UA
MAPMYRUN
TRAINING PLANS
In 2015 Under Armour acquired the Copenhagen based fitness app Endomondo. And with it, our team at MapMy integrated the personalized Endomondo training plans feature.
While it was an incredible training tool that adapted to a user’s individual running performance, in 2018 it was time for an update. There were huge user drop offs in the sign up flow and so I was tasked with cleaning up the entire experience for a feature that was designed and developed by a team from another country.
In the end, it was our most successful feature update.
Problem
MAUs have continued to decline YoY since 2016 and looking at new user rates and retention, it's clear our problem lies with retention and less so on getting new users.
Looking at users that engage with our Training Plan feature, their retention rates are XX% higher than all other users. Because of this and the convenient overlap with connected footwear offerings, we've identified Training Plans as the best starting point for addressing the decrease in MAUs while also benefiting the connected footwear experience.
KPIs
Increase in users tapping on training plan options (Current Daily Avg: ~1,200)
Increase the % of users viewing the 'My Plan' tab that are in an active Training Plan (~18%)
Increase total views of the 'My Plan' tab per active Training Plan user (0.73 views per user)
Indirect: Increase number of paired workouts per active TP user (indicating more people are actively following their plan) (0.21 pairing actions per active TP user)
Reduce TP signup drop-off rates, particularly on the 'Schedule Plan' (6.31%), 'Training Plan Goal' (4.16%)., 'Weekly Distance' (3.37%), and the 'Plan List' (1.66%) screens.
Increase the number of users saving a TP. (Last 90 Day Avg: 43,000)
Initial Work
As you can see, the original design was pretty dated. Even at this point in 2018. Most of this UI could be dated back to around 2013. And as mentioned before, this was all done by a completely different team in Copenhagen. Most of the historical knowledge about this feature was lost by this point also.
Anyone who has worked on a bolted on feature in an app can probably relate. But our biggest challenges were deciding how we wanted to shape the experience without any real baseline besides current usage metrics, and of course significant engineering constraints. No one on the current team had worked on this feature and there was little to no documentation on how things worked. We would essentially be rebuilding this as we went. Of course this meant that scoping was quite large before we even got to touch the UI.
So my job as the designer would be to assess the current experience and clean it up as much as possible without stumbling into any major code rewrites.
Phase 1
The first phase of the redesign focused on the sign up flow for the training plans. On top of the dated UI, there was a lot of confusing and inconsistent navigation UX. I updated a lot of the simple text table cells to include more content to provide better context around the choices the user was making to help guide them on through the sign up. I also updated the layouts to include more white space to let the new content and choices breathe. This in turn created a much clearer hierarchy to the individual screens, as opposed to the very simple text boxes from the past. And then of course I updated all of the selection UX to be more contemporary. I relied on the Apple HIG and Google Material docs to make decisions about when to use check box selections, picker wheels, toggles, and general UX best practices for each platform. By sticking to mostly visual UX and simple navigation updates, I was able to update the sign up flows without adding any signification scope to the engineering side of things.
Phase 2
The second phase of the redesign effort was to focus on updated the dashboard screen that shows after a user has created a training plan. This would serve as the main entry point and where a user would engage with the feature throughout the length of the plan. Through user research we knew that most users needed a clear calendar experience so they they could efficiently plan their runs for the week and month. The previous training plan experience only had the simplest of list views to show a user upcoming runs, provided no context about where they were in their plan, or even a way to start the workout from the view. So the all new calendar experience I designed included simple views to see what upcoming days had work outs and clear check marks to see progress and completed runs. And of course new actions to start workouts, log workouts retroactively and just generally keep a users plan on track.
Since this was a largely new update to the feature, it of course made up the bulk of the engineering effort. Working with the product and engineering teams, we were able to make decisions like, saving some visual redesigns for future updates, and limiting the number of new actions to include for users, to help keep scope in check.
The overall redesign was a huge success for discoverability of training plans and app engagement as a whole.
Increased user sign up sucess and engagement by over 200%
~70% more users completed planned training sessions
~40% more completions of planned sessions happening over the year since launch
Even though this project is fairly dated at this point in my career. I still like to reference this one as an example of how to do a lot with minimal resources. Since this project, it has not been uncommon to have projects that were focused on making updates to features that were developed by contractors or older teams where historical knowledge was lost. The training plans redesign showed we could make significant improvements to our metrics and the users experience, by just simple updates following best practice guidelines and looking at an experience holistically from start to finish.
The success of this project continued for the rest of my years at Under Armour. Ultimately enabling us to pair this experience with our connected footwear experience to see users make really meaningful improvements to their running performance.