DoorDash DASH
Communication Services · Internet Content & Information · Synthos Deep Dive · 2026-08-04
The Overview
DoorDash delivers restaurant meals and, increasingly, groceries. It does not own restaurants or vehicles; it runs the app that connects a hungry person, a merchant and a driver, and takes a fee from all three. Last October it bought Deliveroo, the British equivalent, so it now runs three of these marketplaces across more than thirty countries.
It is growing very fast. Revenue rose 33% in the most recent quarter — more than twice the rate of any other company in this research batch — and it has grown every year for at least six.
It has also, finally, started making money. Until 2024 it lost money every year. In 2025 it made an operating profit of $723 million and a net profit of $935 million.
So why is the share price down 18% over the past year while the market rose 24%?
Two reasons.
The first is arithmetic. At $202 you pay about 80 times what analysts expect the company to earn this year. For that to make sense, profit has to rise very fast — analysts expect it to grow 77% in 2027 and 59% in 2028, far faster than revenue. That requires the business to become much more profitable per order, and the most recent quarter went the wrong way: revenue grew 33% while operating profit actually fell slightly.
The second is that the profit is not quite as real as it looks. DoorDash paid its staff $1,051 million in shares last year — more than its entire reported net profit — and it paid almost no tax, at an effective rate of 0.6%. Both of those will normalise.
There is a genuine long-term idea underneath, and it is the most interesting one in this batch. Independent analysis in our knowledge base found that on a typical $31 order DoorDash is "lucky to make $1" — because most of the money goes to the person driving the car. DoorDash executives describe an in-house delivery robot, 300 pounds, 20 miles per hour, driving itself, that has been operating in Phoenix for two years. If machines eventually do the driving, the company that owns the customers and the routes owns the savings. That is worth a great deal. It is also, at this stage, something we can only see through management's own description, and an independent analyst said in 2023 that the robotics programme had already been scaled back.
We think fair value is about $225 against $202.
The company reports tomorrow, and it has come in below expectations by about 19% in two of the last four quarters. That is not a day to act.
- Downside Risk 7/10. Net cash and no debt problem — but 80x earnings, a 1.78 beta and a 0.6% tax rate.
- Growth Quality 9/10. The highest in this batch. 33% revenue growth and seven straight profitable quarters.
- Exponential Potential 6/10. The highest in this batch. Real network effects, real data, and a contested autonomy programme.
Putting a number on it: our fair-value estimate is $225 against a current price of $236.74 — consistent with our call to stay away or wait for a better setup.
Our summary metrics
"Rated 7 — the risk is entirely in the valuation and the earnings quality, not in the balance sheet. Supports: cash of $4,378 million plus short-term investments of $1,128 million against $3,290 million of total debt gives NET CASH of $2,216 million; a 1.43x current ratio; no dividend obligation; and $2,431 million of operating cash flow in fiscal 2025 on $257 million of capital expenditure. Against that: the shares carry a beta of 1.779, the highest in this batch by 55%, and trade at 79.9x the current-year consensus, so the multiple alone is the dominant risk. Earnings quality is the second issue — stock-based compensation of $1,051 million is 7.7% of revenue and 112% of GAAP net income of $935 million, the accumulated deficit is $4,320 million, and the effective tax rate is 0.64%, so the reported profit is not yet taxed at a normal rate and will be. Execution risk is third and it is fresh: the company spent $4,151 million on acquisitions in fiscal 2025, completing the purchase of Deliveroo plc on 2025-10-02, and now runs three marketplaces (DoorDash, Wolt, Deliveroo) with three membership programmes (DashPass, Wolt+, Deliveroo Plus) across more than the number of countries the vendor payload records anywhere. And the consensus has been missed by 19.5% and 18.6% in two of the last four quarters. The structural risks are named in the knowledge base by an independent analyst rather than by us: a roughly 3.1% blended contribution margin, and gig-labour regulation that at New York's driver minimum implies roughly $7 of additional fee on a $31 order."
"Rated 9 — the highest in this batch and by a wide margin. Revenue: $2,886M (2020), $4,888M (2021), $6,583M (2022), $8,635M (2023), $10,722M (2024), $13,717M (2025) — a 36.5% compound rate over five years with no down year, and the FY2025 rate was still 27.9%. The March 2026 quarter delivered $4,036M against $3,032M, up 33.1%, which is more than double the growth rate of any other name in these twelve dives. The profit inflection is equally sharp and more recent: operating income by fiscal year ran MINUS $436M, MINUS $452M, MINUS $1,124M, MINUS $579M, MINUS $38M and then PLUS $723M in FY2025, with net income of $935M against $123M. Quarterly operating income turned positive in Q3 2024 and has stayed positive for seven consecutive quarters. Geography is a genuine second engine: non-US revenue grew from $332M (2022) to $854M (2023) to $1,319M (2024) to $2,257M (2025), and the Deliveroo acquisition completed 2025-10-02 adds a fourth year of that. Consensus wants revenue of $17,633M in FY2026 (32 analysts, +28.6%), $21,276M in FY2027 (33, +20.7%) and $25,241M in FY2028 (24, +18.6%), with EPS of $2.533, $4.486 and $7.117 — 12.4%, then 77.1%, then 58.6%. A 9 rather than a 10 because a material part of the recent revenue growth is acquired rather than organic, and because two of the last four quarters missed consensus earnings by roughly 19%."
"Rated 6 — the highest exponential score in this batch. DoorDash is a three-sided marketplace with real network effects (more merchants attract more consumers attract more Dashers), operating three marketplaces across more than thirty markets, and it is building two things on top that are genuinely non-linear. The first is data: the knowledge base carries a management claim of 10 billion deliveries of real-world data, which is the kind of proprietary corpus that cannot be bought. The second is autonomy, and here the specificity is unusual — a management voice describes an in-house DOT robot at 300 pounds and 20 miles per hour with level-4 autonomy, live in Phoenix for two years. If local delivery is eventually performed by machines, the party that owns the demand aggregation, the merchant relationships and the route data is the natural owner of the economics, and DoorDash's blended contribution margin of roughly 3.1% on a $31 order — the independent analyst's figure — is precisely the number that autonomy attacks. Against all of that: both autonomy claims are MANAGEMENT claims and are half-weighted; the same store carries an independent 2023 claim that DoorDash's robotics initiatives were already dead or significantly scaled back; and the sharpest bear claim in the lane says AI agents disintermediate aggregators rather than empower them. A 6: genuine network economics, a genuinely proprietary data asset, an autonomy programme we can only see through management's eyes, and a credible argument that the aggregator layer is the thing that gets removed."
What does “fair value” mean?
Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.
The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, We don’t publish a reverse-DCF cross-check for pre-profit companies — negative or missing earnings break that math — so take this number on our modeling alone.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
The Road Ahead
What we expect to matter in each window, and the evidence that would prove us wrong.
Short term 0-6 months
Neutral- Driver
- "An earnings print tomorrow, on the highest-beta name in this batch, after two nineteen-per-cent misses in the last four quarters. DoorDash closed 2026-08-04 at $202.37, up 0.93%, which is 14.8% above a 50-day moving average of $176.35 and 6.3% above a 200-day average of $190.34. It is 28.2% below the 52-week high of $281.74 and 38.0% above the low of $146.60. RSI is 58.8 and MACD +5.36 — mid-range and positive. Three-month return is +17.3% against SPY's +7.6%; six-month MINUS 2.6%; twelve-month MINUS 18.5% against SPY's +24.3%, a 43-point deficit. So the twelve-month record is poor, the last quarter has been strong, and the stock has recovered 38% off its low. Beta is 1.779, the highest in these twelve by 55%. Consensus for tomorrow is $0.4681 of EPS on $4,339.6M of revenue, the latter implying +32.1% growth. The street carries 29 buys, 9 holds and ZERO sells with a $248.50 target. We do not take positions the day before a print on a 79.9x current-year multiple with a demonstrated miss record."
- What we’re watching
- "The 2026-08-05 print against consensus EPS of $0.4681 and revenue of $4,339.6M. Within it: REVENUE GROWTH against the +33.1% just delivered — consensus wants +32.1%, so the bar is essentially the current rate and any deceleration is visible immediately; the contribution of Deliveroo, acquired 2025-10-02, which is inorganic and should be separable; operating income against the $151M of the March quarter, which was DOWN from $155M a year earlier despite 33% revenue growth; stock-based compensation against $1,051M for FY2025, 7.7% of revenue; and whether the effective tax rate begins to normalise from 0.64%. Also watch the international line, which grew from $1,319M to $2,257M in FY2025 and now includes three marketplaces. And note the insider record: the chief financial officer sold across five tranches on 2026-07-08 at $183.36 to $185.66."
- Confidence
- Low
Medium term 6-24 months
Tailwind- Driver
- "The medium-term case is the earnings ramp, and it is steep enough that most of the valuation lives in it. Consensus models EPS of $2.533 in FY2026, $4.486 in FY2027 and $7.117 in FY2028 — 12.4%, then 77.1%, then 58.6% growth — against revenue growing 28.6%, 20.7% and 18.6%. That is operating leverage on a scale nothing else in this batch requires: earnings growing three to four times faster than revenue for two consecutive years. The mechanism is credible on the reported record — operating income went from MINUS $38M (FY2024) to PLUS $723M (FY2025) on 27.9% revenue growth, and has been positive for seven consecutive quarters — and the marketplace model does have the shape that produces it, because incremental orders over an existing network carry very high margins. Layered on top: three marketplaces after Deliveroo, non-US revenue growing from $1,319M to $2,257M in a year, and $2,216M of net cash with no dividend obligation. Against it: two of the last four quarters missed by roughly 19%, March-quarter operating income FELL year on year despite 33% revenue growth, stock-based compensation is 7.7% of revenue, and the tax rate of 0.64% must normalise."
- What we’re watching
- "Whether operating leverage actually arrives. The March 2026 quarter is the warning: revenue +33.1% and operating income $151M against $155M — DOWN. If that pattern repeats, the FY2027 consensus of $4.486 is unreachable and a 45x multiple on it is meaningless. Watch: quarterly operating margin, which ran 3.7% in the March quarter against 5.1% a year earlier; stock-based compensation as a share of revenue against 7.7%; the Deliveroo integration and whether international contribution margins converge on the domestic ones; the effective tax rate normalising from 0.64%, which alone removes roughly a fifth of net income at a 21% rate; gig-labour regulation, which the knowledge base's independent analyst quantifies at roughly $7 of extra fee per $31 order under New York's driver minimum; and any disclosure of the autonomy programme's economics beyond the management description of an in-house level-4 robot in Phoenix."
- Confidence
- Low
Long term 2+ years
Tailwind- Driver
- "Long-run DoorDash is a bet that local commerce gets aggregated and then automated, and that the aggregator captures the automation. The aggregation half is largely won in its home market and is being bought abroad — three marketplaces, three membership programmes, non-US revenue up 71% in a year. The automation half is where the return would come from, because the independent analysis in the knowledge base puts the current blended contribution margin at roughly 3.1% on a $31 order, most of which is the cost of a human driving a car; removing that cost is the whole prize, and a management voice describes an in-house level-4 delivery robot that has been live in Phoenix for two years alongside 10 billion deliveries of route data. Both halves are contested and the store contains the contests. An independent analyst said in 2023 that DoorDash's robotics initiatives were already dead or significantly scaled back — flatly contradicted by the 2026 management description, which is why both are quoted and neither is underwritten. And the sharpest bear claim argues that AI agents will disintermediate aggregators and middlemen by transacting peer-to-peer, which would remove the layer DoorDash occupies rather than automate it."
- What we’re watching
- "Whether contribution margin rises above the roughly 3.1% the independent teardown found. That is the only long-run number that matters and the company does not disclose it in the form the analyst constructed it. Beyond that: whether autonomous delivery produces disclosed unit economics rather than management description; whether grocery, described in 2023 as losing about 10% per order, reaches parity with restaurant; whether gig-labour regulation spreads beyond New York; whether the three-marketplace structure consolidates onto one platform or persists as three cost bases; whether the accumulated deficit of $4,320 million is eliminated, which at the current pace takes about four years; and whether stock-based compensation falls below 5% of revenue, which is the test of whether the reported earnings are real."
- Confidence
- Low
Exponential Potential
"Rated 6 — the highest exponential score in this batch. DoorDash is a three-sided marketplace with real network effects (more merchants attract more consumers attract more Dashers), operating three marketplaces across more than thirty markets, and it is building two things on top that are genuinely non-linear. The first is data: the knowledge base carries a management claim of 10 billion deliveries of real-world data, which is the kind of proprietary corpus that cannot be bought. The second is autonomy, and here the specificity is unusual — a management voice describes an in-house DOT robot at 300 pounds and 20 miles per hour with level-4 autonomy, live in Phoenix for two years. If local delivery is eventually performed by machines, the party that owns the demand aggregation, the merchant relationships and the route data is the natural owner of the economics, and DoorDash's blended contribution margin of roughly 3.1% on a $31 order — the independent analyst's figure — is precisely the number that autonomy attacks. Against all of that: both autonomy claims are MANAGEMENT claims and are half-weighted; the same store carries an independent 2023 claim that DoorDash's robotics initiatives were already dead or significantly scaled back; and the sharpest bear claim in the lane says AI agents disintermediate aggregators rather than empower them. A 6: genuine network economics, a genuinely proprietary data asset, an autonomy programme we can only see through management's eyes, and a credible argument that the aggregator layer is the thing that gets removed."
What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.
Deeper analysis
Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.
> ## ⚠ EARNINGS TOMORROW — 2026-08-05
>
> DoorDash reports second-quarter 2026 results on 2026-08-05, ONE DAY after this dive is struck. Vendor consensus is $0.4681 of EPS and $4,339.6M of revenue — the latter implying +32.1% year-on-year growth against the +33.1% just delivered, so the bar is essentially the current run rate.
>
> DoorDash MISSED consensus earnings per share by 19.5% in the September 2025 quarter and by 18.6% in the December 2025 quarter, before beating by 15.8% in March.
>
> Beta is 1.779 — the highest in these twelve dives by 55%. The verdict below is set with that timing as a direct input.
Reference table
| Street consensus | $248.50 (+22.8%) · median $237.50 (+17.4%) · high $350 (+73.0%) · low $190 (−6.1%) · 1 strong buy / 28 buy / 9 hold / ZERO sells across 38 analysts · vendor consensus label Buy |
| Valuation | 79.9x FY2026E ($2.533) · 45.1x FY2027E ($4.486) · 28.4x FY2028E ($7.117) · 95.3x trailing · EV/EBITDA 48.4x · free-cash-flow yield 1.99% · no dividend |
| THE GROWTH | March-quarter revenue $4,036M against $3,032M, +33.1% — more than double the rate of any other name in this batch. Five-year revenue compound rate 36.5%. Non-US revenue $2,257M (FY2025) against $1,319M (FY2024), +71% |
| THE PROFIT INFLECTION | Annual operating income: −$436M, −$452M, −$1,124M, −$579M, −$38M, and PLUS $723M in FY2025. Positive for seven consecutive quarters |
| THE WARNING INSIDE IT | March-quarter operating income was $151M against $155M — DOWN — on 33.1% revenue growth. Operating margin 3.7% against 5.1% |
| Balance sheet | Cash $4,378M + short-term investments $1,128M = $5,506M against $3,290M of debt → NET CASH $2,216M. Accumulated deficit −$4,320M. Stock-based compensation $1,051M — 7.7% of revenue and 112% of GAAP net income. Effective tax rate 0.64% |
| Corporate action ABSENT from the payload | The acquisition of Deliveroo plc COMPLETED 2025-10-02, part of $4,151M of fiscal-2025 acquisition spending. DoorDash now operates three marketplaces (DoorDash, Wolt, Deliveroo) and three membership programmes (DashPass, Wolt+, Deliveroo Plus) |
| Knowledge base — richest and most defect-laden in this batch | 12 name-level claims; 9 used; THREE documented defect classes fire at once. TWO management voices (a co-founder and a colleague, same 2026-07-23 interview, both excluded and half-weighted); ONE eli_lilly_ceo mis-attribution excluded on content; and the sharpest independent analysis is BEARISH — "[o]n an average ~$31 order DoorDash is lucky to make $1; blended contribution margin ~3.1%" |
| Technicals | +14.8% above the 50-DMA ($176.35), +6.3% above the 200-DMA ($190.34); −28.2% from the 52-week high of $281.74, +38.0% above the low of $146.60; RSI 58.8; MACD +5.36; beta 1.779 — highest in batch; 12-month −18.5% vs SPY +24.3% |
What the experts actually said 6 traceable claims on DASH · showing the highest-conviction voices
“DoorDash's 10 billion deliveries of real-world data plus in-house DOT robot (300lb, 20mph, L4, live in Phoenix 2yrs) are a hard-to-replicate moat for autonomous local commerce”
“Owning and operating end-to-end (first-party, e.g. DashMart) is longer-term more sustainable than sitting on top of third-party merchants; DashMart is a top-10 merchant in every city it operates.”
“Gig-labor regulation is a huge headwind; NYC's ~$22/hr driver minimum in 2024 implies ~$7 extra delivery fee on a $31 order, becoming unsustainable.”
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
Price & moving averages 12 months · 50 & 200-day averages · 52-week range
Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.
Data summary: last close $236.74, 19% above the 50-day average ($199), 26% above the 200-day average ($188) — an uptrend. 16% below the 52-week high of $282, 61% above the 52-week low of $147.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Data summary: price $236.74 is currently inside the band (band $198–$239).
RSI (14) momentum gauge · 0–100
Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 72.
MACD 12 / 26 / 9 · trend & momentum
The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.
Data summary: MACD is currently above its signal line by 1.09, positive momentum.
Relative performance vs S&P 500 & its sector (XLC (sector)), set to 100 a year ago
Solid = DASH · dashed = S&P 500 · dotted = XLC (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
1. The growth, the inflection, and the warning inside it
| Fiscal year | Revenue | growth | Gross profit | Operating income | Op. margin | Net income | Diluted EPS |
|---|---|---|---|---|---|---|---|
| 2020 | $2,886M | — | $1,518M | −$436M | −15.1% | −$461M | −$7.39 |
| 2021 | $4,888M | +69.4% | $2,550M | −$452M | −9.2% | −$468M | −$1.39 |
| 2022 | $6,583M | +34.7% | $2,995M | −$1,124M | −17.1% | −$1,365M | −$3.68 |
| 2023 | $8,635M | +31.2% | $4,046M | −$579M | −6.7% | −$558M | −$1.42 |
| 2024 | $10,722M | +24.2% | $5,180M | −$38M | −0.4% | $123M | $0.29 |
| 2025 | $13,717M | +27.9% | $6,979M | +$723M | +5.3% | $935M | $2.13 |
A 36.5% five-year compound revenue rate with no down year, and a profit inflection in the last two.
The quarterly record, and the last line is the one to read carefully:
| Quarter | Revenue | YoY | Operating income | Op. margin | GAAP diluted EPS | Reported EPS | vs est |
|---|---|---|---|---|---|---|---|
| Q1 2024 | $2,513M | — | −$61M | −2.4% | −$0.06 | — | — |
| Q2 2024 | $2,630M | — | −$201M | −7.6% | −$0.38 | — | — |
| Q3 2024 | $2,706M | — | $107M | 4.0% | $0.38 | — | — |
| Q4 2024 | $2,873M | — | $117M | 4.1% | $0.33 | — | — |
| Q1 2025 | $3,032M | +20.7% | $155M | 5.1% | $0.44 | $0.44 | vs $0.391 |
| Q2 2025 | $3,284M | +24.9% | $163M | 5.0% | $0.65 | $0.65 | vs $0.438 (+48.3%) |
| Q3 2025 | $3,446M | +27.3% | $258M | 7.5% | $0.55 | $0.55 | vs $0.683 (MISS −19.5%) |
| Q4 2025 | $3,955M | +37.7% | $147M | 3.7% | $0.48 | $0.48 | vs $0.590 (MISS −18.6%) |
| Q1 2026 | $4,036M | +33.1% | $151M | 3.7% | $0.42 | $0.42 | vs $0.363 (+15.8%) |
Three readings and they run in different directions.
The revenue line is exceptional and it accelerated. +20.7%, +24.9%, +27.3%, +37.7%, +33.1%. Nothing else in these twelve dives is within 15 points of that. Part of the Q4 and Q1 acceleration is Deliveroo, acquired 2025-10-02, and the payload does not permit the organic figure to be isolated.
The operating-margin line is the warning and it is not subtle. Operating margin ran 5.1%, 5.0%, 7.5%, 3.7%, 3.7% — and in the March 2026 quarter operating income of $151M was LOWER than the $155M of a year earlier, on 33.1% more revenue. The entire valuation case requires operating leverage; the most recent quarter delivered negative operating leverage. Deliveroo integration is the plausible explanation and the archive does not confirm it.
The consensus record is genuinely poor for a name at this multiple. +12.4%, +48.3%, −19.5%, −18.6%, +15.8%. Two misses of roughly 19% in four quarters, from 29 covering analysts, on a business whose forward multiple is 45x. That combination — a demanding multiple and an unpredictable print — is the single most important reason this dive does not act ahead of tomorrow's report.
Note that earn_cal epsActual matches inc_q GAAP diluted EPS EXACTLY in all five quarters ($0.44, $0.65, $0.55, $0.48, $0.42). Consensus for DoorDash is struck on GAAP, which means the $1,051M of stock-based compensation IS inside every earnings figure and multiple in this dive — a genuine quality point, and the same finding recorded on Airbnb in this batch.
2. Three marketplaces — and the acquisition the payload never saw
From the 10-Q filed 2026-05-06, verbatim:
> "The Company's primary offerings include the DoorDash Marketplace, the Wolt Marketplace, and the Deliveroo Marketplace (together, the 'Marketplaces'), and its Commerce Platform... The Company's Marketplaces also include consumer membership programs, DashPass, Wolt+, and Deliveroo Plus, which aim to lower transactional friction by reducing the delivery and service fees charged."
> "On October 2, 2025, the Company completed the acquisition of substantially all of the outstanding equity interests of Deliveroo plc ('Deliveroo'), which was accounted for under the acquisition method of accounting. The acquisition will strengthen the Company's position as a leading global platform in local commerce."
cf_a records acquisitionsNet of −$4,151 million for FY2025, against zero in FY2024 and FY2023 — the Deliveroo purchase plus SevenRooms. Goodwill rose from a lower base to $5,519 million and intangibles to $2,260 million; total assets rose from a smaller balance sheet to $19,659 million; and total debt rose from $522 million (2024) to $3,290 million (2025).
Nothing about the acquisition appears in any vendor field other than those balance-sheet and cash-flow movements. There is no seg_prod detail — the block reports a single line called "Reportable Segment" — so the Deliveroo contribution cannot be separated from organic growth anywhere in this payload. That is the single most consequential data gap in this file, because it is the difference between 33% organic growth and something materially lower.
Geography is the one place the payload does help, and it ties exactly:
| FY2025 | FY2024 | growth | FY2023 | FY2022 | |
|---|---|---|---|---|---|
| United States | $11,460M | $9,403M | +21.9% | $7,781M | $6,251M |
| Non-US | $2,257M | $1,319M | +71.1% | $854M | $332M |
| Sum | $13,717M | $10,722M | $8,635M | $6,583M | |
inc_a revenue | $13,717M | $10,722M | $8,635M | $6,583M | |
| Tie | EXACT | EXACT | EXACT | EXACT |
Non-US revenue has grown from $332M to $2,257M in three years — 89% compound — and is now 16.5% of the total. Note that Deliveroo closed on 2025-10-02, so it contributes only one quarter to the FY2025 figure; the FY2026 international line will step again for reasons that are inorganic.
3. Balance sheet, cash flow and earnings quality
| ($M) | 2025-12-31 | 2024-12-31 |
|---|---|---|
| Cash and equivalents | $4,378 | $4,019 |
| Short-term investments | $1,128 | — |
| Cash + short-term investments | $5,506 | — |
| Total current assets | $8,643 | — |
| Goodwill | $5,519 | — |
| Intangible assets | $2,260 | — |
| Total assets | $19,659 | — |
| Short-term debt | $0 | — |
| Long-term debt | $2,724 | — |
| Capital lease obligations | $566 | — |
| Total debt | $3,290 | $522 |
| NET CASH (incl. short-term investments) | $2,216 | $2,134 |
| Total stockholders' equity | $10,033 | $6,806 |
| Accumulated deficit | −$4,320 | −$5,154 |
Net cash of $2,216 million, no short-term debt, and an accumulated deficit that improved by $834 million in the year. Note that the vendor's netDebt of −$1,088M deducts cash only and omits the $1,128M of short-term investments — the documented omission — so true net cash is $2,216M and corrected enterprise value is approximately $85.96 billion against the vendor's $86.89 billion.
Cash flow:
| Year | OCF | Capex | FCF | SBC | SBC / revenue | Buyback |
|---|---|---|---|---|---|---|
| FY2022 | $367M | −$346M | $21M | $889M | 13.5% | −$400M |
| FY2023 | $1,673M | −$324M | $1,349M | $1,088M | 12.6% | −$750M |
| FY2024 | $2,132M | −$330M | $1,802M | $1,099M | 10.2% | −$224M |
| FY2025 | $2,431M | −$257M | $2,174M | $1,051M | 7.7% | $0 |
Free cash flow is computed by subtraction and the arithmetic is correct in all four years; capital expenditure of $257M is 1.9% of revenue, appropriate for a marketplace. The free-cash-flow yield of 1.99% (trailing, per km_ttm) stands as reported and no correction is applied.
Three earnings-quality facts that belong together and are stated together:
Stock-based compensation of $1,051 million is 7.7% of revenue and 112% of GAAP net income of $935 million. The trend is genuinely improving — 13.5%, 12.6%, 10.2%, 7.7% of revenue — and it is still the case that the company's reported profit is smaller than the equity it issued to employees.
The effective tax rate is 0.64%. effectiveTaxRateTTM is 0.006438 and FY2025 tax expense was $4 million on $939 million of pre-tax income — the company is still using accumulated losses. At a normalised 21% rate, FY2025 net income would have been approximately $742 million rather than $935 million, and the trailing multiple would be roughly 120x rather than 95x. Consensus presumably models the normalisation; the reader should know the reported figure does not.
And the buyback stopped. $400M, $750M, $224M, then zero in FY2025 — because the cash went to Deliveroo. There is no shareholder yield here at all: no dividend, no buyback, and a share count rising from 405.5M to 442.3M diluted over nine quarters.
4. Valuation — priced in or room?
At $202.37 (market cap $88.18B, ~435.7M shares implied; 442.3M diluted weighted-average in the March quarter):
| Trailing | FY2026E | FY2027E | FY2028E | FY2029E | |
|---|---|---|---|---|---|
| Consensus EPS | $2.124 | $2.533 (29) | $4.486 (29) | $7.117 (18) | $10.632 (10) |
| EPS range | — | $2.045–$3.455 | $3.541–$5.992 | $5.336–$9.019 | — |
| Range width | — | 69% | 69% | 69% | — |
| EPS growth | — | +12.4% (vs FY2025E $2.253) | +77.1% | +58.6% | +49.4% |
| P/E | 95.3x | 79.9x | 45.1x | 28.4x | 19.0x |
| Consensus revenue | $14,721M (TTM) | $17,633M (32) | $21,276M (33) | $25,241M (24) | $30,194M (23) |
| Revenue growth | — | +28.6% | +20.7% | +18.6% | +19.6% |
| EV/EBITDA | 48.4x | — | — | — | — |
| Free-cash-flow yield | 1.99% | — | — | — | — |
| Shareholder yield | ZERO (no dividend, no buyback) | — | — | — | — |
Estimate coverage is deep — 29 analysts on FY2026 and FY2027 EPS, 32-33 on revenue, 18 on FY2028 — and the FY2029 and FY2030 rows (10 and 8 analysts on EPS) are used only as directional cross-checks. Every forward EPS range is roughly 69% wide, which is unusually consistent and reflects genuine disagreement about the margin path rather than about revenue (the FY2027 revenue range is only 12.9% wide).
Read the two growth columns together, because that is where the entire valuation lives. Consensus models revenue growing 28.6%, 20.7% and 18.6% while EPS grows 12.4%, 77.1% and 58.6%. In FY2027 and FY2028, earnings are modelled to grow roughly THREE TIMES faster than revenue for two consecutive years. That is a very large operating-leverage assumption, and the March 2026 quarter delivered the opposite: 33.1% revenue growth with operating income down 2.6%.
est.ebitdaAvg and est.ebitAvg are REJECTED — fixed-ratio fabrication, and the magnitude here is the largest in this batch. ebitdaAvg is exactly 33.21% of revenueAvg and ebitAvg exactly 26.68% in every year FY2026 through FY2030. Tested against the actual: FY2025 EBITDA per inc_a was $1,689M on $13,717M — a 12.31% margin — against the block's 33.21%. The fabricated ratio is 170% too high, the worst distortion recorded in these twelve dives. A 33% EBITDA margin describes a mature software company, not a delivery marketplace running a 3.1% blended contribution margin. All forward valuation uses epsAvg. Twenty-fourth consecutive name.
Peer set is unusable. AMX, BIDU, CMCSA, NBIS, RDDT, RELX, SPOT, T, TME. There is not one delivery or local-commerce marketplace in it — no Uber, no Instacart, no Just Eat — and Comcast and AT&T are in it. Comcast is in this same research batch and carries DoorDash in ITS peer set too, which is a reciprocal error rather than a comparison. No peer multiple is drawn.
4a. What today's price assumes (the inversion)
At $202.37 — 79.9x FY2026 consensus, 45.1x FY2027, 28.4x FY2028, a 1.99% free-cash-flow yield and zero shareholder yield — the price embeds:
- Revenue of $17,633M in 2026 and $21,276M in 2027. (Consensus; 32 and 33 analysts.) The March quarter delivered $4,036M and tomorrow's consensus is $4,339.6M, so FY2026 needs roughly $9,258M in the second half, an average of $4,629M a quarter. Achievable at the current rate, and it includes a full year of Deliveroo against one quarter last year.
- Operating margin roughly triples by FY2028. (Our derivation from the EPS and revenue paths.) This is the most fragile assumption in the price, and the March quarter contradicts it directly: revenue +33.1%, operating income $151M against $155M, margin 3.7% against 5.1%. Consensus needs EPS to grow three times faster than revenue for two years; the most recent evidence is negative operating leverage.
- Stock-based compensation keeps falling as a share of revenue. (Our derivation.) From 13.5% (FY2022) to 7.7% (FY2025) — a genuine and continuing improvement — but still 112% of GAAP net income, and consensus is struck on GAAP so every dollar of it is already in the earnings figure.
- The tax rate normalises without breaking the earnings path. (Our derivation.) At 0.64% today and a statutory 21%, normalisation alone removes roughly a fifth of net income, and consensus must be absorbing it inside the FY2027 and FY2028 ramp.
- The market keeps paying 28-45x forward earnings on a growth path this steep. (Our number.) At 35x FY2027E the stock is $157; at 55x it is $247. DoorDash has no stable historical multiple — it was loss-making until 2024 — so there is no band to anchor on, which is why the fair-value range below is wide.
4b. The return bridge (why the multiple moves)
Expected return over the next twelve months decomposes as: EPS growth (+77.1%, from FY2026E $2.533 to FY2027E $4.486) + multiple drift (COMPRESSION; our base of $225 is 50.2x FY2027E against 45.1x today, but that comparison is mechanical — on a constant estimate year the base implies the market pays 31.6x FY2028E against 28.4x today) + shareholder yield (ZERO).
Note two things about this bridge that make DoorDash different from every other name in this batch.
First, there is no shareholder yield at all — no dividend, no buyback since FY2024, and a diluted share count that has risen 9.1% over nine quarters. Every dollar of return must come from earnings and the multiple.
Second, the mechanical rolldown is enormous and must not be mistaken for re-rating. 79.9x FY2026 becoming 45.1x FY2027 and 28.4x FY2028 at a constant price is arithmetic, driven by a consensus that models earnings nearly tripling in two years. Our base of $225 applies 31.6x to FY2028E, essentially the multiple the market pays today on that year, so it assumes NO re-rating whatever — the return is entirely the earnings arriving.
Which is the honest framing: DoorDash's expected return is a bet that consensus earnings are approximately right, with no yield and no multiple support underneath it.
4c. Variant perception (where we differ, what would surprise)
- On the level we are 9.5% BELOW the street's $248.50 and 11.2% above spot. We are more cautious than a covering community that carries 29 buys, 9 holds and ZERO sells — a distribution with no dissent at all on a stock that has fallen 18.5% in twelve months and missed consensus twice in four quarters. A zero-sell rating distribution is itself a caution.
- Our principal departure is on operating leverage, and it is evidence-based rather than theoretical. Consensus needs EPS to grow 77.1% in FY2027 and 58.6% in FY2028 against revenue growing 20.7% and 18.6%. The March 2026 quarter grew revenue 33.1% and operating income FELL from $155M to $151M. Watchable number: operating margin in tomorrow's print against the 3.7% of the March quarter and the 5.1% of a year earlier. A quarter above 6% would restore the leverage case; another sub-4% quarter on 30%+ revenue growth makes the FY2027 consensus of $4.486 look unreachable and 45x meaningless.
- We think the earnings quality is under-weighted by a covering community using GAAP consensus. Stock-based compensation of $1,051M is 112% of GAAP net income, and the effective tax rate is 0.64% against a 21% statutory rate — normalisation alone is roughly a fifth of net income. Both are inside the reported number and neither is inside the intuition of a 45x forward multiple.
- We treat the autonomy story as genuinely important and as unverifiable from this archive. A management voice describes an in-house DOT robot at 300 pounds and 20 miles per hour with level-4 autonomy, live in Phoenix for two years, alongside 10 billion deliveries of route data. If true and scalable, it attacks the 3.1% blended contribution margin an independent analyst measured on a $31 order, which is the entire economic problem of this business. It is a management claim, it is half-weighted per house rule, and the same store contains an independent 2023 claim that DoorDash's robotics initiatives were "already dead or significantly scaled back." Watchable event: any disclosed unit economics for autonomous delivery, which would be the first outside-verifiable evidence.
- Positive surprise that would force a re-rate: an operating margin above 6% in tomorrow's print with revenue growth above 30%, which would demonstrate the leverage; or separate disclosure of Deliveroo's contribution letting the organic growth rate be seen; or disclosed autonomous-delivery unit economics.
- Negative surprise that would break the thesis: a third earnings miss in five quarters; operating margin below 3.5% on 30%+ revenue growth; stock-based compensation rising as a share of revenue; gig-labour regulation spreading beyond New York on the scale the knowledge base quantifies (roughly $7 of extra fee on a $31 order); or evidence that Deliveroo is dilutive to contribution margin.
Synthos fair values
All three anchors are built off the FY2028 consensus EPS distribution (mean $7.117, low $5.336, high $9.019, 18 analysts) rather than FY2027, because the FY2027 figure of $4.486 is itself mid-ramp and applying a multiple to it double-counts the growth the market is already paying for. Cross-checked against FY2027 and FY2026.
- Bear ~$145 — 27.2x the FY2028 consensus LOW of $5.336, i.e. today's FY2028 multiple applied to the bottom of the range. Cross-check: 32.3x FY2027E mean; 57.2x FY2026E; 1.1% BELOW the 52-week low of $146.60. The scenario: operating leverage does not arrive, the March quarter's negative margin trend persists, the Deliveroo integration dilutes contribution margin, gig-labour regulation spreads, and a business growing 20% with a 4% operating margin gets a growth-disappointment multiple. −28.3%.
- Base ~$225 — 31.6x the FY2028 consensus MEAN of $7.117, i.e. the multiple the market pays today on FY2028, applied to FY2028 — no re-rating assumed. Cross-check: 50.2x FY2027E; 88.8x FY2026E; 20.1% below the 52-week high of $281.74; 9.5% below the street's $248.50. The scenario: revenue grows roughly as consensus models, operating leverage arrives on schedule from FY2027, stock-based compensation keeps falling as a share of revenue, and the multiple compresses mechanically as the earnings arrive. +11.2%.
- Bull ~$310 — 34.4x the FY2028 consensus HIGH of $9.019. Cross-check: 69.1x FY2027E; 29.2x FY2029E; 10.0% above the 52-week high. The scenario: the leverage arrives faster than modelled, international contribution margins converge on domestic, autonomous delivery produces disclosed unit economics, and the market pays a platform multiple for a business that has demonstrated it can compound revenue at 30% while expanding margin. +53.2%.
Base is 11.2% above spot; asymmetry 1.88:1 (28.3% down, 53.2% up), with ZERO shareholder yield. 1.88:1 is the second-best payoff ratio in this batch and it does not reach the 2:1 Tactical bar — and even if it did, taking a position the day before a print on a name that has missed by roughly 19% twice in four quarters would be indefensible. Watch, with tomorrow's operating margin as the trigger.
5. Knowledge base — twelve claims, three defect classes, and the bears are the specific ones
Raw hits: 15. Name-level entity claims on DoorDash/DASH: 12. Used: 9. Excluded or discarded: 6.
The sweep ran entity terms DASH, DoorDash, Wolt, Deliveroo, Instacart and CART, plus free text on DoorDash, food delivery, last-mile delivery and Deliveroo. This is the second-deepest name-level lane in this batch after Airbnb's, and it fires three documented defect classes simultaneously.
Defect one — TWO MANAGEMENT VOICES, correctly tagged, from a single interview
> 2026-07-23 · bullish 70 · horizon: thesis · entity: DASH · channel: no_priors · SPEAKER: Andy Fang · ROLE: MANAGEMENT
> "Ask DoorDash natural-language interface driving latent demand: 50% of restaurant trajectories order from new places, ~40% larger grocery baskets"
> 2026-07-23 · bullish 75 · horizon: thesis · entity: DASH · channel: no_priors · SPEAKER: Stanley Ting · ROLE: MANAGEMENT
> "DoorDash's 10 billion deliveries of real-world data plus in-house DOT robot (300lb, 20mph, L4, live in Phoenix 2yrs) are a hard-to-replicate moat for autonomous local commerce"
Andy Fang is a DoorDash co-founder. Both claims carry speaker_role: management and both are correctly tagged — which is worth recording, because the standing house note is that speaker_role is unreliable and must be judged by content. Here the field and the content agree, and the content is unmistakably company advocacy. Both are EXCLUDED from the conviction pool per house rule and quoted at half weight.
They are nonetheless the most specific claims in the lane and they are quoted for that reason. The DOT robot description — 300 pounds, 20 miles per hour, level-4 autonomy, live in Phoenix for two years — is falsifiable in a way that almost no management claim in this programme is, and the "10 billion deliveries of real-world data" figure is checkable. If both are accurate, the exponential score in this dive is too low; we have not raised it on management's word.
Defect two — the eli_lilly_ceo MIS-ATTRIBUTION, excluded exactly as documented
> 2020-03-30 · bullish 70 · entities: Amazon, DoorDash, Uber · channel: eli_lilly_ceo
> "COVID accelerates digital replacing bricks-and-mortar; a substantial number of Americans will stick with new online ordering/delivery patterns permanently."
A claim about pandemic-driven e-commerce substitution naming Amazon, DoorDash and Uber, filed under a pharmaceutical chief executive's channel. This is the documented eli_lilly_ceo mis-attribution defect — the same channel previously found carrying Pfizer, Uber, Starbucks and a direct-to-consumer pharmacy claim, one of which surfaced under CDNS. EXCLUDED on content mismatch, exactly as the policy requires, and reported rather than silently dropped.
Defect three — a stance question that resolves in favour of the tag
> 2025-09-13 · BEARISH 65 · entities: Google, Amazon, DoorDash · channel: cognitive_revolution
> "AI agents will disintermediate aggregators and middlemen (search, delivery, even regulators like the FDA) by transacting directly peer-to-peer on behalf of users."
Tagged bearish, and correctly so — DoorDash IS the aggregator this claim describes being removed. Unlike the Airbnb stance inversion caught elsewhere in this batch, no re-scoring is needed here. It is the sharpest structural bear case in the lane and it is the mirror image of the a16z bull claim below.
The independent lane — and the specific claims are the bearish ones
BULL:
> 2026-01-29 · bullish 85 · entities: DoorDash, Toast, Mercury, Sierra · channel: invest_like_the_best
> "AI-era stickiness comes from network effects (DoorDash), money flow (Mercury/Toast payments), hardware (Toast), or a unique asset."
> 2026-04-01 · bullish 82 · entities: Block, DoorDash · channel: a16z
> "Near/medium-term moats are distribution, network effects, licenses, regulatory posture, and hardware — can't vibe-code 50-60M monthly actives or Square hardware."
> 2025-12-16 · bullish 80 · entities: DoorDash, Affirm, Toast · channel: invest_like_the_best — a sophisticated private holder "persisting and buying more across the curve."
> 2025-08-22 · bullish 65 · entities: Uber, DoorDash · channel: all_in — the doubted-then-dominant pattern.
BEAR — and note that all three of the most quantitatively specific claims in the entire lane come from one independent analyst:
> 2023-05-25 · neutral 60 · horizon: fact · entity: DASH · channel: business_breakdowns · speaker: Matt Newburgh · role: independent
> "On an average ~$31 order DoorDash is lucky to make $1; blended contribution margin ~3.1% (losing ~10% on grocery, ~+5% on restaurant orders)."
> 2023-05-25 · bearish 68 · entity: DASH · same source
> "Gig-labor regulation is a huge headwind; NYC's ~$22/hr driver minimum in 2024 implies ~$7 extra delivery fee on a $31 order, becoming unsustainable."
> 2023-05-25 · bearish 62 · entity: DASH · same source
> "DoorDash's ghost-kitchen and robotics initiatives are already dead or significantly scaled back; pure-delivery demand isn't enough to fill them."
And one more, dated and market-based:
> 2026-03-24 · bearish 60 · entity: DoorDash · channel: compound_and_friends
> "DoorDash in a 45% drawdown reflects a hurting consumer — delivery is the first discretionary spend people cut when gas/costs rise."
Three observations that matter.
The bear claims are three years old and the most specific in the store. A 3.1% blended contribution margin on a $31 order is the kind of number this programme exists to surface, and it is the economic frame for everything else — including the autonomy case, which is an attack on exactly that number.
One bear claim is DIRECTLY CONTRADICTED by a management claim in the same lane. business_breakdowns (2023): robotics "already dead or significantly scaled back." no_priors (2026, management): a level-4 in-house robot live in Phoenix for two years. Both are quoted. Neither is underwritten. The three-year gap makes both potentially true.
The bulls are principled and the bears are numerical. The bull lane argues from network effects and distribution moats in general terms at conviction 80-85; the bear lane argues from unit economics and regulation with figures. In this programme, a numerical claim from an independent analyst outweighs a principled claim at higher conviction, and the score reflects it.
Conclusion. Breadth 6 channels, 9 used claims, net conviction POSITIVE-SPLIT. Concentration test: business_breakdowns/Matt Newburgh supplies 3 of 9 used claims (33%), at the threshold and all from a single 2023 interview — the concentration is noted and the claims are old rather than promotional, so no discount is applied beyond the age weighting.
6. Data integrity — what we rejected and why
Seven findings.
1. est.ebitdaAvg and est.ebitAvg carry the fixed-ratio fabrication signature, and the magnitude is the LARGEST in this batch. ebitdaAvg is exactly 33.21% of revenueAvg and ebitAvg exactly 26.68% in every year FY2026 through FY2030. Tested against the actual: FY2025 EBITDA of $1,689M on $13,717M is a 12.31% margin against the block's 33.21% — the fabricated figure is 170% too high. For context, the same test on MPC produced an 11.8% overstatement and on ABNB 75%. A 33% EBITDA margin describes a mature software business; the independent analysis in the knowledge base puts DoorDash's blended contribution margin at roughly 3.1%. All forward valuation uses epsAvg.
2. seg_prod contains a single line labelled "Reportable Segment" — and here the gap is consequential. DoorDash reports one segment, so the block is not wrong. But the company completed the acquisition of Deliveroo plc on 2025-10-02 and now operates three marketplaces, and NOTHING in the payload permits the acquired revenue to be separated from organic growth. The 33.1% March-quarter growth rate — the headline number of this dive — cannot be decomposed. Stated as the single most consequential data gap in this file. (Note the block also relabels across years: "Marketplaces and Platform Services" (2020, 2022, 2023), "Core Business" (2021) and "Reportable Segment" (2024, 2025) are the same single line under three names.)
3. seg_geo ties EXACTLY in all four available years — a clean result. United States $11,460M + Non-US $2,257M = $13,717M against inc_a revenue of $13,717M, and the same in FY2024, FY2023 and FY2022. It also carries the only decomposable growth signal in the payload: non-US revenue up 71.1% year on year and 89% compound over three years.
4. A $4,151 MILLION ACQUISITION PROGRAMME IS VISIBLE ONLY IN THE CASH FLOW AND BALANCE SHEET. cf_a records acquisitionsNet of −$4,151M for FY2025 against zero in FY2024 and FY2023 — the Deliveroo plc acquisition completed 2025-10-02 plus SevenRooms — and total debt rose from $522M to $3,290M, goodwill to $5,519M and intangibles to $2,260M. No vendor field names the transaction, describes the perimeter, or separates its contribution.
5. netDebt omits short-term investments — the documented omission. The vendor reports −$1,088M (cash $4,378M less debt $3,290M + minor items); adding the $1,128M of short-term investments gives true net cash of $2,216M, and corrected enterprise value of approximately $85.96 billion against the vendor's $86.89 billion — 1.1% lower. The corrected figure is used.
6. The effective tax rate of 0.64% makes every trailing profit metric temporary. effectiveTaxRateTTM is 0.006438 and FY2025 tax expense was $4 million on $939 million of pre-tax income. At a 21% statutory rate FY2025 net income would have been approximately $742 million rather than $935 million. netIncomePerShareTTM of $2.124, the 95.3x trailing multiple and netIncomePerEBTTTM of 0.9925 are all correct and all reflect a tax position that will normalise. Stated wherever a trailing figure is used.
7. Peer set contains no comparable and one reciprocal error. AMX, BIDU, CMCSA, NBIS, RDDT, RELX, SPOT, T, TME. There is not one delivery or local-commerce marketplace in the list. Comcast appears in DoorDash's peer set and DoorDash appears in Comcast's — both in this same batch — which is a reciprocal miscategorisation rather than a comparison. No peer multiple is drawn.
Capex and free cash flow — CHECKED AND CLEAN. FY2025 operating cash flow $2,431M less capex $257M equals free cash flow $2,174M; the subtraction is correct in all four years; capex of 1.9% of revenue is appropriate for a marketplace with no owned fleet. The 1.99% trailing free-cash-flow yield stands as reported.
Consensus basis — CHECKED AND CLEAN, and it is a quality point. earn_cal epsActual matches inc_q GAAP diluted EPS exactly in all five available quarters ($0.44, $0.65, $0.55, $0.48, $0.42), and the FY2025 epsAvg of $2.253 sits 5.8% above actual diluted EPS of $2.13. Consensus for DoorDash is struck on GAAP, so the $1,051M of stock-based compensation is inside every earnings figure and every multiple in this dive. The same finding was recorded on Airbnb in this batch and it is unusual in technology names.
Quote versus technical blocks. quote.yearHigh/yearLow report $285.50 / $143.30; tech.hi52/lo52 report $281.74 / $146.60. Differences of 1.3% and 2.3%. tech is used throughout. tech.max_dd_from_peak (−28.171%) is identical to tech.pct_from_hi.
Non-equity tripwire — checked and passed. Class A common stock, NASDAQ-listed since 2020-12-09. Price $202.37; beta 1.779 — the highest in this batch by 55%; volume 3.44M shares (~$697M of turnover); 52-week band $146.60-$281.74, a 92% range; no dividend. This is common equity. Note the multi-class structure — the founders hold super-voting stock not represented in the traded security.
7. Technicals, insiders and verdict
Technicals
- Price $202.37, +14.8% above the 50-day moving average of $176.35 and +6.3% above the 200-day average of $190.34.
- −28.2% from the 52-week high of $281.74; +38.0% above the low of $146.60. Position in the annual range: 41st percentile.
- RSI 58.8 and MACD +5.36 — mid-range and positive, consistent with a recovery in progress.
- Maximum drawdown from peak: −28.2%, identical to the current distance from the high.
- Beta 1.779 — the highest in this batch by 55% (the next is Emerson at 1.235). On a name reporting tomorrow, that is the operative fact.
- Relative performance: 3-month +17.3% vs SPY +7.6%; 6-month −2.6% vs SPY +11.1%; 12-month −18.5% vs SPY +24.3% and QQQ +30.8% — a 43-point deficit. A poor year, a strong quarter.
- Sentiment: 1 strong buy, 28 buy, 9 hold, ZERO sell across 38 analysts; consensus label Buy. Target consensus $248.50 (+22.8%), median $237.50, high $350 (+73.0%), low $190 — 6.1% below spot. A zero-sell distribution on a stock down 18.5% in twelve months that has missed consensus twice in four quarters is a caution, not a support.
Today's move. DASH closed UP 0.93% at $202.37 from $200.50 on 3.44M shares. No company-specific filing is dated 2026-08-04.
The honest read: the entry is unremarkable and the timing is the whole issue. 41st percentile of the annual range, RSI 59, fair value 11.2% above spot — and a print tomorrow on a 1.779-beta stock at 79.9x current-year earnings.
Insiders — the chief financial officer sold five tranches in one day
The file is dominated by Ravi Inukonda, Chief Financial Officer, who on 2026-07-08 exercised 4,068 options at $7.66 and then sold across at least five tranches at $183.36, $184.46, $185.66 and higher, taking the reported holding from 271,538 down through 269,638, 267,938, 266,038 and 263,933.
The reading, with the qualifications. The exercise price of $7.66 against sale prices above $183 is a 24-fold gain on an option granted long before the 2020 listing, so the transaction is an exercise-and-sell of a deeply in-the-money legacy award rather than a view on the current price. The tranching across five fills on a single day is a single order executing. And the prices — $183 to $186 — are 8-9% BELOW today's close.
There is not one open-market purchase in the file. No transaction by chief executive Tony Xu appears. A chief financial officer monetising a legacy option four weeks before a results date is routine and carries little signal; it is reported because it is the only discretionary transaction in the file and because there is no offsetting purchase.
Verdict
Watch.
Struck the day before a quarterly report, with that timing as a direct input.
The arithmetic. At $202.37 DoorDash trades at 79.9x the FY2026 consensus of $2.533, 45.1x FY2027's $4.486 and 28.4x FY2028's $7.117, with a 1.99% free-cash-flow yield and zero shareholder yield. Our base fair value of $225 is 11.2% above spot with 1.88:1 asymmetry ($145 bear, $310 bull) — the second-best payoff ratio in this batch and short of the 2:1 bar.
What is genuinely exceptional. Revenue growth of 33.1% in the March quarter — more than double any other name in these twelve dives — on a 36.5% five-year compound rate with no down year. A profit inflection from MINUS $38M of operating income in FY2024 to PLUS $723M in FY2025, positive for seven consecutive quarters. Non-US revenue up 71% in a year to $2,257M. Three marketplaces after the Deliveroo acquisition completed 2025-10-02. $2,216M of net cash, no dividend obligation, capital expenditure of 1.9% of revenue. A geography block that ties exactly. Consensus struck on GAAP, so the stock-based compensation is inside the earnings figure. And the highest growth and exponential scores in this batch.
What keeps it at Watch.
First, operating leverage has not appeared and the valuation requires a great deal of it. Consensus needs EPS to grow 77.1% in FY2027 and 58.6% in FY2028 against revenue growing 20.7% and 18.6% — roughly three times faster — and the March 2026 quarter grew revenue 33.1% while operating income FELL from $155M to $151M and the margin fell from 5.1% to 3.7%.
Second, the print is tomorrow and the record is unreliable. Two misses of 19.5% and 18.6% in the last four quarters, on a stock with a 1.779 beta.
Third, earnings quality has two known offsets. Stock-based compensation of $1,051M is 112% of GAAP net income, and the 0.64% effective tax rate must normalise — roughly a fifth of net income at a statutory rate.
Fourth, the most specific claims in the knowledge base are the bearish ones — a 3.1% blended contribution margin on a $31 order, and gig-labour regulation implying roughly $7 of extra fee — while the two most bullish claims are DoorDash management voices from a single interview, excluded from the conviction pool and half-weighted.
Fifth, there is no shareholder yield and no valuation floor. No dividend, no buyback since FY2024, a diluted share count up 9.1% over nine quarters, and no historical multiple band because the company was loss-making until 2024.
Pre-registered KILL criteria — what would take this to Avoid:
- A third earnings miss in five quarters.
- Operating margin below 3.5% on revenue growth above 30%, which would establish negative operating leverage as the pattern rather than the exception.
- Revenue growth below 20%, which would break the FY2027 consensus revenue path.
- Stock-based compensation rising as a share of revenue, after falling from 13.5% to 7.7% over four years.
- Gig-labour regulation spreading beyond New York on the scale the knowledge base quantifies.
- Evidence that the Deliveroo integration is dilutive to contribution margin, which the payload's single-segment structure would hide.
Pre-registered UPGRADE conditions — what would take this to Buy — Tactical:
- An operating margin above 6% in tomorrow's print with revenue growth above 30%. This is the operative trigger and it resolves in one day.
- A price below roughly $175 with the growth intact, which is 24.6x FY2028E and takes asymmetry above 2.4:1.
- Separate disclosure of Deliveroo's contribution, letting the organic growth rate be measured.
- Disclosed unit economics for autonomous delivery, which would be the first outside-verifiable evidence for the claim that the 3.1% contribution margin can be attacked.
- Two consecutive quarters of expanding operating margin on 25%+ revenue growth.
Where DASH fits in the Synthos Framework Portfolio. No position today, and the correct action is to let tomorrow's print resolve the only question that matters. This is a genuine growth asset — the fastest-growing and most structurally interesting business in these twelve dives — priced for an operating-leverage ramp that the most recent quarter did not deliver. A 1.5% starter on an operating margin above 6% with revenue growth above 30%, or on a price below $175; nothing on the anticipation of either. A 1.779-beta stock at 79.9x current-year earnings the day before results is a coin flip with leverage attached. Logged as a tracked Synthos call (Watch) as of 2026-08-04 at $202.37, with the fair-value anchors, kill criteria and both triggers all gradeable.
Single biggest risk: operating leverage that has not yet appeared. DoorDash's valuation is almost entirely a claim on future margin. Consensus models earnings per share of $2.533 (FY2026), $4.486 (FY2027) and $7.117 (FY2028) against revenue growing 28.6%, 20.7% and 18.6% — earnings compounding roughly three times faster than revenue for two consecutive years. The most recent evidence points the other way: the March 2026 quarter grew revenue 33.1% and operating income fell from $155 million to $151 million, with the operating margin down from 5.1% to 3.7%. The independent analysis in this knowledge base explains why that is hard — on a roughly $31 order the blended contribution margin is about 3.1%, losing roughly 10% on grocery and making roughly 5% on restaurant, because most of the economics go to the person driving the car. The company's own answer, from a management voice, is a 300-pound level-4 robot that has been operating in Phoenix for two years. That answer may well be right, and this dive will not underwrite it on management's word. Tomorrow's operating margin is the first evidence either way.
Provenance & disclosures
- Traceability: 12 knowledge-base claims name DoorDash or DASH as an entity; 9 used, 6 excluded or discarded, from 15 case-sensitive raw hits (breadth 6 channels, net conviction POSITIVE-SPLIT). This is the second-deepest name-level lane in this batch after Airbnb's and it fires THREE documented defect classes at once. (1) TWO MANAGEMENT VOICES, correctly tagged
role: management, from a single 2026-07-23no_priorsinterview: speaker Andy Fang — a DoorDash co-founder — bullish 70 on a natural-language ordering interface ("50% of restaurant trajectories order from new places, ~40% larger grocery baskets"), and speaker Stanley Ting, bullish 75, on "10 billion deliveries of real-world data plus in-house DOT robot (300lb, 20mph, L4, live in Phoenix 2yrs)". BOTH EXCLUDED from the conviction pool per house rule and quoted at half weight — and recorded as a case where thespeaker_rolefield and the content AGREE, which the standing note says cannot be assumed. (2) THEeli_lilly_ceoMIS-ATTRIBUTION: a 2020-03-30 claim on that channel naming Amazon, DoorDash and Uber on pandemic-driven digital substitution — not an Eli Lilly management claim at all — EXCLUDED on content mismatch exactly as the documented policy requires, and reported rather than silently dropped. (3) A STANCE CHECK that resolves in favour of the tag:cognitive_revolution2025-09-13, bearish 65, entities Google/Amazon/DoorDash — "AI agents will disintermediate aggregators and middlemen (search, delivery...) by transacting directly peer-to-peer" — correctly tagged, since DoorDash IS the aggregator described, and no re-scoring is needed (unlike the Airbnb inversion caught elsewhere in this batch). THE INDEPENDENT LANE: bulls areinvest_like_the_best2026-01-29 (85, network effects),a16z2026-04-01 (82, "can't vibe-code 50-60M monthly actives"),invest_like_the_best2025-12-16 (80) andall_in2025-08-22 (65); bears arebusiness_breakdowns/Matt Newburgh 2023-05-25 — THREE claims, and the most quantitatively specific in the entire lane: "[o]n an average ~$31 order DoorDash is lucky to make $1; blended contribution margin ~3.1%", gig-labour regulation implying "~$7 extra delivery fee on a $31 order", and robotics initiatives "already dead or significantly scaled back" — pluscompound_and_friends2026-03-24 (60) on a hurting consumer. Note that the robotics bear claim is DIRECTLY CONTRADICTED by the management claim above; both are quoted, neither underwritten, and the three-year gap makes both potentially true. Concentration test: Matt Newburgh supplies 3 of 9 used claims (33%), at the threshold, all from one 2023 interview; noted, and no discount applied beyond age weighting because the claims are numerical rather than promotional. - Data as-of: fundamentals — annual statements through 2025-12-31 from the 10-K filed 2026-02-18; quarterly results and the Deliveroo acquisition disclosure through 2026-03-31 from the 10-Q filed 2026-05-06 · estimates 2026-08-04 · prices 2026-08-04, quote timestamp 1785873601 = 2026-08-04T20:00:01Z ($202.37, +0.93%; 50-DMA $176.35; 200-DMA $190.34; RSI 58.8; MACD +5.36; beta 1.779) · knowledge-base claims 2026-08-04. The fiscal year is the calendar year. All figures come from the Synthos vendor data file for DASH or from the SEC filings in the DASH archive; no figure comes from memory, recall or external retrieval.
- Filing archive contents: 10-K filed 2026-02-18 (fiscal 2025); 10-Q filed 2025-11-06 (September 2025 quarter); 10-Q filed 2026-05-06 (March 2026 quarter — the source of the three-marketplace description and the Deliveroo acquisition disclosure); 8-K filed 2026-02-18 (fourth-quarter results; no extractable item content); 8-K filed 2026-05-06 (first-quarter results; no extractable item content); 8-K filed 2026-06-12 (Items 5.07 and 8.01 — annual-meeting voting results and updated proxy-proposal contact details for the 2027 annual meeting). All six documents carry preserved
[TABLE]statement data. The June 2026 quarter is not in this archive and is not reported anywhere in this dive; it is released 2026-08-05. - Where the vendor was REJECTED:
est.ebitdaAvg/est.ebitAvg, exactly 33.21% and 26.68% ofrevenueAvgin all five forward years against an FY2025 ACTUAL EBITDA margin of 12.31% — the fabricated ratio is 170% TOO HIGH, the largest distortion recorded in these twelve dives (for comparison, MPC's was 11.8% and ABNB's 75%); all forward valuation usesepsAvg, and the FY2029 (10 analysts) and FY2030 (8) rows are used only as directional cross-checks.seg_prod, a single line relabelled across years ("Marketplaces and Platform Services", "Core Business", "Reportable Segment") — technically correct for a single-segment reporter and CONSEQUENTIALLY incomplete, because the Deliveroo acquisition completed 2025-10-02 cannot be separated from organic growth anywhere in the payload, so the headline 33.1% March-quarter growth rate cannot be decomposed.netDebtof −$1,088M, which omits $1,128M of short-term investments — true net cash is $2,216M and corrected enterprise value approximately $85.96bn against the vendor's $86.89bn. The peer set, which contains no delivery or local-commerce marketplace at all and includes Comcast — which is in this same batch and carries DoorDash in ITS peer set, a reciprocal miscategorisation; no peer multiple is drawn.quote.yearHigh/yearLow($285.50/$143.30) differ fromtech.hi52/lo52($281.74/$146.60) by 1.3% and 2.3%;techis used. - Where the vendor was CLEAN and it is recorded:
seg_geoties EXACTLY to the dollar in all four available years and carries the only decomposable growth signal in the payload (non-US revenue up 71.1% year on year and 89% compound over three years); capital expenditure and free cash flow, with the subtraction correct in all four years and capex of 1.9% of revenue appropriate for a marketplace, so the 1.99% free-cash-flow yield stands as reported; and the consensus basis —earn_calepsActualmatchesinc_qGAAP diluted EPS EXACTLY in all five available quarters, so consensus is struck on GAAP and the $1,051M of stock-based compensation is inside every earnings figure and multiple in this dive. That last point is unusual in a technology name and is recorded as a quality finding, as it was on Airbnb in this batch. - Corporate action absent from the payload — ONE, and it is large: the acquisition of substantially all of the outstanding equity interests of Deliveroo plc, COMPLETED 2025-10-02, part of $4,151 million of fiscal-2025 acquisition spending (against zero in FY2024 and FY2023), which took total debt from $522M to $3,290M, goodwill to $5,519M and intangibles to $2,260M, and which gives DoorDash three marketplaces (DoorDash, Wolt, Deliveroo) and three membership programmes (DashPass, Wolt+, Deliveroo Plus). No vendor field names the transaction, describes the perimeter or separates its contribution.
- An earnings-quality note that is not a defect: the effective tax rate is 0.64% (
effectiveTaxRateTTM0.006438; FY2025 tax expense of $4 million on $939 million of pre-tax income). At a 21% statutory rate FY2025 net income would have been approximately $742 million rather than $935 million, and the trailing multiple roughly 120x rather than 95.3x. Every trailing profit figure in this dive reflects a tax position that will normalise, and it is stated wherever one is used. Separately, stock-based compensation of $1,051 million is 7.7% of revenue and 112% of GAAP net income — improving from 13.5% of revenue in FY2022 and still larger than the reported profit. - Estimate coverage: 29 analysts on FY2026 and FY2027 EPS and 32-33 on revenue — deep — with every forward EPS range approximately 69% wide while the FY2027 revenue range is only 12.9% wide, i.e. the disagreement is entirely about margin and not about revenue; 18 on FY2028 EPS; 10 and 8 on FY2029 and FY2030, used only as directional cross-checks.
- Fair-value caveat: the $145 / $225 / $310 anchors are multiples of the FY2028 consensus EPS distribution (mean $7.117, low $5.336, high $9.019, 18 analysts) rather than FY2027, because the FY2027 figure of $4.486 is itself mid-ramp and applying a multiple to it double-counts the growth the market is already paying for — 27.2x the low, 31.6x the mean, which is the multiple the market pays TODAY on FY2028, so the base assumes NO re-rating whatever, and 34.4x the high. Cross-checked against FY2027 and FY2026 and against the 52-week range. Stated arithmetic, not a discounted cash flow. Note that the mechanical rolldown from 79.9x FY2026E to 28.4x FY2028E at a constant price is arithmetic driven by a consensus that models earnings nearly tripling in two years, and is nowhere in this dive described as re-rating. Note also that DoorDash has NO shareholder yield — no dividend, no buyback since FY2024, and a diluted share count up 9.1% over nine quarters — so the entire expected return is earnings arriving.
- Timing: second-quarter 2026 results are due 2026-08-05, ONE DAY after this dive, against consensus of $0.4681 in EPS and $4,339.6M of revenue (+32.1% year on year). DoorDash MISSED consensus EPS by 19.5% in the September 2025 quarter and by 18.6% in the December 2025 quarter, beating by 15.8% in March. The only discretionary insider transaction in the file is chief financial officer Ravi Inukonda's exercise-and-sell of 2026-07-08, converting options struck at $7.66 and selling across five tranches at $183.36 to $185.66 — 8-9% below today's close — with no offsetting purchase and no transaction by chief executive Tony Xu. 2026-08-04 carried no company-specific filing.
- Accessibility note: no information in this dive is conveyed by colour. All emphasis is carried by bold text, table structure and explicit labelling.
- Not investment advice. Independent research, educational and informational only, never personalised. No recommendation to buy, sell or hold any security is made to any person.
- Version: 2026-08-04-full.