Yesterday, an article by Eric Sefert, head of marketing at Wooga, appeared on the pages of Mobile Dev Memo. In it, Eric talked about the art of “soft” launching mobile applications and the importance of traffic sources. We offer you the Russian version of the material.
Most economic models assume that a product at a certain point in its development reaches a stable state in which users and income from them are in a constant range of values. Such an assumption at the prototyping stage of the product is useful, it allows you to predict the situation.
In reality, everything is different. Metrics do not “settle” in a static framework during periods of growth, moreover, their values constantly require decision-making for further work on the product. At these stages of the product’s life, the size and nature of new user cohorts can change radically from day to day, greatly affecting the size of daily revenue and the level of engagement.
In other words, in conditions of constant growth, metrics are moving targets.
The central problem in modeling and forecasting metrics is the nature of visibility in app stores: an app can become more popular by climbing the chart. As a result, traffic sources are mixed. This makes setting up indicators at the time of a “soft launch” difficult.
We mean that during a soft launch, the user base, as a rule, consists entirely of users attracted by an advertising campaign in one small country (for example, Canada). A mix of promotion channels and demographic changes makes the release a global launch. Especially if the mix includes stimulated installations and organic installations that come from the promotion of the game by the platform itself.
But the point is that metrics should be predictable throughout the soft launch, so that then, based on the results obtained, it would be possible to make informed decisions after the global launch. To be predictable, they must be pure (from a single source).
The essence of solutions, as a rule, revolves around the product’s compliance with market requirements: whether the LTV project exceeds the marketing costs that will then be spent during global promotion, how much the project can potentially earn with a certain number of users, and so on.
The scale of the marketing campaign determines the cost of marketing costs – the greater the coverage of the marketing campaign in the region (i.e., the more users the developer hopes to acquire through marketing), the higher the installation cost will be. The steeper the marketing campaign in a certain segment, the smaller the difference between LTV and CPI in it.
Mixed traffic and difficulties with user engagement, which affect LTV, are becoming the main challenge for developers.
LTV, in theory, is understood in the context of the intrinsic value of the application relative to the user, but this ignores the fact that different marketing channels reach different users under different conditions. This is most obviously demonstrated by the difference between motivated and unmotivated users. Motivated people download the app only in order to receive the promised reward in another application. In other words, they are clearly not motivated by a desire to try the application.
The difference in motivation and enthusiasm for the product is less obvious between users acquired from other sources or channels. For example, the behavior of organic traffic, users attracted by word of mouth and fullscreen advertising is more similar. But still, it is impossible to assume that users acquired in different ways will behave the same way.
When the user base is large enough and the number of new daily users is relatively constant, this difference cannot significantly affect key metrics, unless, of course, there have been major changes in the traffic source.
But such an approach cannot be applied when switching from a softlatch to a global launch or in the first weeks after a global launch. Traffic sources, in terms of not only marketing channels, but also geography, change significantly after the world release, especially if the project is being implemented in any region. The decision to launch the application on the global market requires foresight, understanding how the release will affect the revenue of the application, how LTV can be predicted for each individual channel, how the channel size can be predicted and how it can affect the size of the user base and how much each of these channels will cost you.
Be that as it may, the only main metric is what benefits the product brings. The peaks of downloads that occur after a celebrity has written about the application or after promoting the application in the store or mentioning it in some popular publication reduces the average retention. The situation is also fair, like the one when targeted advertising at a soft launch gives out inflated indicators that cannot be scaled in the future.
App revenue is absolute, all other metrics can change, which is why user segmentation is such an important task in mobile marketing.