“Mobile attribution” is what allows us to trace app installs and interactions on mobile devices to a specific source, like a marketing channel or campaign. The key to getting these app insights is data collection, both from the app stores and the end users’ devices. While each individual event, like a single app download or a single menu tap, doesn’t help understand user behavior, the analysis of large groups of events over time can yield useful findings.
Organizations and mobile developers use mobile attribution to understand and optimize their paid marketing funnels. Learning which customer groups respond to particular ads is invaluable: It allows teams to make decisions around which marketing activities to invest in and which to dial down.
As a result of understanding attribution for a mobile app, organizations get more insight into their customers’ personas and behaviors. For example, knowing that a specific marketing message outperforms others in a given region can help better serve customers in that region.
Let’s have a look at how mobile attribution works in marketing funnels.
Marketers frequently see that engagement metrics on a marketing campaign look like a funnel, which is wide at the beginning and narrows out towards the end. For example:
In this example, the funnel starts at a width of 1000 and ends at a width of 4, a much narrower audience than what we started with.
There are different levers that you can use to change the drop-off rate between funnel stages, like tweaking ad copy, improving app screenshots in the app store, and implementing drip notifications to keep users engaged. Since there are so many potential actions to take, more information about what’s happening at each funnel stage lets us know what’s truly going on and what tweaks might be possible.
In the context of this marketing funnel we described above, mobile attribution would help us by providing more detail for each acquisition channel being used. Grouping ad displays and conversions by ad group, keyword, device type, region, and other characteristics allows us to achieve the following aims:
To summarize, by using mobile attribution organizations and developers collect insights that go well beyond the number of app downloads or installs. The metrics we obtain through mobile attribution can help distinguish power users, understand the specific actions that users tend to take within our app, and otherwise get more actionable insights into the app user base.
To deliver useful insights, a mobile attribution funnel needs to to be able to pinpoint the marketing activity responsible for bringing a particular user onto our app. One of the key elements in this process is the identification of each mobile user.
As web users, we’re well-accustomed to the standards and conventions that allow organizations to track the performance of their marketing campaigns, namely through the use of cookie files in web browsers. For example, Google Analytics, a free tool that’s used on millions of websites worldwide, gives website administrators visibility into factors like the number of page views, top referrers, top visited pages, and time spent on the website by user.
These standards and conventions are not available in the same way in mobile ecosystems like Android and iOS. Mobile ecosystems tend to be more locked down, with iOS being the ultimate example of limited tracking options. Mobile developers rarely get access to user data through the app stores, and cannot implement cookies unlike with web tracking.
To address this information problem, mobile attribution platforms like Kochava and Adjust offer software development kits (SDKs) for developers to integrate directly into their apps. As a developer, you ship your app to your end users with the attribution SDK included. Your application code lets the attribution SDK know when different actions take place inside your app, and the SDK sends an event back to the mobile attribution service.
The attribution service then aggregates all events based on additional information that it’s able to collect, like a unique tracking ID (where possible), device type, operating system, time of day, region, and so on. Mobile OS provides basic tools for user identification, and these identifiers like AAID on Android and IDFA on iOS are key to the mobile attribution process.
Mobile attribution resides at the intersection of marketing, data analytics, and engineering. A mobile attribution platform will produce valuable insights only to the extent it can trace marketing campaigns to particular user actions in the app. Most of the value in mobile attribution is unlocked by building attribution funnels that extend from the marketing channels (e.g., by using UTM parameters) to app interactions (by submitting in-app events) and to data analysis(connecting data points from marketing and app analytics).
Building such funnels frequently requires cross-team collaboration between marketers, app developers, and data analysts.
When companies get started with mobile attribution by taking advantage of a mobile attribution platform, that platform should generate reports that will provide initial data insights. When properly configured, mobile attribution dashboards in popular tools can provide visibility into factors like the following:
In the longer term, teams usually export the data available in the mobile attribution tool, through integrated reporting options or an API, and plug the data directly into their data warehouse.
If related information like product analytics data and marketing channel spend are available in the data warehouse, it’s possible to create powerful mobile attribution dashboards and one-off research on top of a data warehouse. Other valuable data points can include financial data, customer service interactions, satisfaction scores, and referrals.
With the release of iOS 14.5 in April 2021, Apple has significantly limited the information that’s available to mobile attribution platforms. With iOS being one of the major mobile platforms, this change will impact many organizations worldwide.
Previously, Apple made available an IDFA identifier, and attribution platforms could use it to track per-device actions. More importantly, unique tracking IDs could be shared with other platforms like Google Ads or Facebook Ads to create traceable mobile attribution funnels.
As of iOS 14.5, the IDFA identifier is no longer available by default, and apps must ask the user for permission to track them.
Because of the limited options in the dialog and the specific wording being used, users normally don’t perceive any direct benefit from allowing tracking. Developers can prompt the dialog only once, effectively giving just one chance to get the user’s permission for tracking. As a result, only 20-40% of app users are expected to allow tracking.
Learn more about what this means for your mobile attribution in our guide to mobile attribution in iOS 14.5+.
Within the mobile attribution platform space, there are a few key players that offer varying functionality.
Here are a few platforms we often come across:
Each of these mobile attribution providers allow you to measure initial clicks, attribute these to an install, and notify businesses of installs and events. You’re able to demo any of the four at no cost, typically for up to one month. However, whereas Adjust and Branch offer monthly contracts for their service, Appsflyer and Kochava are available on an annual basis only. Of the four, only Branch and AppsFlyer are currently offering open pricing options.
If you’re looking to implement a mobile attribution funnel, Mighty Digital is here to help.
Thanks to our experience in building data warehouses and implementing mobile attribution in apps, we’re able to addresscomplex mobile attribution scenarios with sophisticated, growth-focused solutions.
Get in touch with us today! If you’re looking for a comprehensive introduction to mobile attribution, we’ve got you covered!