
Hold, $170 target: the Q3 rebound in users is in the price, leaving 5% upside
We initiate on Duolingo at Hold with a $170 target, 5.4% above the $161.27 close. Duolingo runs a language-learning app with 153.1m monthly users and monetises 13.1m subscribers out of 64.9m daily users. DAU growth re-accelerated to 29% in Q3 and the shares rose 26% the next day; our DCF says that is now in the price.
- Users are the engine: DAUs grow 29% to 64.9m; we reach 93.8m by FY28E. The DAU path alone spans $113.3 to $171.1 a share (Debate 1).
- Margin is the cost: adj. EBITDA margin dips to 26.5% in FY26E. We rebuild 1pt a year to 32.5%; 1pt is worth $5.29 a share (Debate 2).
- The price pays for it: $161.27 implies 4.0% terminal growth, we use 3.0%. Base DCF $152.2; EV/sales cross-check $175.8.
What the market is missing. Little near term: we are 2.8% above FY27E revenue consensus ($1,417m vs $1,379m) and 3.5% above on EPS; the dispute is how long 20%+ DAU growth lasts.
Catalysts. 11 Jan 2027 Apple v. Epic hearing on app-store fees; Feb 2027 Q4 results and first FY27 guide (we need margin guide of 27.5%+); May 2027 Q1 subscribers.
Valuation and target. 10-year DCF after stock pay, 9.75% WACC, 3.0% growth: bear $78.3, base $152.2, bull $205.8 (20/50/30%), weighted $153.5, rolled to $170. Upside to downside 0.9x.
Risks. AI tutors erode the free app (DAU bear −$38.9); margins stuck near 26% (−$26.9). Against us: a 1pt lower discount rate gives $193.4, a Buy.
| Year end 31 Dec, US$ | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Revenue ($m) | 748 | 1,038 | 1,222 | 1,417 | 1,664 |
| Growth y/y | 40.8% | 38.7% | 17.7% | 16.0% | 17.4% |
| Adj. EBITDA ($m) | 192 | 306 | 324 | 390 | 474 |
| Margin | 25.7% | 29.5% | 26.5% | 27.5% | 28.5% |
| Net income ($m) | 89 | 414 | 137 | 179 | 234 |
| EPS, diluted ($) | 1.97 | 8.88 | 2.69 | 3.46 | 4.42 |
| EPS growth y/y | n/a | +351% | −70% | +28.4% | +27.9% |
| Free cash flow ($m) | 264 | 360 | 379 | 426 | 541 |
| At $161.27 | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| P/E | 81.9x | 18.2x | 59.9x | 46.7x | 36.5x |
| EV/adj. EBITDA | 35.0x | 21.9x | 20.7x | 17.2x | 14.1x |
| P/B | 9.9x | 6.1x | 5.3x | 4.3x | 3.5x |
| FCF yield | 3.2% | 4.4% | 4.6% | 5.2% | 6.6% |
| Dividend yield | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| ROE (year-end equity) | 10.7% | 30.7% | 8.8% | 9.3% | 9.9% |
| Net debt/EBITDA | −4.6x | −4.2x | −4.8x | −5.1x | −5.3x |
| P/E at our target ($170) | 86.3x | 19.1x | 63.1x | 49.2x | 38.4x |
Source: Company filings (A); Penmark Research model, 5 Nov 2026 (E). FY24A and FY25A EPS from filings (FY25A includes a $222.7m one-time Q3 2025 tax benefit). Multiples at the 4 Nov 2026 close; only the last row is at our target. Shaded = estimates. Consensus in Exhibit 10; peers in Exhibit 11.
Base-case value is $152.2 a share at 30 Sep 2026 and $170 in 12 months: +41% to bull, −46% to bear
Source: Yahoo Finance daily closes to 4 Nov 2026; Penmark Research estimates (DCF values rolled forward at the 9.75% cost of equity to 4 Nov 2027)
Management plan on time: 100m DAUs in Q4 2028, penetration back to 23%, +3% ARPU, margin to 35%.
Guidance delivered; DAU growth 20% then 15%, ~94m DAUs end-2028, penetration to 22%, margin to 32.5% by FY32.
AI tutors erode the free-app moat: DAU growth ~3% by FY30, penetration 18.5%, no price rises, margin 26–27%.
| Income statement, $m | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Revenue | 748 | 1,038 | 1,222 | 1,417 | 1,664 |
| Gross profit | 544 | 749 | 887 | 1,013 | 1,190 |
| Adj. EBITDA | 192 | 306 | 324 | 390 | 474 |
| SBC-related costs | 120 | 149 | 171 | 191 | 216 |
| EBIT (GAAP) | 63 | 136 | 134 | 177 | 233 |
| EBIT margin | 8.4% | 13.1% | 11.0% | 12.5% | 14.0% |
| Net income | 89 | 414 | 137 | 179 | 234 |
| Diluted EPS ($) | 1.97 | 8.88 | 2.69 | 3.46 | 4.42 |
| Cash flow, $m | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Total bookings | 871 | 1,158 | 1,297 | 1,513 | 1,786 |
| Operating cash flow | 286 | 388 | 416 | 461 | 583 |
| Capex incl. software | −21 | −27 | −37 | −35 | −42 |
| Free cash flow | 264 | 360 | 379 | 426 | 541 |
| FCF margin | 35.3% | 34.7% | 31.0% | 30.0% | 32.5% |
| D&A | 11 | 14 | 17 | 21 | 25 |
| Non-cash SBC | 110 | 137 | 162 | 182 | 206 |
| Deferred revenue | 373 | 496 | 572 | 667 | 789 |
| Balance sheet, $m | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Cash and investments | 878 | 1,276 | 1,560 | 1,986 | 2,527 |
| Accounts receivable | 129 | 163 | 183 | 213 | 250 |
| Total assets | 1,302 | 1,992 | 2,308 | 2,764 | 3,340 |
| Deferred revenue | 373 | 496 | 572 | 667 | 789 |
| Total liabilities | 477 | 645 | 751 | 846 | 983 |
| Stockholders’ equity | 824 | 1,347 | 1,557 | 1,918 | 2,357 |
| Financial debt | 0 | 0 | 0 | 0 | 0 |
| Net cash | 878 | 1,276 | 1,560 | 1,986 | 2,527 |
| Per share and KPIs | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Diluted shares (m) | 44.9 | 46.6 | 50.8 | 51.8 | 52.9 |
| Diluted EPS ($) | 1.97 | 8.88 | 2.69 | 3.46 | 4.42 |
| EPS growth y/y | n a | +351% | −70% | +28.4% | +27.9% |
| Q4 average DAUs (m) | 40.5 | 52.7 | 68.0 | 81.6 | 93.8 |
| DAU growth y/y | 50.6% | 30.1% | 29.0% | 20.0% | 15.0% |
| Paid subscribers (m) | 9.5 | 12.2 | 13.6 | 15.9 | 18.3 |
| Subscribers / DAUs | 23.5% | 23.1% | 20.0% | 19.5% | 19.5% |
| Revenue / subscriber ($) | 75 | 81 | 82 | 84 | 85 |
Source: Company filings (A: 10-K, shareholder letters); Penmark Research model, 5 Nov 2026 (E). Shaded = estimates. Adj. EBITDA adds back SBC-related costs (company definition). FY25A net income includes a $222.7m one-time tax benefit. FY24A and FY25A shares and EPS from filings.
Returns are healthy and the balance sheet is net cash, but at 46.7x FY27E earnings we are 3.5% above Street EPS and the stock is priced for it
| Growth and margins | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Growth y/y | |||||
| Sales | 40.8% | 38.7% | 17.7% | 16.0% | 17.4% |
| Adj. EBITDA | – | 59.4% | 5.8% | 20.4% | 21.7% |
| EBIT | – | 116.6% | −0.9% | 31.8% | 31.5% |
| EPS | n a | +351% | −70% | +28.4% | +27.9% |
| DPS | – | – | – | – | – |
| Margins | |||||
| Gross margin | 72.8% | 72.2% | 72.6% | 71.5% | 71.5% |
| Adj. EBITDA margin | 25.7% | 29.5% | 26.5% | 27.5% | 28.5% |
| EBIT margin | 8.4% | 13.1% | 11.0% | 12.5% | 14.0% |
| FCF margin | 35.3% | 34.7% | 31.0% | 30.0% | 32.5% |
| Interest income ($m) | 43 | 45 | 48 | 59 | 74 |
| Returns and valuation | FY24A | FY25A | FY26E | FY27E | FY28E |
|---|---|---|---|---|---|
| Returns and leverage | |||||
| ROE | 10.7% | 30.7% | 8.8% | 9.3% | 9.9% |
| ROA | 6.8% | 20.8% | 5.9% | 6.5% | 7.0% |
| ROCE (EBIT / equity) | 7.6% | 10.1% | 8.6% | 9.2% | 9.9% |
| Net debt / equity | −1.1x | −0.9x | −1.0x | −1.0x | −1.1x |
| Net debt / EBITDA | −4.6x | −4.2x | −4.8x | −5.1x | −5.3x |
| Valuation at $161.27 | |||||
| P/E | 81.9x | 18.2x | 59.9x | 46.7x | 36.5x |
| EV / adj. EBITDA | 35.0x | 21.9x | 20.7x | 17.2x | 14.1x |
| FCF yield | 3.2% | 4.4% | 4.6% | 5.2% | 6.6% |
| Dividend yield | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| P/E at target ($170) | 86.3x | 19.1x | 63.1x | 49.2x | 38.4x |
Source: Company filings (A); Penmark Research model, 5 Nov 2026 (E). ROE, ROA and ROCE on year-end balances (Dec-24 equity $824.5m, assets $1,302m); ROCE = EBIT / equity (no financial debt). Multiples at the 4 Nov 2026 close; last row at our target. FY25A ROE includes the one-time tax benefit. Duolingo pays no dividend; net interest is interest income on cash.
| US$ | 4Q26E | FY26E | FY27E |
|---|---|---|---|
| Revenue ($m) | |||
| Penmark estimate | 322.3 | 1,221.7 | 1,417.4 |
| Consensus | 318.2 | 1,210 | 1,379 |
| Prior-year period | 282.9 | 1,037.6 | 1,221.7 |
| Difference vs consensus | +1.3% | +1.0% | +2.8% |
| Penmark growth y/y | +13.9% | +17.7% | +16.0% |
| Diluted EPS ($) | |||
| Penmark estimate | – | 2.69 | 3.46 |
| Consensus | – | 2.78 | 3.34 |
| Prior-year period | – | 8.88 | 2.69 |
| Difference vs consensus | – | −3.1% | +3.5% |
| Penmark growth y/y | – | −70% | +28.4% |
Source: Penmark Research model; consensus: Tickflow.io (revenue, Sep 2026), WallStreetZen (EPS, Oct 2026), Yahoo Finance (4Q26E revenue); all unofficial and before the 3 Nov results. FY25A EPS includes the tax release.
| Duolingo vs peers | DUOL | Peer median | Diff. |
|---|---|---|---|
| Valuation (EV, 30 Oct peer prices) | |||
| EV / FY25 revenue | 6.5x | 5.9x | +8.7% |
| EV / FY25 GAAP EBITDA | 44.7x | 19.6x | +128% |
| Regression-implied EV / sales | 6.5x | 7.2x | −10.0% |
| Growth and margins, FY25 | |||
| Revenue growth | 38.7% | 27.6% | +11.1pp |
| GAAP EBITDA margin | 14.5% | 27.0% | −12.5pp |
Source: Signals (SEC-derived financials; peer closes 30 Oct 2026), Penmark model. Peers: Match, Roblox, Life360, Netflix, Grindr; EBITDA median of three meaningful peers. Diff. = % for multiples, pp for ratios.
The next test is February: the stock is not cheap on sales or cash flow, and the rebound needs 29% DAU growth to hold
- 7 DecLead-plaintiff deadline in the securities class action filed 7 Oct 2026 (class period May 2025 to Feb 2026); allegations only, not findings.
- 11 JanUS Supreme Court hears Apple v. Epic; the permissible external-link fee decides how much of the 85% Apple and Google share of revenue can be rerouted.
- 25 FebQ4 FY26 results and first FY27 guide (date estimated). We expect Q4 revenue $322m, DAU growth ~29% and FY27 adj. EBITDA margin 27.5%; consensus FY27 revenue $1,379m vs our $1,417m.
- MayQ1 FY27 results: subscribers after the New Year promotion; we model penetration at 19.5% of DAUs in FY27E (Debate 3).
Source: Company calendar and filings; court dockets; Penmark Research estimates.
- AI tutors erode the free app. DAU growth fades to ~3% by FY30. Likelihood Medium; −$38.9/share vs base (DAU bear).
- Margin stays near 26%. AI cost per user and marketing do not fall. Likelihood Medium; −$26.9/share vs base.
- Higher discount rate. +1pt on the 9.75% cost of equity. Likelihood Medium; −$17.3 on the $170 target.
- Against our rating: cheaper capital or on-time plan. A 1pt lower discount rate gives $193.4; the bull case is $227.9. Likelihood Medium; +$23.4/share vs the $170 target.
Source: Penmark model, value per share, one input at a time; target values from the Market Check tab.
Source: Yahoo Finance weekly closes; trailing-four-quarter revenue and diluted shares from the Penmark model historicals; Penmark Research calculation, Jan 2025 to 4 Nov 2026
Source: Signals (SEC-derived financials; closes 30 Oct 2026); Penmark model. Five peers; R-squared 0.61
| Indicator | Thesis confirmed if | Thesis broken if |
|---|---|---|
| Q4 DAU growth y/y (Feb 2027) | 29% or higher; FY27 goal path of 20% | Below 21%, the Q1 FY26 low |
| Paid subscribers / DAUs (Q1 FY27) | At or above 19.5% | Below 18.5%, our bear level |
| FY27 adj. EBITDA margin guide | 27.5% or higher | 25.5% or lower, our bear level |
| Revenue per subscriber (FY27) | Up 2% or more | Flat, as in our bear case |
| Share price against our $170 target | Above $188.9 would put total return below −10% (Sell) | Below $154.5 would put total return above +10% (Buy) |
Source: Penmark Research. Price levels are the $170 target divided by 0.90 and 1.10, the band limits for Sell and Buy.
The stock follows DAU growth, not margin: the trailing multiple has halved while the engine still runs at 29%
Source: Company shareholder letters (DAU growth); Yahoo Finance quarter-end closes; Penmark Research calculation; Dec 2024 to Sep 2026, n = 8
Source: Company filings (A); Penmark Research estimates (E), base case; goal of 100M DAUs in 2028 per Q4 FY25 letter
Source: Company filings (A); Penmark Research estimates (E)
Source: Company filings (A); Penmark Research estimates (E); consensus: Tickflow.io aggregate, Sep 2026, unofficial and pre-Q3 results
Source: Yahoo Finance weekly closes; trailing-four-quarter FCF (company definition) and diluted shares from the Penmark model historicals; Penmark Research calculation
Source: Signals (SEC-derived financials); Penmark model. GAAP EBITDA after stock-based compensation
Three debates carry the call: the market is worth at most $5.0 a share above us on any one of them
| Debate | Market view | Penmark view | Evidence | What changes our view | Resolves |
|---|---|---|---|---|---|
| 1 AI: tailwind or substitute | Q3 DAU growth of 29% shows AI features lift engagement; DAU growth stays near that pace. Worth $157.2. | DAUs normalise: 94m in Q4 FY28E, 160m by FY36E. Worth $152.2. | DAU/MAU at 42.4% (37.3% a year ago); Q3 DAUs 64.9m, +29%. | Q4 DAU growth under 21% (bear) or above 29% for two quarters. | Feb 2027 |
| 2 Margin rebuild after the 2026 investment year | Per-unit AI costs fall faster; margin is ~0.5pt above our path. Worth $154.8. | Rebuild 1pt a year from 26.5% to 32.5% in FY32E, the midpoint of the 30% to 35% target. Worth $152.2. | Q3 adj. EBITDA margin 25.5% vs 29.5% a year ago; gross margin steady at 72.9%. | FY27 margin guide above 28% (better) or below 25.5% (bear). | Feb 2027 |
| 3 Monetising AI in broader tiers | Video Call in Super lifts ARPU beyond price and mix; revenue ~1% above ours. Worth $153.5. | Subscribers per DAU recover only to 22%; revenue per subscriber +2% a year. Worth $152.2. | Subscribers per DAU 20.2% from 22.8%; Max sunset, Video Call to all Super by end-2026. | Q1 FY27 subscribers per DAU at or above 19.5% (confirms) or below 18.5% (breaks). | May 2027 |
Source: Penmark model (values per share at 30 Sep 2026); Q3 FY26 shareholder letter. Market view is our reading of what the price discounts.
| Line item | Penmark | Company guidance | Consensus (pre-Q3) | Penmark vs guidance | Penmark vs consensus |
|---|---|---|---|---|---|
| FY26E revenue | $1,222m | $1,221m | $1,210m | +$1m | +1.0% |
| FY26E adj. EBITDA | $324m | $323m | – | +$1m | – |
| FY26E free cash flow | $379m | Above $350m | – | +$29m above floor | – |
| FY27E revenue | $1,417m | Not guided | $1,379m | – | +2.8% |
Source: Q3 FY26 shareholder letter (3 Nov 2026); Penmark model; consensus Tickflow.io, Sep 2026, unofficial and before the Q3 results
Source: Company calendar (dates after 3 Nov 2026 estimated from prior-year timing); court dockets; Penmark Research
Duolingo is a subscription business: 84% of revenue is subscriptions and 85% of revenue passes through two app stores
Source: Company FY25 10-K (revenue by product; Apple 61.6% and Google 23.4% of revenue processed); Penmark Research. US$m, FY25A
Source: Q3 FY26 shareholder letter (3 Nov 2026)
What it sells, and why customers buy
Duolingo sells a free, game-like language app and charges for removing friction. 153.1m people use it monthly and 13.1m (8.6%) pay for Super or Max, which remove ads, add unlimited hearts and AI practice. Subscriptions were $873m, 84% of FY25 revenue; advertising $80m, the English Test $42m and in-app purchases $40m.
Pricing and payment
Most customers pay by annual plan through Apple or Google. Upfront annual payments built $496m of deferred revenue at Dec 2025 and $514m at Sep 2026. Apple processed 61.6% and Google 23.4% of FY25 revenue and take 15% to 30% of in-app payments; Duolingo books revenue gross and records the fee in cost of revenue.
Cost to deliver
72.2% gross margin in FY25 is stable; the margin pressure sits in operating costs. Cost of revenue was $288m (store fees, hosting and AI inference). R&D was $306m (29.5% of revenue), sales and marketing $126m (12.1%), G&A $182m (17.5%). Stock compensation, $137m, is 13.2% of revenue and is counted as a cost in our valuation.
Cash versus profit
Cash runs ahead of profit because customers prepay. FY25 operating cash flow was $388m against $136m of GAAP operating profit, and net operating working capital is negative $255m. Capex is 2.5% to 2.8% of revenue, so $360m of FY25 free cash flow is 34.7% of revenue.
Volume drivers
Daily users set subscribers. Q4 average DAUs grew 30% in FY25 to 52.7m and subscribers 28% to 12.2m. DAU/MAU is 42.4%, up from 37.3%, so existing users are using the app more often. In FY26 subscribers per DAU have fallen to 20.2% from 22.8% as free users were added faster than payers.
Price drivers
Price and mix are a minor lever. Revenue per subscriber rose 6.7% in FY25 to $80.5 and we model +2% a year. Management is moving Video Call into all Super plans by end-2026 and has stopped new Max sales from November 2026. Advertising adds $1.71 per DAU a year and is 7.7% of revenue.
Growth comes from more users, not higher prices: average subscribers rose 35% in FY25 and revenue per subscriber 6.7%
Source: Company FY25 10-K; Penmark Research. US$m; GAAP, after $137.4m stock-based compensation
Source: Company filings; Penmark model Revenue Build tab. Highlighted: drivers we forecast (blue)
Source: Company FY25 10-K; US$m
Source: Company FY25 10-K (revenue by geography; share of revenue processed by Apple and Google)
Where the money goes
Cost of revenue and R&D take 57 cents of each dollar. Sales and marketing is only 12 cents, which is why the free product can keep growing users cheaply. Operating profit was $135.6m in FY25 after $137.4m of stock compensation.
Concentration
The US is 38% of revenue and falling; two app stores collect 85%. Rest-of-world revenue grew 48% in FY25 against 26% in the US. Subscriptions are the only line above 8% of revenue.
The market is small and contested, but Duolingo has 3x the revenue of its nearest rival and a habit others lack
Source: Company FY25 10-K; Q3 FY26 shareholder letter; Penmark Research
Source: Company 10-K; Business of Apps estimates (Babbel €, Busuu, Memrise); Forbes via Yahoo, Nov 2025 (Speak annualised revenue, above $100m). Mixed currencies and bases
Market size
Language apps are a small, fast-growing market, and size estimates differ by a factor of five. Business of Apps puts 2025 global language-app revenue at $1.54bn on 327m downloads; Straits Research puts the wider market, including institutions and web, at $7.36bn in 2025 and $8.55bn in 2026. Duolingo's FY25 revenue of $1,038m is not comparable with either, so we do not state a share.
Threat: AI
AI tutors are a risk to the free app, not yet a measured loss. Google Translate has piloted AI speaking practice since August 2025 and Gemini Live and ChatGPT Voice offer conversation practice. Duolingo DAUs still grew 29% in Q3, so AI has not yet shown up as lost users; one broker flagged AI as a risk after slower downloads. The bear case models DAU growth fading to 3%.
| Source of advantage | Evidence | Durable? | What would break it |
|---|---|---|---|
| Habit | DAU/MAU 42.4% in Q3, up from 37.3%; 64.9m DAUs, +29% | Yes, while engagement rises | DAU/MAU back below 37% |
| Scale | 153.1m MAUs; Babbel passed 30m cumulative subscriptions, Speak about 15m cumulative downloads | Yes: 13.1m payers fund R&D of $306m | A rival with a free AI tutor at scale |
| Cost position | 72.9% gross margin; sales and marketing 12.1% of FY25 revenue | Mostly: store fees are the swing factor | Apple and Google fee changes |
| Brand | 64.9m DAUs, +29%; game partnerships brought millions of new and returning users in Q3 | Yes: the brand brings users in cheaply | Partnership-led user growth stalls |
| Switching costs | Streaks and progress history; no contract lock-in | Weak: monthly or annual plans | Competitor imports progress |
Source: Company Q3 FY26 letter and FY25 10-K; Business of Apps and company estimates for competitors (revenue and user bases mix currencies and bases); Penmark Research assessment.
Capital is kept, not returned: $1.2bn of operating cash flow since FY21 became cash, and founders control the vote
Source: Company filings: operating cash flow and capex FY21 to 9M26; acquisitions FY24 $6.6m, FY25 $33.1m, 9M26 $5.0m (acquisitions shown from FY24); buybacks and net-settlement taxes $89.3m, 9M26 net financing outflow. US$m
Source: SEC Form 4 filings via Yahoo Finance, 5 Nov 2025 to 5 Nov 2026; sales only; the one purchase was a director buying 5,000 shares at $99.76
Source: Penmark model: reported revenue vs midpoint of guidance given a quarter earlier
| Area | Fact | Why it matters |
|---|---|---|
| Control | Class B carries 20 votes a share against one for Class A. As of April 2026 the founders held 40.0% and 35.8% of votes. | Minority holders cannot remove management; the board is the check. |
| Management | Luis von Ahn is CEO and chair. Gillian Munson became CFO on 23 February 2026, succeeding Matthew Skaruppa. | Every period guided since she joined has been beaten. |
| Board | Ten directors, including both founders. | Founders plus eight others. |
| Capital return | $400m buyback authorised February 2026; $82.0m repurchased to 30 October (781,439 shares). No dividend. | Buyback is 1.0% of market cap; it offsets dilution, not more. |
| Litigation | Securities class action filed 7 October 2026 (W.D. Pa., 2:26-cv-02197) on a class period of 2 May 2025 to 26 February 2026. Allegations only. | No finding; value effect not modelled. |
Source: Company 10-K and 8-K filings, governance page and Q3 FY26 letter; court docket; Penmark Research. Votes: third-party analysis, Apr 2026.
The stock trades on how long 29% DAU growth lasts: it fell 25% when growth slowed and rose 26% when it re-accelerated
| Driver | What moves the stock | Tests confirmed | Market reaction vs our model | Debate | Next number to watch |
|---|---|---|---|---|---|
| D1 DAU growth | DAU growth sets the multiple, not margin | 3 of 3: event day, decomposition, correlation 0.83 | +26.0% on 3Q26 print; market view $157.2 vs ours $152.2 (+$5.0, 3.3%) | 1 | Q4 DAU growth, 25 Feb 2027; we model 29% |
| D2 Next-year outlook | Bookings and margin guide on results days | 2 of 3: event day, decomposition | −14.0% after 4Q25 (FY26 margin guide ~25%); market view $154.8 vs ours $152.2 (+$2.6, 1.7%) | 2 | FY27 margin guide, Feb 2027; we model 27.5% |
Source: Yahoo Finance prices and weekly returns; Company letters; Penmark model. Event-day move = close before to close after each release.
Source: Yahoo Finance closes; move from the close before to the close after each release (after-market releases)
Source: Yahoo Finance closes; trailing-four-quarter revenue 3Q25 $964.3m to 3Q26 $1,182.3m (Penmark model historicals); Penmark Research calculation
Source: Penmark model Market Check tab; value per share at 30 Sep 2026. Market view = our reading of what the price discounts
Market: +26.0% the day after 29% DAU growth (3Q26), −25.5% when growth slowed (3Q25); correlation with price 0.83.
Model: $152.2 vs market view $157.2, a $5.0 (3.3%) gap; bear $113.3, bull $171.1.
Next: Q4 DAU growth, 25 Feb 2027; we model 29%.
Market: −14.0% after 4Q25 (FY26 margin guide near 25%), −9.4% after 2Q26.
Model: $152.2 vs market view $154.8, a $2.6 (1.7%) gap; margin bear variant $125.3.
Next: FY27 adj. EBITDA margin guide, Feb 2027; we model 27.5%.
Three-quarters of the stock’s weekly movement is its own news: rates and the S&P 500 explain under 15%
Source: Yahoo Finance weekly closes, 104 weeks to 5 Nov 2026 (Friday closes); OLS R-squared of weekly log returns (yield: change in 10-year yield); Penmark Research calculation
What the factor test says
Three-quarters of weekly moves are specific to Duolingo. The software sector (IGV) explains 23.7% of weekly variance, the S&P 500 12.8% and rates 0.9%. All three together explain 24.5%, so 75.5% of weekly price variance is stock-specific news: results days and growth expectations.
What it means for value
A macro or sector view will not move our target. The two drivers do: the DAU path ($113.3 bear to $171.1 bull) and the margin path ($125.3 to $164.7).
Source: Penmark model; value per share at 30 Sep 2026, one input at a time
| Candidate | Test result | Verdict |
|---|---|---|
| S&P 500 | Weekly R² 12.8% alone; adds 0.4pt once software sector is included | Dropped: sector explains it |
| Software sector (IGV) | Weekly R² 23.7% alone; 24.1% with the S&P 500 | Kept as context, not a driver |
| 10-year Treasury yield | Weekly R² 0.9%; 24.5% with all three, a gain of 0.4pt | Dropped |
| Adj. EBITDA margin on its own | Quarterly margin vs price correlation 0.35; Q3 margin fell 4pts y/y while the stock rose 26% | Dropped as a standalone driver; kept inside D2 as guidance |
| Subscribers per DAU | Correlation 0.55 with price, but fell to 20.2% in the quarter the stock rose 26% | Dropped: fails the event-day test |
| Buybacks and insider sales | $82.0m repurchased; insiders sold $25.3m in 12 months | Dropped: too small against an $8.2bn market value |
Source: Yahoo Finance weekly closes (104 weeks to 5 Nov 2026), Penmark model Historicals; Penmark Research calculation.
DAU growth is durable enough to reach 94m by FY28E, but the market already pays for it
Source: Company filings (A); Penmark model (E): Q4 average DAUs, millions
1 What the market believes
The price needs more growth than we model. At $161.27 the market implies FY30E revenue of $2,395m against our $2,233m (7.2% higher).
2 The evidence, from three sources
- Reported: Q3 DAUs 64.9m, +29% y/y, accelerating from +21% in Q1 FY26.
- Engagement: DAU/MAU 42.4% vs 37.3% a year ago (Exhibit 47).
- Management: Q4 DAU growth expected similar to Q3; goal of 100m DAUs in Q4 2028.
Against: MAUs grew only 13% and subscribers 14%, so most new users are free.
Source: Company shareholder letters; Penmark model Historicals (quarterly average DAUs and MAUs)
3 What it is worth
The DAU path is worth $38.9 down and $18.9 up per share. Base reaches 93.8m DAUs in Q4 FY28E, 6% below the 100m goal; the bear path (DAU growth fading to 3%) gives $113.3, the bull path (goal met) $171.1, against the $152.2 base.
4 Signposts
- 25 Feb 2027: Q4 DAU growth of 29% confirms; under 21% breaks.
- May 2027: Q1 FY27 DAU growth against our 20% FY27E path.
- FY27 outlook: management on the 100m goal.
Margins rebuild by 1pt a year, but the price already assumes more than we do
Source: Company filings (A); Penmark model (E): adj. EBITDA margin; long-term target 30% to 35% set Feb 2023
1 What the market believes
At $161.27 the price already assumes a stronger margin than ours. It needs margins about 1.7pts above our FY27E to FY36E path, or a revenue path 7.2% higher.
2 The evidence, from three sources
- Reported: Q3 adj. EBITDA margin 25.5% vs 29.5% a year ago; gross margin held at 72.9%.
- Guidance: FY26 margin 26.5%, Q4 $84m; stock compensation about 14% of revenue.
- Target: company long-term margin target 30% to 35%, set Feb 2023; FY25 reached 29.5% before the FY26 investment year.
Against: R&D rose 30% y/y in Q3 ($107.7m vs $82.7m).
Source: Company filings (A); Penmark model (E). Adj. EBITDA adds back stock-based compensation incl. payroll tax
3 What it is worth
Each 1pt of margin is worth $5.29 a share. Our base rebuilds 1pt a year to 32.5% in FY32E. A margin 3pts lower is worth $136.3 and 3pts higher $168.1; the margin bear variant is $125.3 and the bull $164.7.
4 Signposts
- 25 Feb 2027: FY27 margin guide at or above 27.5% confirms; at or below 25.5% breaks.
- Each quarter: R&D growth below revenue growth.
- FY28E: margin 28.5%, back to FY25 level in FY29E.
Subscribers per DAU recover only slowly, and paid conversion is the smallest of the three levers
Source: Company filings (A); Penmark model (E): paid subscribers / Q4 average DAUs
1 What the market believes
The market pays for modest upside here. Our debate gap is $1.3 a share: revenue about 1% above our path if AI tiers lift ARPU.
2 The evidence, from three sources
- Reported: subscribers per DAU 20.2% at Q3 vs 22.8% a year ago; subscription bookings +16% vs DAUs +29%.
- Product: Video Call to all Super subscribers by end-2026; Max sunset, no new Max sales from November 2026.
- Revenue mix: subscriptions 87% of FY26E revenue (Exhibit 51).
Against: management expects minimal bookings effect from the Max change, so no uplift is guided.
Source: Company filings (A); Penmark model Revenue Build tab (E); US$m
3 What it is worth
Subscribers per DAU is worth $16.5 down and $5.4 up. Base: 19.5% in FY27E, recovering to 21.5% by FY32E. The penetration bear variant gives $135.7, the bull $157.6. Revenue per subscriber (+2% a year in base): bear $134.6, bull $162.1.
4 Signposts
- May 2027: Q1 FY27 subscribers per DAU at or above 19.5% confirms; below 18.5% breaks.
- Feb 2027: FY27 bookings growth against our $1,513m.
- Mid-2027: revenue per subscriber up 2% or more.
Reported numbers are high quality: cash conversion is 118% and the main distortion is a one-off $222.7m tax benefit
| FY21A | FY22A | FY23A | FY24A | FY25A | |
|---|---|---|---|---|---|
| Growth | |||||
| Revenue ($m) | 251 | 369 | 531 | 748 | 1,038 |
| Revenue growth | n/a | 47.3% | 43.7% | 40.8% | 38.7% |
| From more subscribers (volume) | n/a | n/a | 61.2% | 49.1% | 34.8% |
| From revenue per subscriber (price, mix) | n/a | n/a | −8.2% | 0.7% | 6.7% |
| Margins | |||||
| Gross margin | 72.4% | 73.1% | 73.2% | 72.8% | 72.2% |
| Adj. EBITDA margin | −0.4% | 4.2% | 17.6% | 25.7% | 29.5% |
| GAAP EBIT margin | −23.9% | −17.6% | −2.5% | 8.4% | 13.1% |
| Gap: adj. EBITDA less EBIT margin | 23.5pp | 21.8pp | 20.1pp | 17.3pp | 16.4pp |
| Cash | |||||
| Free cash flow ($m) | 3 | 44 | 140 | 264 | 360 |
| FCF / adj. EBITDA | n/m | 282% | 149% | 138% | 118% |
| Stock comp (SBC-related) / revenue | n/a | 20.5% | 18.7% | 16.1% | 14.3% |
Source: Company 10-K filings FY21 to FY25 (A); Penmark model. Volume = growth in average paid subscribers; price = growth in subscription revenue per average subscriber. Growth rows are subscription-driven, not total revenue.
Source: Company filings (A); Penmark model (E)
What the record says
Profit quality is good: cash conversion is above 100% of adj. EBITDA. Free cash flow was 118% of adj. EBITDA in FY25 because customers prepay. GAAP EBIT margin rose from −17.6% in FY22 to 13.1% in FY25 while stock compensation fell from 20.5% to 14.3% of revenue.
What flatters the result
The adj. EBITDA to EBIT gap is 16.4pp in FY25; stock compensation is most of it. We value on cash flow after stock compensation, so this does not inflate our target.
| Item | What it does | Size |
|---|---|---|
| Stock compensation add-back | Adj. EBITDA excludes SBC-related costs of $148.6m in FY25 (14.3% of revenue); we deduct it in the DCF. | $148.6m FY25 |
| One-time tax benefit | Q3 2025 net income included a $222.7m tax benefit, so FY25 EPS of $8.88 and ROE of 30.7% are not a base for FY26E. | $222.7m |
| Deferred revenue | Annual plans paid upfront put $496m on the balance sheet; bookings ($1,158m) ran $120m ahead of revenue in FY25. | $120m |
| Capitalised software | Small: $9.3m in FY25, 0.9% of revenue, so little profit is deferred. | $9.3m |
| Interest income | $45m in FY25 is 25% of pre-tax profit ($182m); it falls as a share as operating profit rises. | $45.2m |
Source: Company FY25 10-K and Q3 FY26 shareholder letter; Penmark model.
Every forecast starts from the DAU path: subscribers and ads follow it, and costs rebuild margin by 1pt a year
Source: Company filings; Penmark model Revenue Build tab. Highlighted: drivers we forecast (blue)
| Line | Penmark | Guidance / Street | Why we differ |
|---|---|---|---|
| Q4 FY26 revenue | $322.3m | Guide $322m | On guidance; Q3 beat midpoint by 2.3%. |
| FY26E revenue | $1,222m | Guide $1,221m; Street $1,210m | Guidance raised after Street snapshot. |
| FY26E adj. EBITDA | $324m (26.5%) | Guide $323m (26.5%) | Margin set to guidance. |
| FY26E free cash flow | $379m | Floor above $350m | Prepaid annual plans add $75m of deferred revenue. |
| Q4 FY26 DAU growth | 29% | Similar to Q3 (29%) | Management guide used as the point estimate. |
| Q4 FY28E DAUs | 93.8m | Goal 100m | Base fades growth to 15%; goal met only in bull case. |
| FY27E revenue | $1,417m | Street $1,379m (+2.8%) | Street pre-dates the Q3 beat and raised Q4 guide. |
| FY26E / FY27E EPS | $2.69 / $3.46 | Street $2.78 / $3.34 | FY26E lower on 24% tax and SBC; FY27E higher on revenue. |
Source: Q3 FY26 shareholder letter (3 Nov 2026); Tickflow.io and WallStreetZen (unofficial, pre-Q3); Penmark model. Costs: gross margin 71.5% to 72.0%, R&D, S&M and G&A fall as a share of revenue; capex 2.5% of revenue; no debt.
| Driver (one at a time) | Bear input | Bull input | Value range ($/share) | Swing |
|---|---|---|---|---|
| Q4 DAU path | DAU growth fades to 3% | Goal of 100m DAUs met | $113.3 to $171.1 | $57.8 |
| Adj. EBITDA margin | Stays near FY26 level | Higher path | $125.3 to $164.7 | $39.5 |
| Revenue per subscriber | Flat | Higher growth | $134.6 to $162.1 | $27.6 |
| Subscribers per DAU | Lower path | Recovers faster | $135.7 to $157.6 | $21.9 |
| Discount rate | +1pt (10.75%) | −1pt (8.75%) | $135.5 to $174.7 | $39.2 |
Source: Penmark model; value per share at 30 Sep 2026; each driver moved alone with all others at base.
The DCF says $152.2 and the cross-check $175.8: the $170 target sits between them and 5.4% above the price
Source: Penmark model; Yahoo Finance (Street targets, 52-week range) as at 4 Nov 2026. Model values at 30 Sep 2026; target at 4 Nov 2027
Source: Penmark model; $ per share at 30 Sep 2026; base case outlined
Source: Penmark model; FY27E to FY36E offsets; $ per share at 30 Sep 2026; first-order
| Method | Why it fits | Value / share |
|---|---|---|
| Primary: 10-year DCF after stock compensation | Subscription cash flows are long-lived and recurring; stock compensation (13% to 14% of revenue) is a real cost, so it is deducted and future grants are not also diluted. | $152.2 base; $153.5 weighted |
| Cross-check: EV/sales against growth, 5 peers | Peers are subscription or freemium apps; growth explains 61% of their EV/sales (R² 0.61, n=5). Applied at DUOL FY25 growth of 38.7% gives 7.2x. | $175.8 |
Source: Penmark model; values per share at 30 Sep 2026.
Terminal value is 55% of the $6.2bn enterprise value, and the price only needs 1pp more terminal growth than we use
Source: Penmark model, base case, $m at 30 Sep 2026; 50.8m diluted shares give $152.2 a share
Source: Penmark model; DCF value per share at 30 Sep 2026
| Reverse DCF | Model | What the price needs | Gap |
|---|---|---|---|
| Terminal growth | 3.0% | 4.0% (4.011%) at 9.75% WACC | +1.0pp |
| Discount rate | 9.75% | 9.31% (9.309%) at 3.0% growth | −0.44pp |
| Adj. EBITDA margin, FY27E to FY36E | Path to 32.5% | Path +1.7pts (to 34.2% in FY36E) | +1.7pts |
| FY30E revenue | $2,233m | $2,395m | +7.2% |
| Multiple at price, FY27E: EV/adj. EBITDA, P/E | At DCF value: 16.0x, 44.0x | At price: 17.2x, 46.7x | +1.2x, +2.6x |
Source: Penmark model; value at 30 Sep 2026; price $161.27 at 4 Nov 2026. Implied margin and revenue offsets are first-order.
How the cases tie to the drivers
Bear (20%, $78.3): DAU growth fades to 3%, penetration and revenue per subscriber stay low, margin near 26%; Debates 1 to 3 all go against us.
Base (50%, $152.2): 94m DAUs in FY28E, margin to 32.5%, penetration to 22%.
Bull (30%, $205.8): the 100m DAU goal is met (100.4m in FY28E) and margins and penetration run above base. Weights lean to Bull as Duolingo beat guidance every period tracked; they are judgement.
| Step | Value / share | Link |
|---|---|---|
| Bear / Base / Bull DCF at 30 Sep 2026 | $78.3 / $152.2 / $205.8 | Probabilities 20% / 50% / 30% |
| Probability-weighted value, 30 Sep 2026 | $153.5 | 0.2 x 78.3 + 0.5 x 152.2 + 0.3 x 205.8 |
| Rolled to 12 months at 9.75%, 400 days | $86.7 / $168.5 / $227.9 | No dividend; price date 4 Nov 2026 |
| Probability-weighted 12-month target | $170.0 ($169.97) | Page 1 call box |
| Expected total return vs $161.27 | +5.4% (5.395%) | Hold: between −10% and +10% |
Source: Penmark model Summary, Scenarios and Market Check tabs.
The next 12 months turn on one print: February’s DAU growth and FY27 margin guide decide whether Hold becomes Buy or Sell
| Date | Event | Penmark expectation | Consensus | Rating effect |
|---|---|---|---|---|
| 7 Dec 2026 | Class-action lead-plaintiff deadline (filed 7 Oct; allegations only) | No value effect modelled | n/a | None |
| 11 Jan 2027 | US Supreme Court hears Apple v. Epic; fee outcome unresolved | No fee saving in base | n/a | Lower fees would lift value |
| 25 Feb 2027 (est.) | Q4 and FY27 guide. D1: Q4 DAU growth 29%; D2: FY27 margin 27.5% | Q4 revenue $322m; FY27E revenue $1,417m | FY27E revenue $1,379m; DAU and margin n/a | Under 21% DAU growth: toward Sell |
| May 2027 (est.) | Q1 FY27. Debate 3: subscribers per DAU 19.5% | n/a | Under 18.5%: bear path | |
| Aug 2027 (est.) | Q2 FY27: margin on track for 28.5% in FY28E | n/a | Margin below 25.5%: bear | |
| Nov 2027 | 12-month target date: $170 | Street mean $137.57 | Price above $188.9 or below $154.5 changes rating |
Source: Company calendar and filings; court dockets; Penmark model. Dates after 3 Nov 2026 estimated from prior-year timing. Consensus: Tickflow.io (unofficial, pre-Q3).
| Risk | Likelihood | Value effect / share | What ends the thesis |
|---|---|---|---|
| AI tutors substitute for the free app | Medium | −$38.9 (DAU path bear, $113.3) | DAU growth below 21% with penetration under 18.5% |
| Margin stuck near 26% | Medium | −$26.9 (margin bear, $125.3) | FY27 margin guide at or below 25.5% |
| Lower paid conversion or flat price | Medium | −$16.5 penetration; −$17.6 revenue per subscriber | Subscribers per DAU below 18.5% |
| Higher discount rate (+1pt) | Medium | −$17.3 on the target ($152.7) | Not a thesis break |
| Against our rating: lower discount rate (−1pt) or bull path | Medium | +$23.4 on the target ($193.4); bull $227.9 | Would justify Buy above +10% return |
| App-store fees and platform rules | Not rated | Not quantified | Fee rise on 85% of revenue |
| Securities class action; founder control | Not rated | Not quantified; allegations only | Adverse ruling or governance change |
Source: Penmark model (variants, value per share at 30 Sep 2026; target values). Likelihood is Penmark judgement, not a statistic.
Holders are concentrated and the short base is large: 20% of the float is short against a Street that mostly says Hold
Source: Yahoo Finance analyst price targets, 4 Nov 2026; Penmark Research
Source: Yahoo Finance recommendation counts, 4 Nov 2026
| Holder | Shares (m) | % of shares | Source date |
|---|---|---|---|
| Baillie Gifford | 4.82 | 11.95% | 30 Jun 2026 |
| BlackRock | 3.87 | 9.58% | 30 Jun 2026 |
| AQR Capital | 2.33 | 5.78% | 30 Jun 2026 |
| Vanguard Capital Management | 1.75 | 4.32% | 30 Jun 2026 |
| All institutions | n/a | 98.70% | Yahoo Finance, Nov 2026 |
| Insiders (as reported) | n/a | 1.49% | Yahoo Finance, Nov 2026 |
Source: Yahoo Finance institutional holder filings (13F) as of 30 Jun 2026; ownership totals as at 5 Nov 2026. Percentages are of shares outstanding as reported by the data provider.
| Item | Value | Read-across |
|---|---|---|
| Short interest | 7.1m shares, 20.05% of float (15 Oct 2026) | One in five free-float shares is short; days to cover 7.2. A beat can squeeze, which fits the +26.0% move on 4 Nov. |
| Liquidity | $176m average daily value, 3 months | A $10m position equals 6% of one day's average traded value. |
| Street ratings | 6 Buy, 16 Hold, 2 Sell; mean target $137.57 | Street mean is 14.7% below the price; our $170 is 23.6% above it. |
| Buyback | $82.0m of $400m used to 30 Oct 2026 | $318m left; 3.9% of market value. |
| Control | Class B 20 votes per share | Founders control the vote; free float is the Class A. |
Source: Yahoo Finance (short interest 15 Oct 2026; ratings and targets 4 Nov 2026; price history); company filings; Penmark Research calculation.
The model: revenue compounds at 17% a year to FY28E and margins rebuild, while quarterly margin has lagged user growth
| FY24A | FY25A | FY26E | FY27E | FY28E | FY29E | FY30E | |
|---|---|---|---|---|---|---|---|
| Income statement | |||||||
| Revenue ($m) | 748 | 1,038 | 1,222 | 1,417 | 1,664 | 1,938 | 2,233 |
| Growth | 40.8% | 38.7% | 17.7% | 16.0% | 17.4% | 16.4% | 15.3% |
| Gross profit | 544 | 749 | 887 | 1,013 | 1,190 | 1,395 | 1,608 |
| Gross margin | 72.8% | 72.2% | 72.6% | 71.5% | 71.5% | 72.0% | 72.0% |
| Adj. EBITDA | 192 | 306 | 324 | 390 | 474 | 572 | 681 |
| Margin | 25.7% | 29.5% | 26.5% | 27.5% | 28.5% | 29.5% | 30.5% |
| Stock-based comp. incl. payroll tax | 120 | 149 | 171 | 191 | 216 | 233 | 246 |
| EBIT (GAAP) | 63 | 136 | 134 | 177 | 233 | 310 | 402 |
| Interest income | 43 | 45 | 48 | 59 | 74 | 95 | 119 |
| Net income | 89 | 414 | 137 | 179 | 234 | 308 | 396 |
| Diluted EPS ($) | 1.97 | 8.88 | 2.69 | 3.46 | 4.42 | 5.71 | 7.20 |
| Cash flow and balance sheet | |||||||
| Operating cash flow | 286 | 388 | 416 | 461 | 583 | 701 | 789 |
| Capex incl. software | 21 | 27 | 37 | 35 | 42 | 48 | 56 |
| Free cash flow (company definition) | 264 | 360 | 379 | 426 | 541 | 652 | 733 |
| Cash and investments | 878 | 1,276 | 1,560 | 1,986 | 2,527 | 3,180 | 3,913 |
| Deferred revenue | 373 | 496 | 572 | 667 | 789 | 925 | 1,073 |
| Equity | 825 | 1,347 | 1,557 | 1,918 | 2,357 | 2,886 | 3,515 |
| Users | |||||||
| Q4 average DAUs (m) | 40.5 | 52.7 | 68.0 | 81.6 | 93.8 | 105.1 | 115.6 |
| Paid subscribers (m) | 9.5 | 12.2 | 13.6 | 15.9 | 18.3 | 21.0 | 23.7 |
| Subscribers / DAUs | 23.5% | 23.1% | 20.0% | 19.5% | 19.5% | 20.0% | 20.5% |
Source: Company filings (A); Penmark Research model, 5 Nov 2026 (E). US$m unless stated. FY25A net income includes a $222.7m one-time tax benefit. Q4 DAUs; subscribers at period end. Shaded = estimates.
| 4Q24 | 1Q25 | 2Q25 | 3Q25 | 4Q25 | 1Q26 | 2Q26 | 3Q26 | |
|---|---|---|---|---|---|---|---|---|
| Revenue ($m) | 209.6 | 230.7 | 252.3 | 271.7 | 282.9 | 292.0 | 298.5 | 309.0 |
| Total bookings | 271.6 | 271.6 | 268.0 | 281.9 | 336.8 | 308.5 | 289.1 | 316.1 |
| Gross margin | 71.9% | 71.1% | 72.4% | 72.5% | 72.8% | 73.0% | 72.6% | 72.9% |
| EBIT (GAAP) | 13.9 | 23.6 | 33.4 | 35.2 | 43.5 | 44.5 | 33.9 | 24.2 |
| Adj. EBITDA | 52.3 | 62.8 | 78.7 | 80.0 | 84.3 | 83.4 | 77.3 | 78.7 |
| Margin | 25.0% | 27.2% | 31.2% | 29.5% | 29.8% | 28.6% | 25.9% | 25.5% |
| Net income | 13.9 | 35.1 | 44.8 | 292.2 | 42.0 | 43.5 | 33.2 | 27.2 |
| Operating cash flow | 83.3 | 105.6 | 90.7 | 84.2 | 107.3 | 150.8 | 88.3 | 89.4 |
| Free cash flow | 80.9 | 103.0 | 86.3 | 77.4 | 93.7 | 147.8 | 78.6 | 75.3 |
| DAUs (m, quarterly average) | 40.5 | 46.6 | 47.7 | 50.5 | 52.7 | 56.5 | 58.7 | 64.9 |
| MAUs (m) | 116.7 | 130.2 | 128.3 | 135.3 | 133.1 | 137.8 | 140.6 | 153.1 |
| Paid subscribers (m) | 9.5 | 10.3 | 10.9 | 11.5 | 12.2 | 12.5 | 12.7 | 13.1 |
| Subscribers / DAUs | 23.5% | 22.1% | 22.9% | 22.8% | 23.1% | 22.1% | 21.6% | 20.2% |
Source: Company shareholder letters (A); Penmark model. US$m unless stated; net income in 3Q25 includes the $222.7m tax benefit. Forward quarters are not modelled beyond the Q4 FY26 stub (revenue $322.3m, adj. EBITDA $84.3m).
Reference: every adjusted figure reconciles to GAAP, and every estimate traces to the 5 November model
| Reconciliation | FY25A | Q3 FY26 |
|---|---|---|
| GAAP EBIT | $135.6m | $24.2m |
| Add: depreciation and amortisation | $14.4m | $4.3m |
| Add: SBC-related costs incl. payroll taxes | $148.6m | $49.1m |
| Add: other items, less other income (net) | $7.3m | $1.1m |
| Adjusted EBITDA (company definition) | $305.9m | $78.7m |
Source: Company 10-K and Q3 FY26 shareholder letter; Penmark model. US$m.
Glossary
- DAU / MAU: daily and monthly active users; fiscal-year figures are Q4 averages.
- Paid subscribers: users on Super or Max plans at period end; penetration = subscribers / Q4 DAUs.
- Bookings: revenue plus change in deferred revenue; subscription bookings are cash billed for plans.
- Adj. EBITDA: company measure that adds back D&A, SBC-related costs and some other items.
- UFCF: unlevered free cash flow after SBC; FCF (company) = operating cash flow less capex and capitalised software.
Ratio definitions
- ROE: net income / year-end equity; ROA: net income / year-end assets; ROCE: EBIT / year-end equity (no financial debt).
- FCF yield: company FCF / market cap at $161.27 ($8,192m).
- P/E: price / diluted EPS; EV/adj. EBITDA: EV $6,709m at the price date.
- Net debt / EBITDA: negative when cash exceeds debt (no debt).
Peer set
Match Group, Roblox, Life360, Netflix, Grindr: consumer subscription or freemium businesses, chosen by business model and growth. EV at 30 Oct 2026 closes; FY25 financials from SEC filings.
Methodology
Primary method is a 10-year DCF after stock compensation (Q4 FY26 stub plus FY27E to FY36E, mid-period, valued at 30 Sep 2026, 9.75% cost of equity, 3.0% terminal growth, 25% terminal ROIC), rolled forward 400 days to the 4 Nov 2027 target. Bear, base and bull are weighted 20%, 50% and 30%. Cross-check: EV/FY25 revenue against growth for five peers. Peers are compared on GAAP EBITDA. The base case stays below the 100m DAU goal.
| Source | As of | Used for |
|---|---|---|
| Duolingo FY25 10-K and Q3 FY26 shareholder letter (SEC) | 3 Nov 2026 | Financials, KPIs, guidance, geography, app-store share |
| Penmark model (Signals SEC-derived data) | 5 Nov 2026 | All estimates, DCF, scenarios, peers |
| Yahoo Finance | 4 to 5 Nov 2026 | Prices, index, Street ratings and targets, holders, insiders, short interest |
| Tickflow.io, WallStreetZen (unofficial) | Sep to Oct 2026 | Consensus revenue and EPS, pre-Q3 |
| Business of Apps, Straits Research; court dockets | 2025 to 2026 | Market scale, litigation, fee risk |
Source: Penmark Research. Model data cut-off 5 Nov 2026; price date 4 Nov 2026.
Analyst certification and important disclosures
Ratings
Ratings reflect the expected total return to our 12-month price target, including dividends, weighted across bull, base and bear scenarios.
| Buy | Expected total return above +10% |
| Hold | Expected total return between −10% and +10% |
| Sell | Expected total return below −10% |
We review ratings at each set of results and whenever the facts change materially. A share price can move outside its band before the rating changes, and targets may not be reached.
Analyst certification
The Penmark Team certifies that (1) the views in this report accurately reflect its views about the subject company and its securities, and (2) no part of its compensation was, is or will be directly or indirectly related to the specific ratings, price targets or views in this report.
Contact
Penmark Research
Questions about our research:
research@penmarkresearch.com
All our research and coverage:
penmarkresearch.com
Basis of preparation
This report uses information publicly available up to its publication date. Share prices are closing prices on the trading day before the publication date, unless stated otherwise. Forecasts, valuations and price targets are Penmark Research’s own, and forecast periods are marked E. Sources are shown beneath each exhibit.
Valuation and price targets
Price targets are based on the valuation methods set out in each report and are weighted across bull, base and bear scenarios. The main risks to each rating and price target are described in the report. Penmark Research updates its coverage after each set of results but is not obliged to update a report at any other time.
Independence and conflicts of interest
Penmark Research is independent. It does not provide investment banking, brokerage, market-making or asset management services, does not trade securities, and does not accept payment from any company for research coverage. Companies do not see or review Penmark Research reports before publication.
No member of the Penmark Team holds a long or short position, directly or through derivatives, in the securities of any company Penmark Research covers.
Regulatory status
Penmark Research publishes general, impersonal investment research on a regular basis. It is not registered as an investment adviser with the US Securities and Exchange Commission or any state securities authority, is not a registered broker-dealer and is not a member of FINRA. Penmark Research does not give advice tailored to any person’s circumstances.
Not investment advice or an offer
This report is for information only. It is not personalised investment advice and is not an offer or solicitation to buy or sell any security or other financial instrument. It does not take into account any reader’s objectives, financial situation or needs. Readers should make their own investment decisions and seek independent professional advice where appropriate.
Forward-looking statements and risk
Forecasts, valuations and price targets are estimates based on assumptions that may prove wrong, and they may not be achieved. Opinions are current as of the publication date and may change without notice. Investing in equities involves risk, including the loss of capital. Share prices can be volatile, and past performance is not a reliable indicator of future results.
Data and sources
Company financials, operating data and filings come from company filings, earnings releases and earnings calls. Market data and any consensus estimates come from third-party sources named beneath the relevant exhibit; consensus figures from aggregators are unofficial. Penmark Research does not use other brokers’ research. Third-party data belong to their providers and may not be extracted from this report.
Accuracy and liability
Information is taken from sources believed to be reliable, but Penmark Research makes no representation or warranty as to its accuracy, completeness or timeliness. To the fullest extent permitted by law, Penmark Research accepts no liability for any loss arising from use of this report. Descriptions of litigation and regulatory matters reflect allegations and public records, not findings, unless stated. If a published report is corrected, the corrected version carries an “Updated” date at penmarkresearch.com.
Distribution and access
Penmark Research reports are available to registered users of penmarkresearch.com. Downloads and programmatic access are governed by the Penmark Research subscription terms. This report is not directed at, or intended for, any person in Singapore or India, and access from those countries is blocked. It is not intended for distribution to, or use by, any person in any jurisdiction where that would be contrary to law or regulation or would subject Penmark Research to registration or licensing requirements there.
Copyright and permitted use
© Penmark Research. All rights reserved. Penmark Research is a trade name of Brightriver Labs. Readers may share a link to this report but may not copy, reproduce, redistribute, sell or use its content, including for text and data mining, without written consent. Contact: research@penmarkresearch.com.