privacy regulations, technological updates and also seasonality can all have an impact on CPI.
Here’s how our mobile game CPI has changed over the course of the past 12 months, due to different reasons. For instance, the sharp decrease in 2023 Q4 is something that frequently happens to many apps, due to budget limitations.
Countries represent another dimension which can significantly affect the cost per install in mobile games. Indeed, by running user acquisition campaigns for our action game worldwide, we can spot the difference in the cost of acquiring new users.
As we expected, the United States and Canada lead the race with the highest CPI, followed by major western Europe countries. Instead, Japan and South Korea show the lowest value, both under $1.
However, these differences are related to the mobile game category, more specifically to the action game subcategory. Other mobile game categories, such as hyper-casual, can show totally different values, based on their different target users. Furthermore, the difference in the monetization strategy (IAPs, in-app purchases vs IAA, in-app advertising), affects CPI too, because it tends to be more expensive to acquire paying users who are willing to spend money in-app
We are taking a step further in the next paragraph, which compares cost per install by countries and also app categories.
The fluctuations of cost per install within different mobile game categories become even larger when mobile games are compared with other app categories. In this case, we combined data from our mobile game UA campaign with Appsflyer database in order to look at the full picture.
We have chosen three different app categories: shopping, finance and dating. Apps belonging to these categories monetize in different ways. Shopping apps make most of their money out of transactions made by users to purchase items. Finance apps can be divided into multiple subcategories, such as mobile banking, remittance apps, trading and so on. Based on the subcategory, there are several monetization events like deposits, money transfers, trades, etc. Finally, dating apps monetize through both in-app ads and in-app purchases, which are often subscriptions.
All these monetization strategies reflect different target audiences, each at its own cost. Please find below a chart which breaks down the cost per install by both app category and country.
As we mentioned earlier in this article, one key dimension which affects cost per install is the platform, or operating system. Indeed, iOS users come at a higher cost, since they are supposed to have a higher purchasing power than android users. Therefore, they could be willing to spend money in apps, through IAPs or subscriptions, and become high-LTV users. Hence, their higher CPI.
On the other hand, android smartphones have a high market share in most of the countries, so there are more users to target compared to iOS, and at a lower cost. In addition, recent privacy regulations introduced by Apple have made mobile attribution, and as a consequence user acquisition campaigns, more difficult when targeting iOS users.
This difference in cost per install between iOS and Android occurred to us too, when running user acquisition campaigns for our action game. As you can see from the chart below, there is a significant variation between the two platforms.
We would like to conclude this article by highlighting the growing importance of another metric, which is likely to overcome CPI as the leading measure to assess app campaigns. And we are speaking about Return On Ad Spend, a.k.a. ROAS.
ROAS measures how much advertisers get in terms of in-app revenues from new users acquired through UA campaigns, compared to the cost of acquiring such new users. This metric is highly adopted by app developers for 3 main reasons:
- Effectiveness: it clearly measures the effectiveness of user acquisition by comparing revenue and costs;
- Quality: it provides a better understanding of user quality than any other metric
- Optimization: it adds a new layer of optimization, allowing UA platforms to optimize their performance towards ROAS.
Speaking of the last point, platforms like Mapendo are constantly developing new optimization strategies to meet clients’ expectations and ROAS optimization is amongst them. We are training our ML algorithms to optimize UA campaigns towards ROAS and we can adopt multiple additional strategies:
- Dynamically adjusting CPI in order to meet ROAS target;
- Combining different traffic channels;
- Improving targeting through user segmentation between paying and non-paying users.
In conclusion, the cost per install (CPI) landscape for mobile apps in 2024 is as dynamic as ever, while new emerging metrics are taking the stage.
App marketers are increasingly adopting a data-driven approach, focusing on user acquisition strategies that deliver high-quality users and long-term user value. As a consequence, CPI is losing its appeal compared to ROAS and other quality-driven KPIs.
By carefully selecting advertising channels, optimizing ad creatives, and leveraging machine learning technologies, businesses can enhance their ROI and achieve sustainable growth in the mobile app ecosystem.
Ultimately, staying informed about industry trends and being adaptable to changing conditions will be key for any app seeking to thrive in the competitive mobile market. The future of cost per install will undoubtedly bring both challenges and opportunities, and those who are prepared to innovate and evolve will be best positioned for success
If you are interested in learning more about our CPI and ROAS optimization strategies, reach out to our sales team!
