April 12th, 2018
These days, some of your most loyal and highly converting users are your mobile app users. When you compare mobile app engagement to mobile web engagement, the results are particularly telling—comScore found that time spent in mobile apps is 20x higher, and Criteo found that conversion rate on mobile is 3x higher and that app transactions account for 71% of sales on mobile. Even with the dramatic rise in mobile app conversions and engaged app users, many mobile marketing professionals are still figuring out proper mobile app conversion tracking.
Conversion tracking means more than just logging the right in-app events. Like most mobile professionals, you likely want to know the origins of your converting users. Mobile app installs are no longer the only goal marketers have anymore. Acquiring engaged, retained, loyal users who convert is an essential element of maximizing in-app conversion.
Ideally, of course, you have a large group of highly engaged users that you know came from marketing channel X. Based on that data, you then invest more money in user acquisition from marketing channel X. However, in reality, the mobile ecosystem is far more complex—and, with most attribution providers, the clarity of your attribution data (and the confidence you can have in it) almost always suffers. Unlike user behavior on the web, user behavior on mobile often transcends any one siloed channel or platform. Due to the difficulties that accompany accurate cross-platform web and app attribution, most attribution providers are left to focus on either web or app, rather than both. In the rare cases that legacy app attribution providers also cover on web channels, they rely on requiring a user to sign in on both platforms. Given that users rarely login on both platforms, most could not be attributed to their original web source.
What does this mean for you, the marketer? App user conversions that originate from any source on the web have a good chance of being incorrectly attributed or not attributed at all—which means that web-based marketing campaigns are not receiving credit for the conversions they generate in the app, and vice versa. Single-platform attribution providers are essentially for companies that separate their web attribution from app attribution. As we’ve mentioned, the paradox within this approach is that users never just interact with one or the other.
Why is this the case? Websites offer immediacy, since they don’t require an installation to interact with content, and can be a good fallback if a user does not have the app. Apps offer offline capabilities, stored user context, and pre-rendered components, preserving data usage and giving users a higher-quality experience. Both web and app are necessary in the lifecycle of a user, so it is not simply a matter of prioritizing one platform over the other. On the other hand, businesses that can better cater to consumers’ cross-platform behavior see more than 50% of transactions completed on mobile.
Companies will never fully understand the true user journey until they consolidate web and app measurement into one attribution system. Building a user experience that optimizes for conversion must consider the web and app flow as one whole user experience, not two separate experiences.
So, how can you consolidate web and app measurement and, in doing so, de-duplicate users that interact with your company’s brand on more than one channel? Let’s dive into how you can set up consistently accurate, seamless, and simple mobile app conversion tracking in three steps:
The first step of app conversion tracking is making sure you’re tracking the events that are important to you. The obvious choices here are the events that drive the greatest revenue for your brand. For an e-Commerce company, for instance, that’s the purchase event—but these companies also will likely track the add-to-cart event and product viewing events. For digital media, on the other hand, revenue comes from repeat engagement, so these brands may wish to track articles viewed, videos viewed, and some broader user engagement numbers like MAU (monthly active users) or WAU (weekly active users).
What some marketers might overlook is activation events like sign-ups or registrations. For some mobile apps today, users who download a given app don’t really count as “users” until they sign up. That’s because, depending on the mobile app in question, users who don’t sign up might not be able to complete a revenue event. A user signing up or registering is also a good indication that he or she is likely to stick around, and is therefore at a lower risk for dropping off. If your funnel is long and requires a lot of completed events before users truly convert, tracking activation can be a best practice to make sure you’re acquiring users, and not just mobile app downloaders or visitors.
Ultimately, if you’re looking to optimize for a event in question, you should probably be tracking it.
Alright, you’re tracking every important event in your mobile app—now what? You need to make sure you know where converting users are originating!
First, it’s important to determine all the ways your users could find and access your mobile app. You’ll be surprised by how many marketing channels can apply (and, in practice, do apply) to your app—the mobile web, the desktop web, email, text and SMS, social media platforms (among them, Facebook, LinkedIn, and Twitter), other apps, and paid promotion (including social media ads, Apple Search Ads, and more), among others.
Next, it’s critical to confirm that you are monitoring your strategies not only for engaging with users within and across each of these channels, but also that you’re tracking them reliably and accurately. What’s important to stress is that basic probabilistic modeling and cookie-based methods of measurement and tracking do not allow for reliable cross-channel, cross-platform user attribution.
If you’re currently relying on basic probabilistic modeling or cookie-based methods of attribution, it could well be worth looking into Branch’s attribution engine as a far-superior alternative.
So, you’ve collected the attribution data and the conversion data pertaining to the events that matter to you—but how do you connect them? There are two solutions here, one more complicated and challenging to maintain than the other.
Several companies choose to connect and maintain their own wells of attribution and conversion data. Oftentimes, conversion tracking and attribution tracking are set up separately, and thus remain separate unless a data team works to render the data compatible. As you can imagine, this is a complicated and tedious task, and is not one we’d recommend.
If you already use Branch, you know we offer services ranging from deep linking and deep linked email to personalized web-to-app smart banners and attribution, in order to help you provide the most consistent, personalized, and enjoyable cross-channel and/or single-channel user experiences out there in the digital ecosystem. That said, optimization is more than good strategies, even when you’ve got the right tools—it’s also about consistent review and potential use of accumulated data.
The Branch dashboard offers a single, unified place for marketers and product professionals to access the valuable data they need to assess current and past campaign performance while driving forward-thinking strategies and campaigns. We recognize, however, that the Branch dashboard is not the only place you’ll want to access and review this data. Through our Data Feeds toolkit, we offer four distinct tools for exporting unified attribution data collected by Branch: Data Integrations, Webhooks, a Data Export API, and a Query API.
When Nextdoor needed to ensure they were attributing conversions and engagement back to accurate advertising sources, tapping into Branch’s valuable well of attribution data was a no-brainer. As a social network, Nextdoor depends on highly-engaged users, so it’s vital for them to know where these users originate (and where they didn’t originate) so they can double down on those marketing channels. Branch helps them do just that. You can check out our Nextdoor case study for a closer look.
Next time you’re preparing to run a big marketing campaign, remember that, if you settle for basic probabilistic modeling instead of Branch attribution, you may end up spending money on channels that bring you installs but don’t really bring engaged paying users.
To access more tips, growth hacks, and best practices for attribution in 2018, check out our Ultimate Guide to Web and App User Attribution.