Building a Mobile Attribution Framework Doesn’t have to be Scary

Mobile analytics can be tricky and after years of working with clients on their mobile framework, I've seen major advancements in the mobile analytics field. A solid, mobile attribution framework will help you gather mobile data to improve the user experience, generate personalized content and even build a retargeting program to improve conversion so I’m going to help you breakdown the approach logically so you can not only fully understand your mobile channel but also build your own mobile attribution framework.

But before you start to think about attribution across mobile devices, you first need to wrap your head around the difference between mobile and traditional web analytics. I recently wrote about how to implement analytics tracking for mobile devices, take a look at it here.

Know your game plan before the green light.

You’re probably thinking, "Great Bill, I have to track the what?"  If you're going to build a solid attribution framework for mobile apps, you have to have a game plan. Here are some preliminary questions to help you get started:

         · How does a visitor identify themselves within the mobile app?

         ·  What actions are available within the app for a user to complete?

         ·  What actions can only be done on the web (website or mobile site) and thus would require someone to jump out of the mobile app to complete?

          ·   What investments are being made to promote the app or drive conversion across multiple devices?

          ·   Do you have a sound strategy for tracking all of your digital campaigns?

An important thing to know about your mobile app is that people will naturally feel more comfortable remaining signed in or even having to login in the first place since it is their mobile device. Don’t be afraid to push users to identify themselves when they set up the app for your future use.  If you’re not doing that now, then that should be in your next release of the app. Highlight the benefits of creating an account, make it easy to set up an account (social login options, etc.), and then reap the benefits of being able to collect the right information going forward.

Next, don’t attempt to reinvent the wheel with your analytics tracking within the mobile app (I outlined this in the newsletter as well). It doesn’t matter what screen the user is on within the app, all that matters is what actions they perform within your app. When you’re building an attribution framework, it’s as simple as tracking the following:

·  Users: Track individual users based upon a login or loyalty account

·  Device: Identify the device, OS, and app version

·  Actions: Track the actions within the app as they happen or simply collect them if the device is offline and send the data later


Tackle the tech: look for resources already available to you.

When it comes to the technology for tracking, you have several options. Your web analytics vendors should offer a Mobile Software Development Kit (SDK) to help make it easier to develop. Tag Management Systems (TMS) also offer SDKs, and then you also have the mobile-specific analytics vendors such as Flurry (free) and Localytics (Enterprise) to provide channel-specific information.  If you pick a new technology vendor to track just your mobile channel, there are also several options for exporting that information out of that system and into your web analytics vendor. Ultimately, it doesn’t matter which route you go provided you vet out your business requirements up front and ultimately pick the solution that gives you the most flexibility going forward.


Conclusion: be realistic, linkable and likeable.

At this point, attributing conversions to the users for the events happening within your mobile app should be easy if you’ve taken into account and completed all of the things I’ve discussed thus far. However, the more difficult hurdle emerges - tracking the same person across their mobile device as well as your other channels.

Here are some final bits of advice:

Mobile analytics is not an exact science.

First things first, remember that this will never be an exact science and if you’re looking for 100% accuracy you came to the wrong place. The whole point here is to do the best possible job at collecting information so you can be confident in the decisions you make based off the story the data is telling you.  After all, this is the only reason we collect this information in the first place.

Be linkable (& likeable).

You already have the user identified within your mobile app (hopefully. So, if there are actions that can only be completed on the web then include those links within the mobile app. This makes it possible for a user to easily jump to the site which will pass along identifiers via the link to ensure you can track that activity from start to finish.

If the user doesn’t jump from within the mobile app and just simply comes to your website from a laptop or a mobile browser - ensure you are tracking those actions appropriately within your web analytics solution:

o   If the action requires a login first, then you’re all set

o   If the action doesn’t require a login, then you have to track those actions as they happen and tie it all together down the road once that user provides identifying information (which may happen in a future visit)

If you break down the approach logically like I did in this blog post, you’ll be well on your way to not only understanding your mobile channel but also to building a solid, mobile attribution framework.  Have you started working on or maybe even completed your mobile analytics framework? What is/was your biggest challenge? Start the discussion now!


If you have any questions, please feel free to leave a comment and I’ll try to help as best I can.


By Bill Bruno
About the Author:

Bill Bruno is the CEO - North America, Ebiquity.

Contact Us Now