SPOKE

SPOKE is a Meditation mobile app that aims to break the mold of mindfulness by using music as a mean of making the wellness industry more accessible and relevant to a wider audience. As SPOKE's sole in-house product designer one of my main responsibilities is to constantly track in-app user data in order to evaluate the app's performance within its audience and understand how to best meet our user needs.
‍
The app has been facing a retention problem since it's launch in BETA testing phase: whilst early signs on retention were promising, driving regular usage in first few in-app sessions seemed to be challenging.

This project involved collaboration of all the spheres of the business: Musical Content Creation, Research, Copy, Strategy and App Development. Its aim is to monitor in-app user data in order to uncover performance issues and opportunities for improvement - with the goal to guarantee future user retention on the app, ideally reaching 10% of retention (higher than Calm and Headspace that both have around 8% as previously disclosed by the companies).

Previous Status
In the beginning of 2021 the SPOKE Beta app was an MVP composed of 3 main elements:

SPOKE - Old Session Recommendation via Check In Screen Recorded Video

Session Recommendation via Check In

Once installing the app and opening it for the first time, the user would be prompted to complete their first Check In, when they would answer questions about how they were feeling and what was on their mind by selecting the options given on the screen.
‍
‍The Check In was composed of the following 3 steps:
1st step: picking one of the emotions from the Emotions Axis
2nd step: choosing one or more topics from the Themes Selection
3rd step: answering a Reflective Question

The purpose of going through the Check In process was to collect punctual data about the user's mood and feelings to be matched with one of the Meditation albums/ playlists available on the app and have it recommended as the first session for the user to listen to.
‍
Based on the data inputted by the user on the Check In, a full album/ playlist would be recommended to them, that way each user would have a different initial session on the app that would be personalized to their mood and needs when accessing the app for the first time.

Explore Section

Another way of selecting a session to listen would be by going to the app's main section: the Explore Section and manually choosing a session to be played.

That was also the app's Home Screen, so whenever returning to the app after the first app use, this would be the screen where any user would land.

The Explore Section is where the user can discover the 7 individual albums or playlists available on the app, browse through the different albums and the sessions inside them.

The 7 albums available on SPOKE were:
1. Foundations
2. Desire
3. Heart
4. Purpose
5. Expression
6. Wisdom
7. Meaning
Each one with 4 sessions (tracks) inside.

SPOKE - Old Explore Section Screen Recorded Video

User Profile

‍The User Profile was accessed by tapping on the avatar icon on the app's Explore Section.

On the profile users could track some personal stats about their app use, such as their daily streak and the total amount of in-app listen time. The history of emotions and themes selected by the user on all the check ins completed was also available for their reference.

It was possible to complete a new Check In by scrolling down on the profile and finding the Check In button at the bottom.

On the top of the profile screen there was a user feedback component, allowing customers to send direct feedback to us by clicking on the link and completing a structured feedback form.

In order to identify opportunities for increasing in-app user retention, the approach used for our Discovery phase included:
‍
- Analysis of customer survey results completed voluntarily;
- Execution of structured user interviews, both one-to-one and in group settings;
- Conduction of a longitudinal Diary Study with daily video submissions;
- Monitoring of anonymised data analytics and individual user tracking;

DiscoveryAfter going through this extensive research phase, we were able to list the key reasons impacting user retention:

Unfamiliar Territory

‍The initial app structure was based on the body's chakra points. Each playlist or album inside the app was named after one of the body chakras, meaning that the organization of the content in-app was theme-based and organized into subjective categories that caused confusion for users to interpret. This type of categorization went against the initial design guidelines previously set out (in the Rapitation project), since it made use of elements that are intrinsic to the Mindfulness universe, therefore demands an extent of familiarity with the topic.

Image taken from the 7chakrastore.

The Check In

Although the overall idea and function of the Check In was highly praised by our user base, they showed the need of expanding both the selection of emotions displayed on the Emotions Axis at the first step of the Check In and the list of themes at the 2nd step.

At the last step of completing the Check In, a reflective question would be randomly chosen for the user based on the topics selected on the previous screen. App users didn't seem to understand the purpose of that 3rd step, they questioned the relevance of the questions being asked at that screen and why the app would need to collect that kind of information from them.

Another aspect of the last Check In step was that no data inputted by users on that screen was used to feed the Session Recommendation Algorithm.

Locked Content

Initially, there was a content locking system in place, that meant that the user was only given access to the sessions inside the album recommended through the Check In.
‍
The data inputted by the user on their Check In answers are based on the punctual moment they are accessing the app, the way they feel at that point and the particular topics on their mind at that time, not on permanent or long-lasting facts.

This locking system led to frustration, because although users could see all the content available on the app (read session titles, descriptions and see their artwork) they weren't able to listen to them and were limited to one recommended album. Only after listening to all the sessions on the recommended album, they would be offered the option to do a Check In again and have a different album/ playlist given to them by the app.

iPhone X Screenshot showing content locked in-app.

IdeationFrom the information gathered from our customers and a deeper understanding on how the SPOKE app was performing at the first quarter of 2021, we were able to set out project guidelines to be followed in order to hit the goal of improving user retention in the near future.

SPOKE - Project Guidelines for the Ideation Phase

We delved into the Design Exploration, when we looked into how different content platforms on the market are usually structured – specifically those focused on audio media – with the intention to identify potential solutions that could be applied to our case. From this, wireframes were created in order to try out the initial ideas that later on evolved to interactive prototypes that were constantly shared with a few users who we maintained communication along the way so that these ideas could be iterated according to their feedback.

Because SPOKE is a small early stage fast-paced start-up, we often opt for the Pilot Methodology, this means that the business prefers to work with a launch-then-iterate approach: releasing features and solutions directly into the market first, so we can gather in-app usage data from a wider scope of users that will then inform the refinement of said releases as we iterate and improve on them.
Along the way we keep the user feedback loop open thanks to the feedback component on the user profile section of our app that allows customers to send direct feedback to us so that we're aware of: what things they approve, what issues they're running into and what we might have missed on. From that it's just a constant iteration process.

Delivery
Content Re-structure

We moved away from the Chakra theme based album names and decided to reorganize all the sessions into mood based categories, because although the sessions inside an album addressed the same theme – based around a body chakra –, the style of the tracks could vary greatly. In the case of the user listening to more than one track in one go, dissonant session styles would often affect their meditative state of flow and even put them off carrying on with the session at that point.

Updated content structure:

Besides restructuring the app's content architecture and modifying copy to be more accessible to a wider public, we added tags alongside each track in order to aid session discovery by providing information about the the Meditation techniques and themes covered on the track in a quick and easy to scan way.

We improved the app's Explore Section by displaying richer information about the tracks on the main browsing page, so that it would be easier for users to choose their sessions without wasting time sampling tracks beforehand or having to access each track's details (i.e. track's description or lyrics).

Iteration of the Check-In
‍
In order to define what extra themes and emotions to add to the Check In screens we ran online polls within our user base and selected the most mentioned ones. By maintaining communication with a small sample of our audience, it was possible to test our different solutions and gather feedback on these. Thanks to this, we noticed that by addressing user intentions in the Check In, the second Check In step asked for a different selection of themes according to the intention/goal chosen in order to provide a more natural interaction for the user.

Optimized Check In structure:

SPOKE - Check-In Iteration of 3rd Question

Flexibility is vital

Removing the content locking system meant making all released sessions/ tracks on the app available to all users. That way we expanded the possibility of exploration and session discovery since we allowed users to opt for their preferred approach of consuming content on the app, either passively: by completing a Check In to get a recommended session; or actively: by browsing through the Explore Section and manually choosing a session to listen. By doing so we ensured the meditation session to be the most relevant to and resonant with the user.

We opted for maintaining the existing UI that was working for the app's audience and adding or modifying a few elements where necessary, as keeping the minimal aesthetic was one of the requirements we set initially for the project. Other reasons for deciding on this direction were the available budget and amount of development work that would be involved in Deployment.

Existing design

Updated design

Enhances on the UI involved:
‍
- Removing the lock icons that wouldn't be used anymore due to the decision of lifting the content blocking mechanism;
‍
- Removing the play/ pause icon that appeared below every session's thumbnail and would more often than not cut off longer track titles. This consequently simplified the interface by making the thumbnail the play button for the session and avoided the appearance of repeating elements on the same screen;
‍
- Removing unnecessary or repeated information such as the ordinal numbers before the track's title and the album name below the thumbnail respectively, allowing for more white space on the screen. Since the quantity of sessions contained in each album isn't a numerous amount, the information of the position of the track inside the album doesn't really matter to users. Besides this, each album name already appears as a header on top of the screen, identifying the albums contained in the Explore area of the app, therefore there's no need to display the album's name anywhere else on the screen, if all tracks are contained in the same album.