Intuitive Today

Brief

Optimize the Today section where people get daily personalized recommendations based on genre, and likes from unlocked categories (TV & movies, Books, Podcasts).

Problem

The Today tab in the Likewise app is where people get daily personalized recommendations of their favorited categories based on genre, likes, and people you follow. The section was seeing sluggish retention numbers for day zero. Users who unlocked more than one category in their first day on the app were retained longer than users who only unlocked one category (TV & Movies, books & podcasts but current onboarding infrastructure could only support unlocking one category at a time. How could we surface unlocking multiple categories at the top of the funnel in the live product?.

Old Today Recommendation Screen
Personal Recommendation

My Role

I collaborated with another designer throughout the whole process. My main contribution was working on the user flows, prototype, and final delivery.

Research

Metrics

We scheduled a kickoff meeting with our product manager and UX researcher to learn more about the metrics and qualitive data. Our UX researcher pointed us to some pre-recorded user videos and we began going through them and noting user behavior to understand what roadblocks or frustrations they were having.

One constraint we had to be aware of throughout the process was keeping the scope to the Today section only. Since, the Today tab touched the onboarding flow and user profile, which were big flows, we had to be careful for the project not to balloon to something much bigger where we can't correctly measure it after launch.

Current User Journey Audit

We also created an audit of the current user journey and pointed out any abvious roadblocks.

User Journey Audit
Personal Recommendation

Competitive Research

Lastly, we used competitive comps to learn of any new patterns or learnings that we may of missed.

Competative Research
Personal Recommendation

Synthesis Information

We got together with the team and worked through all of the research to find common themes and to painpoints that currently existed. Here are the top things that we learned and work towards a solution:

This wider range of cross-category recommendations brought a higher value to people which also caused a higher engagement during each session. However, the number of people with multiple unlocked categories was low.

Audience

We decided to optimize the experience for two of our biggest use cases to align with business goals:

Audience
Personal Recommendation

Exploration

We quickly brainstormed a range of middle level fidelity ideas.

We focused on:

v1-User Education
Personal Recommendation
v2-User Onboarding
Personal Recommendation
v3-User Profile
Personal Recommendation
v4-Custom Cards
Personal Recommendation
v5-Header Nav
Personal Recommendation

Iterations

We worked with our partners to get the needed approval and narrowed it down to the option we called the "user loop". This option gave the user full control navigating through all of their recommendations, but also quickly navigate to a category.

Loop Diagram
Personal Recommendation
v6-User Loop Flow
Personal Recommendation

We wire-framed the approved flow to insure a consistent and foolproof user experience. Then we connected with the developer to walkthrough the scenarios and come up with a testing plan.

Full Flow
Personal Recommendation

We also worked through the details of the experience, such as:

Polish-Content Cards
Personal Recommendation
Polish-Empty States
Personal Recommendation
Polish-End Cards
Personal Recommendation
Polish-Transitions
Personal Recommendation
<

High Fidelity Deliverables

We flushed out further details in visual design and presented it to stakeholders and our development team for final handoff.

Visual Gradients
TV Experience
Visual TV Logo
TV Experience
Visual Iconography
TV Experience
Visual Materials
TV Experience
Visual Gradients
TV Experience
Visual TV Logo
TV Experience
Visual Iconography
TV Experience
Visual Materials
TV Experience

Prototype

In addition with transition annotations, I built a prototype in Principle to communicate with the development team and present to stakeholders.

Today Prototype

Post Launch

After a few weeks of launching the product, we saw a 15% increase in engagement and retention rose to an all time hight of 28%. Eventhough this was a successful solution, we found that users simply went through their full Today stack, but didn't go deeper in the product. We would continue to iterate and test cards to learn more.