For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Duolingo, Inc. (DUOL) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-23): Designation Watch · Gem Score +2 (−100…+100 Quality+Value blend) · Quality 66 · Value -50 · Sentiment -56 (timing only, not weighted) · Composite fair value $141.47 vs $134.81 at analysis
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
Duolingo, Inc.
DUOL NASDAQDuolingo, Inc. is a technology company that operates a global digital learning platform focused primarily on language education. The company offers the Duolingo language learning app, which uses gamified lessons, adaptive learning technology, and data-driven personalization to help users build reading, writing, listening, and speaking skills across numerous languages. Its product portfolio also includes Super Duolingo, a premium subscription offering with enhanced features, the Duolingo English Test, an online, AI-driven assessment of English proficiency used by educational institutions and organizations, Duolingo for Schools to support classroom instruction, as well as Duolingo ABC for early literacy and Duolingo Math for numeracy skills. Duolingo, Inc. generates revenue from time-based subscriptions, in-app advertising, the Duolingo English Test, and in-app purchases, serving individual learners, schools, and institutional customers worldwide. Founded in 2011 and headquartered in Pittsburgh, Pennsylvania, the company plays a prominent role in the education technology and software application industry by providing accessible, mobile-first learning experiences at scale.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 8.57
Total Equity: $1.35B
Shares: 48,308,000
Total Debt: $0.00
Cash: $1.04B
EBITDA: $149.96M
Total Debt: $0.00
Cash: $1.04B
Revenue: $1.04B
Revenue: $1.04B
Revenue: $1.04B
Total Equity: $1.35B
Tax Rate: -127.0%
Equity: $1.35B
Total Debt: $0.00
Cash: $1.04B
Current Liabilities: $551.15M
Long-Term Debt: $0.00
Total Debt: $0.00
Total Equity: $1.35B
Shares: 48,308,000
Shares: 48,308,000
CapEx: -$18.10M
Shares: 48,308,000
Stock Price: $134.81
Net Income: $414.07M
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 2, 2026 2:26pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $250.8M | $369.5M | $531.1M | $748.0M | $1.0B |
| Cost of Revenue | $69.2M | $99.4M | $142.1M | $203.6M | $288.1M |
| Gross Profit | $181.6M | $270.1M | $389.0M | $544.4M | $749.5M |
| Operating Expenses | $241.6M | $335.3M | $402.3M | $481.8M | $613.9M |
| Operating Income | -$60.0M | -$65.2M | -$13.3M | $62.6M | $135.6M |
| Net Income | -$60.1M | -$59.6M | $16.1M | $88.6M | $414.1M |
| EBITDA | -$57.3M | -$60.3M | -$6.2M | $73.4M | $150.0M |
| EPS | $-2.57 | $-1.51 | $0.39 | $2.04 | $9.05 |
| EPS (Diluted) | $-2.57 | $-1.51 | $0.35 | $1.88 | $8.57 |
Balance Sheet (Annual)
Last updated: Aug 2, 2026 2:02pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $553.9M | $608.2M | $747.6M | $785.8M | $1.0B |
| Total Current Assets | $619.3M | $697.2M | $897.8M | $1.1B | $1.4B |
| Total Assets | $661.3M | $747.3M | $954.0M | $1.3B | $2.0B |
| Current Liabilities | $119.1M | $181.8M | $277.4M | $422.2M | $551.1M |
| Long-Term Debt | — | — | — | — | — |
| Total Liabilities | $148.3M | $205.3M | $298.5M | $477.2M | $645.2M |
| Total Equity | $513.1M | $542.1M | $655.5M | $824.6M | $1.3B |
| Retained Earnings | -$170.9M | -$230.5M | -$214.4M | -$125.8M | $288.2M |
Cash Flow (Annual)
Last updated: Aug 2, 2026 2:26pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | $9.2M | $53.7M | $153.6M | $285.5M | $387.8M |
| Capital Expenditure | -$3.6M | -$5.6M | -$3.2M | -$12.1M | -$18.1M |
| Free Cash Flow | $5.6M | $48.1M | $150.4M | $273.4M | $369.7M |
| Acquisitions (net) | $0 | -$4.1M | $0 | -$6.6M | -$33.1M |
| Net Debt Issued / (Repaid) | — | — | — | — | — |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | $-868,000 | $0 | $0 | — | — |
| Net Change in Cash | $433.4M | $54.3M | $142.2M | $38.2M | $250.6M |
Growth Trends (YoY %)
Last updated: Aug 2, 2026 2:26pm (21d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | +47.3% | +43.7% | +40.8% | +38.7% |
| Gross Profit Growth | +48.7% | +44.0% | +39.9% | +37.7% |
| Operating Income Growth | -8.6% | +79.7% | +572.1% | +116.6% |
| Net Income Growth | +0.9% | +127.0% | +451.3% | +367.5% |
| EBITDA Growth | -5.3% | +89.8% | +1,291.6% | +104.2% |
Deep Analysis
Risk : Reward — if the last quarter repeats
Live · as of 2026-08-19 08:49Recovery pays +127%; another quarter like the worst recent one costs 0%. Ratio 385.3:1.
| Case | Growth | Margin | Fair value | vs price ($134.81) |
|---|---|---|---|---|
| Bull — recovery | +44% | 35.0% | $306.21 | +127% |
| Base — stabilizes | +29% | 35.0% | $206.39 | +53% |
| Bear — keeps slipping | +15% | 30.5% | $119.14 | -12% |
| Stress — last quarter repeats | +18% | 30.9% | $134.37 | -0% |
Narrative Economics
market-narrative step).
Claude Reading
The raw trajectory here is genuinely impressive and I want to establish that before dissecting the models. Revenue went from $531M (2023) to $748M (2024) to $1.04B (2025) — that's 41% then 39% growth at the billion-dollar scale, which is rare. Q1 2026 at $292M annualizes near $1.17B implying growth is decelerating to ~30-33% YoY (Q1'25 was $230.7M, so 26.6% YoY — actually sharper deceleration than the narrative admits). Operating margin swung from -2.5% (2023) to 13.1% (2025), FCF hit $370M on $1.04B revenue (35.6% FCF margin), and the balance sheet carries $1.04B cash against zero debt. At $134.81 and a $6.28B market cap, EV is roughly $5.24B — that's 5.0x EV/sales and ~14x EV/FCF on trailing numbers. That is not a demanding multiple for a business compounding revenue at 30%+ with 35% FCF margins.
The reported P/E of 15.7x is misleading and every downstream model appears to have swallowed it. 2025 net income of $414M includes the Q3'25 print of $292M NI on $272M revenue — a 107% margin that is almost certainly a deferred tax asset release (valuation allowance reversal), not operating earnings. Strip that out and normalized 2025 NI is closer to $165M, putting the "real" P/E near 38x, not 16x. The synthesis DCF fair value of $263-353 (implying +95% to +162% upside) is aggressive and I don't buy it at face value — it likely extrapolates the CAGR trends without haircut for the visible deceleration (Q1'26 rev growth 26.6% YoY vs 38.7% "recent" figure the momentum module cites, which uses stale comps). A more defensible fair value using 25x normalized FCF of ~$400M forward = $10B EV, or roughly $215/share — still meaningful upside from $135 but half what the synthesis claims.
The contrarian case deserves airtime the models mostly waved off. The bear thesis on GenAI disruption is not obviously wrong: language learning is the single most obvious LLM use case, and while Duolingo has integrated AI into its own product (Max tier), the moat is gamification/habit, not content. If ChatGPT's voice mode gets meaningfully better at conversational tutoring at $20/month vs Duolingo Super at ~$7, the freemium-to-paid conversion funnel gets squeezed from the top. Second, Q1'26 revenue of $292M vs Q4'25 of $283M is only 3.2% sequential — that's a real deceleration signal masked by YoY optics. Third, the insider data shows only awards and small option exercises, no meaningful open-market buying — the "neutral" tag is generous; there's zero conviction signal from management at $135 after a 71% drawdown. Fourth, "cult-favorite" narrative durability being flagged as only moderate matters — when the story cracks on a consumer subscription name, multiple compression can be violent.
Net: I partially agree with the synthesis direction (undervalued) but strongly dissent on magnitude. The models are anchoring on a fake P/E and a stale growth rate. Real setup is a high-quality compounder trading at ~14x EV/FCF with decelerating-but-still-30%+ growth, priced for GenAI-driven terminal decline that hasn't shown up in numbers yet. Fair value is $180-215, not $263-353. At $135 that's still 35-60% upside with a clean balance sheet as downside protection, which is a good risk/reward — but calling this a screaming buy at 2x current price requires ignoring both the sequential deceleration and the AI disruption risk that is genuine, even if overblown. Starter position warranted; the table-pounding case the market-forces module makes is overconfident. Watch Q2'26 for whether YoY growth stabilizes in the high-20s or breaks below 25% — that's the number that decides whether this is a 20x FCF story or a 12x FCF story.
GPT Reading
What jumps out first is that the stock looks cheap only if you take reported 2025 earnings at face value. Duolingo’s annual net income of $414.1M on $1.04B of revenue implies a 39.9% net margin, but the quarterly sequence gives away the distortion: 2025-09-30 shows $292.2M of net income on $271.7M of revenue, a 107.5% margin. That is not the economics of a language-learning app; it is almost certainly a one-off below-the-line gain or tax/event-driven accounting benefit. Strip that out and the business is still good, but different: the surrounding quarters are earning roughly $35M-$45M on $230M-$292M of revenue, i.e. mid-teens net margins. On that cleaner run-rate, Duolingo is not trading at 15.7x earnings; it is closer to the low-30s on an underlying annualized profit base around $165M-$180M. That matters because a lot of the “obvious bargain” case rests on a misleading P/E.
The underlying operating story is still strong. Revenue has compounded from $250.8M in 2021 to $1.04B in 2025, and even the latest quarter at $292.0M is up 26.6% from $230.7M a year earlier. Gross margin of 72.2% and operating margin of 13.1% show real software-like economics, and the cash conversion is excellent: $387.8M of operating cash flow and $369.7M of free cash flow on $1.04B of sales is extraordinary, though I would also be careful not to assume all of that is pure recurring owner earnings without looking at working capital and stock comp details. The balance sheet is pristine with $1.04B of cash and no debt, so enterprise value is only about $5.24B, or roughly 5.0x trailing revenue. For a consumer software platform still growing 25%-40%, that multiple is not demanding. The market clearly does not believe current growth and monetization can persist for very long.
That skepticism is not crazy. The quarterly revenue cadence itself suggests deceleration from the 40%+ narrative: $178.3M, $192.6M, $209.6M, $230.7M, $252.3M, $271.7M, $282.9M, $292.0M. Absolute dollars are still rising, but the slope has moderated recently, and the jump from $282.9M to $292.0M quarter-over-quarter is modest. This is the key contradiction with the hyper-bull valuation outputs: yes, EV/revenue at 5.1x is low versus premium software, but Duolingo is not classic B2B SaaS with contractual revenue durability and deep enterprise switching costs. It is a consumer subscription product in a category where AI can narrow feature advantages quickly. At $134.81, I think the stock is discounting a real risk correctly: that today’s excellent engagement engine can remain popular while growth settles faster than bulls expect. My read is that the shares are somewhat undervalued, but nowhere near the “DCF says 160% upside” conclusion.
The best case against my caution is simple and strong. Even excluding the anomalous quarter, Duolingo has crossed into consistent profitability, with quarterly net margins generally between 12% and 18%, while revenue keeps compounding above 25%. A debt-free company with over $1B of cash, 72% gross margins, and nearly $370M of free cash flow should arguably not trade at just 5x revenue. If management can keep annual growth around 25%-30% for even three more years while lifting operating margin from 13% toward 20%, today’s price would look extremely cheap. The market may also be over-penalizing “AI disruption” when AI could instead improve lesson quality, personalization, and conversion. On that view, the stock deserves a materially higher multiple, especially because the business has already demonstrated that scale brings sharp operating leverage.
What would change my mind is straightforward. If the next few quarters show revenue re-accelerating back above 30% year over year while underlying net income stays in the $45M+ range without one-time items, I would move more bullish and support a valuation north of $170. Conversely, if quarterly revenue stalls in the low-$300M range, year-over-year growth slips toward 20%, or margins fall back below 10% despite scale, then even $135 is not cheap and fair value is closer to $110-$120. For now I see a high-quality company, a messy reported earnings base, and a stock that is modestly but not dramatically mispriced to the upside.
Grok Reading
The numbers tell a story of a business that has completed its profitability transition with unusual cleanliness and is now being priced as if that transition never happened. Duolingo generated $1.04B in 2025 revenue, up 39% from $748M, with free cash flow of $370M — a 35.6% FCF margin on a consumer freemium model that still has no debt and $1.04B in cash against a $6.28B market cap. Operating income scaled from −$13M in 2023 to $136M in 2025 while gross margin held at 72%. Sequential quarterly revenue is still climbing ($231M → $252M → $272M → $283M → $292M), and trailing FCF alone puts the stock at roughly 17x cash earnings. That is not the multiple of a broken growth story; it is the multiple of a company the market has decided will stop compounding.
The distortion that matters most is the Q3 2025 net income spike to $292M (107% margin), which pulls reported annual NI to $414M and manufactures the headline 15.7x P/E. Strip that one-time item and normalized earnings land nearer $160–170M, implying a mid-to-high 30s P/E — still undemanding next to 30%+ top-line growth, 13% operating margins that are still expanding, and FCF that already exceeds operating income because capex is a trivial $18M. The rule-based “mature earner” label is simply wrong: a business printing nearly 40% revenue CAGR and 57% FCF CAGR is not mature. What the market is actually pricing is the AI-commoditization narrative — that ChatGPT-style tutors will hollow out the subscription base — and it has compressed EV/revenue to 5.0x and the stock 60%+ below the internal DCF anchors. That fear is legible but overstated: Duolingo’s moat is behavioral addiction and distribution, not pure content, and the same AI wave that threatens generic tutoring also lowers Duolingo’s content-production cost curve.
The strongest case against this read is the deceleration already visible in the print. Q1 2026 revenue of $292M is only ~27% above the year-ago $231M, a clear step-down from the 35–40% zone, and the secondary signals flag “decelerating” with only moderate confidence. If paid-subscriber growth and ARPU stall while the English Test fails to become a meaningful second engine, 17x FCF becomes fair rather than cheap and the EV/EBITDA of 35x starts to look expensive for a mid-teens grower. Skeptics will also note that net margin and ROIC figures are polluted by the one-time gain, that freemium conversion still leaves monetization thin on a per-user basis, and that a cult-favorite stock which has already drawn down 71% from peak can stay unloved for a long time if the growth narrative keeps cracking. Those points have weight; I simply weigh the fortress balance sheet, the already-realized operating leverage, and the absolute FCF yield more heavily than the deceleration slope at current prices.
I would flip to fairly valued or overvalued if two consecutive quarters show revenue growth sustaining below 20% with FCF margins compressing under 25%, or if management guides to a structural slowdown in Super conversion. Conversely, a re-acceleration above 35% driven by English Test scale or Math/ABC monetization, or a clear print of 15%+ operating margins with stable growth, would push conviction materially higher.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
Revenue has compounded from $250.8M in 2021 to $1.04B in 2025 (roughly 4x in four years) while gross margin held remarkably steady around 72-73%. The more important story is operating leverage: operating margin swung from -23.9% in 2021 to +13.1% in 2025, net income went from -$60.1M to $414.1M, and FCF scaled from $5.6M to $369.7M. OCF/NI of 2.53x and accruals at -10.8% of assets confirm the earnings are cash-backed, not accounting-driven. Altman Z of 7.32 places it deep in the safe zone. The balance sheet is a fortress: $1.04B liquid cash, zero net debt, and the business is self-funding.
Verify before trusting this (6)
- Composition of 2025 net income - was there a large deferred tax asset valuation-allowance release inflating NI relative to FCF?
- 10-K disclosure on daily active users, paid subscriber growth, and retention cohorts to assess engagement durability
- Customer/revenue concentration by geography and by Super Duolingo vs Duolingo Max tiers
- Details of SBC grant structure and expected future dilution from unvested RSUs/options
- Management commentary on competitive response to generative-AI language tools
- Any related-party or unusual transactions behind the M-score -1.7 flag
The composite fair value of $263.66 looks generous once you audit its parts. The anchored-PE of $422.99 leans on a 2025 GAAP earnings print that the quality lens itself flags as juiced by a likely non-recurring item, so it should be heavily discounted. The EPV floor of $39.85 says the current run-rate cash economics alone justify well under a third of today's price - meaning most of the $134.81 is growth optionality. That leaves the DCF at $295.89 as the only serious anchor, and even that requires Duolingo to sustain the recent operating-leverage inflection (OpM -24% to +13%, FCF to $370M) for many more years while user growth is already decelerating per the bear case.
Verify before trusting this (5)
- Whether the 2025 GAAP net income includes a one-time tax benefit or valuation-allowance release (would gut the PE anchor)
- DAU and paid-subscriber growth trajectory in the next two prints - deceleration curve
- English Test revenue disclosure and institutional adoption metrics
- Super ARPU trend and pricing power evidence
- Stock-based comp as a % of revenue vs the reported FCF
The macro tape is basically neutral-to-mildly-constructive (VIX 16, S&P only 1.6% off highs), so this is not a risk-off mauling. With beta 0.88, DUOL isn't especially exposed to the tape anyway. The pressure here is idiosyncratic and narrative-driven: a cult-favorite growth story whose intensity is still strong but whose durability is visibly eroding as user growth decelerates and the AI-tutor / language-app cohort loses its 2024 shine. The gap between a $353 DCF and a $135 tape tells you the market has stopped paying the cult multiple. That is a sentiment de-rating, not a fundamentals collapse. Recent tape action confirms the press: a -4.7% single-day drop on a day the broader market rose is textbook single-name selling into an unsupportive narrative. News flow is quiet but unfavorable in tone, and there is no fresh catalyst (Math, ABC, English Test traction) landing to re-ignite believers. Net: moderate headwind. Not a Strong Headwind because the tape isn't hostile and the cult base still provides a floor of demand; but clearly negative pressure until the growth-reacceleration story gets a new proof point.
Verify before trusting this (4)
- Next earnings print - DAU/MAU growth rate and Super subscriber conversion, the two numbers that would flip the narrative
- Any English Test institutional adoption announcement that would resurrect the moat story
- Sell-side target revisions - is the analyst community cutting or holding into the print
- Whether the consumer-AI cohort as a group reclaims a bid (would lift DUOL passively)
This lens hasn't been run for this ticker yet.