Duolingo's Cheap P/E Hides a Deliberate Trade-Off

Duolingo's stock looks cheap. The company behind the green owl trades at about 16 times its trailing earnings, it grows revenue in the high teens, and its gross margin is over 72%. For a business that still has an open growth story, that reads as a bargain — especially after the shares fell below $90 this year, having traded at $353 as recently as late 2025. Most retail screens stop right there and say "cheap."

The trailing number is a lie. In the quarter ended September 2025, Duolingo released a valuation allowance it had long held against its deferred tax assets, an accounting move that produced a one-time income-tax benefit of $222.7 million. Net income for that single quarter reached $292.2 million, or $5.95 a share. A year earlier the same quarter earned $0.49 a share. That windfall is most of what the cheap multiple is built on, and it will not repeat. Take it out, and the stock trades at roughly 42 times forward earnings — an expensive company, not a cheap one. The market is not underpricing the business; it is pricing a bet.

What makes the bet worth looking at is how unusual the underlying economics are. Duolingo is a rare consumer-internet company that makes a lot of money without spending a lot of money. The whole model runs on a single idea: users pay in one of two currencies. They pay with their time, in the form of ads on the free tier, or they pay with cash, in the subscription that now makes up around four-fifths of revenue. The product is built to capture that time — streaks, leaderboards, reminders — and the numbers say it works. Current-user retention hit a record 84% in the second quarter of 2026, and roughly 80% of its users have always arrived organically rather than being bought. It has passed a billion lifetime downloads. Measured by capital, the machine is almost laughably efficient: free cash flow margin around 35%, capital spending of only about $33 million over the trailing year, and more cash than debt on the books. This is the rare company that is, in the plainest sense, default-alive.

Now the part that makes it interesting. In 2025 the company decided to spend that efficiency. Growth had begun to wobble, and when Duolingo reported the third quarter, daily active users came in at 50.5 million, beneath what Wall Street had forecast. Founder and CEO Luis von Ahn said the company was now tilting its experiments "much more towards user growth," and the guidance it gave scared the market: the shares fell about a quarter in a single day, the worst drop in the stock's public history. The market read it as a sign the hit was maturing. Von Ahn's own managers were saying something similar — that they would deliberately trade near-term money for a bigger audience.

The second quarter of 2026 shows the trades now being executed. Daily active users reached 58.7 million, up 23% from a year earlier, while revenue grew only 18%, to $298.5 million. The company guided 2026 revenue growth to 15–18% and bookings growth to just 10–12%, a sharp step down from the 39% revenue growth of 2025. Users, in other words, are now growing faster than the money they generate. Management is reinvesting its fat margins into the free product rather than pushing prices, betting that the flywheel compounds to a target of 100 million daily users by 2028 and converts a far larger base to paid later. That is a thesis you can take or leave, but it is a real one, and it is not what the trailing P/E advertises.

Take the thesis seriously and one number deserves your attention before any other: the gap between daily-user growth and revenue growth. As long as users outrun money, the bet is working — Duolingo is buying a bigger audience with cash it can actually afford to spend, and the time-and-money engine is intact. The moment those two converge or invert, while the company is still giving up margin to grow, the efficient machine stops looking efficient and the 42-times-forward multiple has no cushion at all.

We already know what a wobble costs here. The stock went from $353 down to below $90 in about five months when growth simply did not meet a forecast, and it has clawed back toward $145 since. A company priced for re-acceleration that is guiding to deceleration is a fragile object. The product earns the loyalty it gets. The price only deserves as much as the growth it buys.