Intro to Incrementality: The Key to Measuring Your Ad Effectiveness
Trends & Insights
November 30, 2022
by
Dahyun Kang

A mind-boggling amount of data – 2.5 quintillion bytes – floods out of the connected devices we use every day. Savvy marketers have set on “data-driven” marketing to capitalize on such data abundance, but impactful insights are rare. Having access to data does not exactly mean knowing what to do with it.

The struggle is obvious when it comes to measuring marketing performance. A survey conducted by DemandGen revealed that more than a third of marketers admitted they needed improvement in understanding marketing effectiveness. What is more, the pandemic has accelerated growth in mobile device usage, pushing marketers to master not only web attribution but also the more complicated mobile attribution.

That is why in this post, we’re walking you through what mobile app attribution is and how it works. Read until the end to effectively drive growth. Let’s dive right in.

What is attribution?

Marketing attribution, in general, is the practice of matching causes such as channels and campaigns with effects such as purchases and revenue. Simply put, it is the process of understanding the impact marketers made on a specific goal.

As Theodore Roosevelt once said, the more you know about the past, the better you are prepared for the future. An accurate analysis of historical data is a prerequisite for business success. Building a winning paid marketing strategy is especially important because it involves cost – companies cannot remain in business without turning a profit. Given that global digital ad spend was around $378 billion in 2020 alone, marketers must identify the best working channels and campaigns to improve return and profitability within a limited budget.

All of this might sound complex and challenging. Thankfully, with marketing attribution, you can clarify the connection between inputs and outputs and unlock insights vital to growth.

Web attribution vs. mobile attribution

Market share of mobile devices was as high as 58.33% in October 2022, but before such domination, marketing measurement was more focused on desktop web. With web attribution using cookies, UTM parameters, and referrers, marketers could capture certain user behaviors and track traffic sources to their websites.

By definition, mobile attribution is the same as web attribution. However, tracking mobile user journeys is much more difficult because user identification is lost as users move from mobile web to app markets. Mobile apps were built without the notion of cookies, and the “walled gardens” in the mobile ecosystem essentially make it impossible to take the approach that works on the web.

For this reason, many app marketers use app attribution tools or Mobile Measurement Partners (MMPs) like Airbridge. They provide a streamlined view of data from multiple sources, reducing the burden of going back and forth between various ad platforms. Preventing ad fraud and adapting to new frameworks like SKAdNetwork are some other benefits of working with attribution providers.

Essential ingredients for mobile attribution

App attribution tools collect and process user-level data, offering insights into the power of marketing efforts. Specifically, they look into user journeys which consist of touchpoints and conversions.

Mobile User Journey

Touchpoints are clicks, impressions, and other external events that lead to app installs or in-app actions. There are three main sources of touchpoint data:

  • A tracking link is a regular URL with tags added to the end. These tags, also known as parameters, contain information on user behavior. You can create a tracking link using an app attribution tool.
  • Server-to-Server (S2S) API, also known as server-side tracking, generates a unique ID when there are app or web activities, sending the data from the user to you before it is transmitted elsewhere. Self Attributing Networks (SANs), in particular, use S2S as they attribute their traffic without involving a third party. Unlike regular ad networks, SANs only report the last-touch activity to attribution providers via API.
  • A referrer is a unique string of characters specific to Android devices. This anonymous ad tracking ID is sent to Google Play when a user clicks on an ad, and if an app install or an in-app event happens consequently, the result is passed back to the app attribution tool.

Conversions refer to in-app events such as install, purchase, and add to cart. Most marketers use app attribution tools or MMPs for conversion tracking.

  • Mobile apps can directly implement SDKs provided by app attribution tools. Your app code lets the attribution SDK know when different actions take place inside your app, and the SDK sends the events back to the attribution provider.
  • Server-to-Server (S2S) API can be used to track not only touchpoints but also conversions.

How does mobile attribution work?

Mobile attribution is all about associating touchpoints with conversions. This process, called identity matching or identity resolution, ties and attributes all a user’s behavior into a single profile. In other words, app attribution tools track user behavior scattered across multiple devices and link their interactions with specific ads. There are two primary matching methods:

  • Deterministic methods use personally identifiable information such as device ID, prioritizing accuracy. The identifier – GAID for Android devices and IDFA for iOS devices – assign an identity to a mobile user to track that person across web and app. Since iOS 14.5, mobile apps have to request user consent to access IDFA.
  • Probabilistic methods create device relationships by compiling individual pieces of information such as a device’s operating system, IP address, Wi-Fi network, etc. Devices are grouped together based on the data and associated with an identity that is “probably” accurate. These methods are often used when deterministic data is not available.

While identity matching is the foundation of mobile attribution, the story is a bit different when it comes to SANs. As explained earlier, these “super publishers” measure performance on their own. When a SAN reports user activity to an MMP through an API after the conversion has taken place, the MMP asks the SAN the device ID of that event, and the SAN provides relevant details so that the MMP can draw additional insights from the conversion data. Note that with IDFA deprecated in iOS 14.5, device IDs are no longer used in reporting on iOS devices.

StoreKit Ad Network (SKAN) is another type of independent attribution. Developed by Apple to attribute iOS campaigns in a privacy-preserving way, the SKAN API hides all user-level or device-specific data and measures on an aggregated level. When a user converts, the winning ad network receives validation postbacks from Apple, and then sends them to an MMP which is responsible for connecting the dots to make sense of the data.

Holistic attribution allows marketers to optimize marketing strategy, but it is extremely demanding for an individual app to integrate, report, verify, visualize, and analyze on its own. Hence, for mobile attribution, it is recommended to work with an app attribution tool or an MMP.

The ground rules of mobile attribution

When pairing touchpoints with conversions as described above, you need to decide how much credit to give to each touchpoint in the conversion path. This is where the concept of attribution model comes into play. By assigning values to users’ interactions with ads, marketers can determine which channels and campaigns are the most powerful.

In the majority of cases, user journeys consist of multiple touchpoints, meaning that there are various ways to attribute. The two primary approaches are rule-based and data-driven. Popular attribution models such as the single-touch and the multi-touch are all rule-based.

Most app attribution tools are built on the simple and heuristic Last-Touch Attribution (LTA) model, which gives full credit to the single last touchpoint. Nevertheless, keep in mind that each tool works under different priorities, rules, and criteria, and thus the attribution result for the same conversion could slightly vary.

👉 Click here to read about attribution models you must know as a marketer.

Drive growth with mobile attribution

Mobile attribution is a retrospective causal analysis that helps you identify your strongest channels and campaigns. While this is in itself meaningful, what is more important is to keep adapting in the ever-changing marketing landscape. There is no one-fits-all approach to marketing success, and only by consistently and proactively modifying your action plan can you seize opportunities.

Airbridge aims to take a marketer-centric approach and build an app attribution tool that actually benefits marketers. As your reliable marketing partner, Airbridge will continue to provide industry-leading solutions to prepare you for the future.

👉 Check out this post to find out about Airbridge’s innovation in tackling marketing challenges.

Want to get more insights?
Get a mail whenever a new article is uploaded.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Dahyun Kang
Product Marketing Manager
Dahyun is Airbridge’s Product Marketing Manager. Being responsible for Airbridge’s blog, social media accounts and webinars, she is passionate about building a brand’s visibility through data-driven content.
Subscribe to the newsletter for marketing trends, insights, and strategies.
Get a mail whenever a new article is uploaded.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Related Posts
Discover true sources of growth with Airbridge’s Unified Measurement Stack.
Discover true sources of growth with Airbridge’s Unified Measurement Stack.