How Duolingo reignited user growth

Jorge Mazal, former CPO of Duolingo, recounts how the company reversed a slide in daily active user growth and eventually grew DAU about 4.5x over four years. The journey began with failed attempts: a Gardenscapes-inspired moves counter and a referral program that did little to move the needle. Those failures pushed the team toward a more rigorous approach, building a user-bucket growth model that revealed current-user retention (CURR) as the lever with outsized impact on DAU. From there, leaderboards, disciplined push-notification optimization, and streak improvements became the main wins. The lessons matter because they show how to adapt borrowed features thoughtfully and how to choose a single North Star lever through modeling.

01Key takeaways

• Before borrowing a feature from another product, ask why it works there and whether the context will translate.
• Use historical user-state data and a bucket model to find the single lever with the largest compounding effect on your target metric.
• Protect fragile channels like push notifications and email by limiting volume, since over-testing causes users to opt out permanently.
• Adapt gamification mechanics by stripping unnecessary complexity, making participation automatic and frictionless.
• Look for big wins in existing features as well as new bets, since optimizing something like streaks can yield major retention gains.

02Key sections

Early gamification failure

The team copied a match-3 game's moves counter to add scarcity to lessons, but it had no effect on retention and split the team. The author concluded they had copied surface similarities without accounting for how the underlying context differed.

Referral program miss

A referral reward modeled on Uber's approach lifted new users by only 3%, because the best active users already held the premium plan that the reward offered. Context mismatch again drove the weak result.

Growth model and North Star

The team segmented users into mutually exclusive buckets and modeled movement between them using retention rates. Sensitivity analysis showed CURR had roughly five times the impact of the next lever on DAU, so it became the North Star.

Leaderboards, notifications, and streaks

Adapted leaderboards with leagues and frictionless opt-in drove large engagement gains. Notification testing was constrained by a rule to protect the channel, and streak features like streak-saver reminders produced further retention wins.

Results and reflections

CURR rose 21%, cutting daily churn among top users by over 40%, and the share of DAU with a 7-day-plus streak nearly tripled. The author stresses adapting ideas to your own context rather than copying them.

03From the post

“1. Growth inflections 2. How to be prepared for layoffs 3. How Coda builds product Subscribe to get access to these posts, and every post. I was at a small event a few months back where Jorge Mazal (former CPO of Duolingo) shared the story behind Duolingo’s growth reaccelerating. I was captivated. I’ve never seen a growth story like this before—4.5x growth for a mature product, driven by a small handful of product changes, rooted in an innovative growth model, and explained in such actionable detail. I asked Jorge if he’d be willing to share (and expand on) the story with a broader audience, and I’m so happy he agreed. Many products already look to Duolingo for inspiration, and I suspect this story will only increase that trend. Enjoy! Follow Jorge for more on LinkedIn and Twitter. I joined Duolingo as the Head of Product in late 2017. Duolingo was already the most downloaded education app in the world, with hundreds of millions of users, fulfilling its mission to “develop the best education in the world and make it universally…”

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.