Loading...
December 15, 2024 • 1hr 28min
Behind the product: Duolingo streaks | Jackson Shuttleworth (Group PM, Retention Team)
Lenny's Podcast: Product | Growth | Career
Jackson Shuttleworth is a Group Product Manager at Duolingo where he leads the retention team and oversees their streak feature. The streak feature, which tracks consecutive days of learning, has been instrumental in Duolingo's growth to nearly 600 million users and a $14 billion valuation. In this conversation, Jackson shares the detailed journey of how this feature evolved through extensive experimentation and iteration.
Behind the product: Duolingo streaks | Jackson Shuttleworth (Group PM, Retention Team)
0:00
-0:00
Key Takeaways
- Streaks are Duolingo's most impactful feature - Other than the core learning product, streaks have been the biggest driver of growth and retention, helping build Duolingo into a $14B company
- Focus on clarity over complexity - Simple, clear mechanics work better than complex systems. The team constantly works to make streak features easily comprehensible
- Test extensively but thoughtfully - The team has run over 600 experiments on streaks in 4 years, but with clear hypotheses and strategic intent
- Balance flexibility and perfection - Features like Streak Freeze provide needed flexibility while Perfect Streak rewards consistency
- Protect the "sanctity of the streak" - Having dedicated people focused on maintaining the feature's integrity helps prevent over-optimization that could diminish its value
Introduction
Jackson Shuttleworth is a Group Product Manager at Duolingo where he leads the retention team and oversees their streak feature. The streak feature, which tracks consecutive days of learning, has been instrumental in Duolingo's growth to nearly 600 million users and a $14 billion valuation. In this conversation, Jackson shares the detailed journey of how this feature evolved through extensive experimentation and iteration.
Topics Discussed
Origins and Evolution of the Streak Feature (09:58)
The streak feature has been part of Duolingo since launch, but has evolved significantly over time. Initially, it was based on XP (experience points) goals, but was simplified to track daily lessons completed.
- Key early change: Moving from XP-based goals to "complete one lesson" dramatically improved engagement
- Failed experiment: Trying to make streaks even easier by counting partial lessons didn't work - users didn't connect with it
- "The unit of use for Duolingo is doing a lesson. If what we care about is users coming back every day and doing a lesson, then it doesn't hurt us to make our streak focus on just do one unit versus multiple" - Jackson Shuttleworth
The Impact of Clear Communication (33:07)
One of the biggest learnings has been the importance of making the streak mechanic easily understandable for users across different cultures and backgrounds.
- Simple copy changes drove massive engagement increases
- First winning experiment: Adding 8 words explaining how streaks work
- "Start a day to extend your streak, but miss a day and it resets" - This simple copy drove over 10,000 additional daily active users
- Cultural considerations matter - the "flame" metaphor doesn't resonate equally across all regions
Balancing Flexibility and Engagement (46:59)
The team discovered the importance of providing flexibility while maintaining the feature's core value.
- Streak Freeze allows users to maintain streaks despite missing days
- Testing revealed two freezes worked better than one or three
- More flexibility is needed early in the streak journey (first 7 days)
- Less flexibility is appropriate for longer streaks where habits are established
- "If you're going to miss a day, I'd rather you come back having missed that day to an intact streak. But if you don't have to miss a day, much rather you don't" - Jackson Shuttleworth
Notification Strategy (54:08)
Notifications play a crucial role in maintaining streaks and user engagement.
- Practice reminder sent 23.5 hours after last practice
- Streak saver notification at 10pm for users at risk of losing streaks
- Users view these as helpful rather than spam because they're protecting something valuable
- Timing based on actual user behavior outperforms user-selected times
Team Operations and Experimentation (1:04:47)
The team's approach to product development emphasizes rapid experimentation with clear strategic intent.
- Metric-focused teams rather than feature-focused teams
- Strong process orientation to manage hundreds of experiments
- Start with simple V1s to test core hypotheses
- Use JIRA automation and detailed roadmaps to coordinate cross-functional work
- "We've run in the last four years over 600 experiments on the streak. So every other day effectively we're running an experiment" - Jackson Shuttleworth
Building Effective Streak Features (1:18:57)
Key principles for implementing successful streak mechanics in other products:
- Core product value must exist first - streaks amplify engagement but can't create it
- Match cadence to natural usage patterns - daily may not be right for every product
- Make streaks highly visible - users care because Duolingo shows it cares
- Test extensively with real users rather than theorizing
- "The reason why users care about your streak so much is because you care about your streaks so much" - Bing Gordon, Board Member
Conclusion
The success of Duolingo's streak feature demonstrates how thoughtful implementation of behavioral mechanics can drive massive business impact. The key is maintaining a balance between engagement optimization and feature integrity while constantly experimenting and iterating based on user behavior and feedback. The lessons learned - from the importance of clarity to the value of flexibility - provide a valuable framework for other products looking to implement similar features.
