Improving User Onboarding with Favorite Channels for iOS

How I Enhanced Early User Engagement and Powered Recommendations by Personalizing Onboarding

Background & Context

Philo faced a cold-start problem: new users often struggled to discover relevant content when first entering the product. This lack of personalization led to a suboptimal early experience, particularly in the guide and home screen, which negatively impacted engagement and retention rates. To address this, we designed a new onboarding flow allowing users to favorite channels or save shows immediately after subscribing.

Key Goals:

1. Seed the Recommendations System: Use favorite channels to populate personalized rows on the home screen, surfacing relevant shows and movies earlier.
2. Enhance the Guide Experience: Automatically highlight favorited channels at the top of the guide, reducing friction in content discovery.
3. Improve First and Second Payment Rates: Personalization during onboarding was hypothesized to lead to better early engagement and retention.

Research & Insights

We analyzed user behavior and gathered insights:

• Cold-Starts: New users often lacked guidance on where to start, leading to lower engagement on the home screen and longer times to discover relevant content.
• High Drop-Off Rates: Users had difficulty navigating the guide, often scrolling far down to find the channels they liked, which increased friction and frustration.

Key Hypothesis:

By allowing users to select their favorite channels during onboarding, we would:

• Seed the recommendations engine faster.
• Improve content discovery and early engagement.
• Lead to higher first and second payment rates.

Ideation & Design Process

Users were prompted to select their favorite channels immediately after subscribing.

Features Implemented:

1. Favorite Channels Screen:
   Users could choose channels from a visually engaging grid.
   Channels selected here were automatically favorited in the guide.
2. Personalized Home Hero Row:
   A new row on the home screen displayed top shows and movies from favorited channels.
3. Collaborations:
   Partnered with Engineering & Data to ensure favorite inputs were integrated into the backend, powering:
   • The guide (flagging channels as “favorited”)
   • Linking recommendations
   • Define and measure success metrics

Launch & Results

The redesigned onboarding flow demonstrated measurable success:

Primary Metric:

• First Payment Rate: Maintained at 31.8% in the treatment group with no drop-off compared to the baseline, validating that the intervention did no harm to initial conversions.

Engagement Metrics:

• 41.9% Increase in Channels Favorited: The number of favorited channels rose significantly, showing the onboarding flow’s success in driving personalization.
• 7.93% Increase in Playback Sessions: Improved engagement as users were more likely to start playback sessions after favoriting channels.
• 5.25% Increase in Live Channel Playback Sessions: More users accessed live content directly from the guide, reducing friction in content discovery.

Guide and Recommendations Impact:

• Favorited channels were prominently flagged in the guide, leading to quicker navigation and better content discovery.
• The personalized home hero row received increased interaction, reflecting the relevance of curated content.

Reflections & Learnings

What Worked:

• The introduction of personalization early in the user journey created a smoother and more engaging experience.
• Users favored channels without drop-offs in first payments, showing that the added step provided value without adding friction.
• Collaborating with Engineering and Data teams ensured seamless backend integration and accurate measurement of success.

What Could Be Improved:

• Extending the experiment to measure second payment rates and long-term retention would provide a clearer picture of the intervention’s broader impact.
• Incorporating additional personalization, such as favorite genres or shows, could further enhance the recommendations system.