Case Study: Strava

Strava is Swedish for “strive,” which epitomizes their attitude and ambition: It’s a passionate and committed team, unified by a mission to build the most engaged community of athletes in the world. Every day, Strava are searching for new ways to inspire athletes and make the sports they love even more fun.

But it’s not only about achieving – Strava is made up of an inclusive team, dedicated to elevating each other and the members of our community. That balanced approach has helped us revolutionize our industry, with millions of athletes using Strava every day.

And for Strava, the unique challenge of attributing real lifetime value to a subscription-based service — with minimal post-install visibility from their networks — was a real hurdle to cross.

This case study is a look at what Strava really needed from an attribution provider, and how Adjust met their individual needs in a way no other solution could.

Strava is an exercise app that doesn’t have a conventional way to monetize users, which means that there is no traditional app conversion flow. Typical apps — like gaming or eCommerce apps — focus on early monetization via in-app purchases after a free download. For those apps, app installs are an important KPI to track performance, but for Strava, app installs aren’t as important as user engagement. It isn’t about finding any user to download, it’s about finding the right user they can monetize.

To illustrate, Strava offers a set of free services that any user can access, as well as several types of subscription services bundled into packs under its Summit product. There are different packs for different types of users — such as a Training pack, a Safety pack, and an Analysis pack.

In addition, a key portion of Strava’s app is the social component. In order to provide better value for its users, they needed users to upload their exercise data to contribute to the wider Strava community and further inform the features within the Summit packs.

This is why Strava prioritizes engagement over acquisition. Getting a user to install is nice, but it isn’t as valuable if that user doesn’t interact with the app. This meant that the usual KPIs — installs, 7DARPU, ROAS — don’t run the full trail of Strava’s conversion window. Getting users to interact and purchase Summit packs is naturally a longer sales cycle than getting users to purchase in-game currency or retail apparel.

The unique nature of Strava’s business model identified two types of users that Strava wanted to capture. They wanted to get users that would interact with the app, and within that subset of users, find the ones that were likely to purchase Summit packs. Naturally, the company wanted more of these types of users, while at the same time being able to manage its campaign spend effectively.

Strava's growth model required the team to identify and set up unique KPIs further along the conversion funnel in order to find the users who converted.

One of the KPIs Strava came up with was Cost-Per Strava Uploading Member in 7 Days (CPSUM7D). This metric tracks the cost of acquiring a user that would upload an activity within seven days of installing the app. Because Strava found strong correlations between initially active users and eventual paying customers, the company wanted to know how much it cost to get the users that interacted with their app. CPSUM7D was seen as one of several core indicators on the performance of Strava's marketing efforts.

CPSUM7D is representative of the kind of KPIs that Strava focuses on — it's not just about acquiring users, it's about acquiring users that become active users of the community which over time will sign up for Summit.

In order to effectively implement its unique KPIs like CPSUM7D, Strava needed an attribution provider that could cover the long tail. Since it runs most of its mobile campaigns on self-attributed networks, there was minimal visibility into the campaigns that were performing the best under their CPSUM7D targets. Without an attribution provider to fill in the blanks, it was extremely difficult to allocate budget and measure performance based on their unique KPIs.

That said, only one half of the equation is identifying the users you want. The other is knowing how to find them.

Michael Nguyen of Strava says, "Paid users come in with very, very different levels of intent, so without Adjust telling us that these are the types of users we should be going after, we would be flying pretty blindly."

Since Strava uses mostly “self-attributed networks” (large advertising networks such as Facebook) when running paid campaigns, there needed to be a partner to continue to illuminate the user journey after they install.

Without an attribution provider that could easily track user activity post-install, it would be hard to judge the performance of a network. Adjust is able to connect attribution with post-install activity, allowing Strava allocate its budget to fully leverage their unique KPIs.

Strava went with Adjust because of the ease of our SDK integration and our robust attribution capabilities. Not only does Adjust have the event tracking, real-time data, and simplified campaign setup they needed, but our native S3 support also allows Strava’s marketing team to pull raw data reports from the S3 buckets daily for deeper analysis.

Integrating Adjust makes it possible for Strava to operate at scale. Without huge marketing and data science departments at its disposal, Strava is still able to execute smart mobile marketing because Adjust’s data flexibility and ease of use lets the company run sophisticated campaigns that push real uplift.