With mobile advertising sweeping the digital marketing space, the option to use a cost per install (CPI) pricing model has become increasingly attractive to many developers. This is similar to a cost per acquisition (CPA) model in that the marketer pays when a new user is acquires, in this case in the form of an install of some piece of software.
This can work well for developers in the mobile space because many applications use a freemium model in which the app itself is free but certain in-app content requires the user to pay. However, this also presents several potential problems for the marketer. In particular, there is usually a three to five day lag between the download and becoming a paying customer; therefore, the marketer could be paying significantly for installs before receiving any revenues. Additionally, focusing solely on installs could mean neglecting user profiles that can determine the likelihood that a user become a paying customer.
To help to address these shortcomings, one can implement a system of optimization pixels. These are indicators tracked by an automated system that can enable marketers to optimize a CPI campaign more efficiently without having to get complex data from the developers.
For example, an optimization pixel may be when the user completes a significant action that indicates a high likelihood of becoming a paying customer. In the case of free-to-play game, this could be finishing the second level. The idea is that player who complete the second level are more likely to keep playing and thus more likely to make in-game purchases. This extracts all of the users who simply download the game without playing it, or those who play the first level but quickly lose interest.
In the case of apps that take a relatively long time to monetize, multiple pixels can be implemented. This can be used to generate detailed profiles of the types of users who will likely become paying customers.
The major advantage of these pixels is that they provide an earlier picture of who is likely to monetize. Going back to the earlier example, by focusing on optimizing for the types of players who complete level two, you could hone in your CPI campaign faster and with more agility than if you waited to see who monetized.
Furthermore, these optimization pixels can also be used to track in-app user experience. This data can then be applied to make a more compelling value proposition within the application itself. For example, if you had two optimization pixels, one for completing the tutorial and the other for completing level two, you can identify drop off. Perhaps you notice that the users who take more than two minutes to complete the tutorial are less likely to reach the second optimization pixel. Or perhaps you realize that non-English speaking players are less likely to become customers because all of your calls to action are in English.
One powerful thing about optimization pixels is that they can be easily implemented into the existing tracking of a CPI campaign. By using them, you can improve how quickly you collect data in order to optimize your advertisements. This is invaluable when you are trying to drive significant sales within a short time period or simply trying to remain agile.