TTDNASDAQThe short version
The Trade Desk, Inc.
The Trade Desk runs the buy-side platform agencies use to buy connected-TV, video, display and audio ads across the open internet, taking a percentage fee on every dollar of spend it routes but never owns.
From a ~$3 IPO in 2016 the shares compounded to an all-time high of $139.51 in December 2024, then fell roughly 86% to $19.53 by July 2026 as growth slowed.
$19.53
Share price
$2.90B
Revenue (FY25)
21.6%
Take rate (rev ÷ spend)
27%
Free-cash-flow margin
SwipeScroll▾
The statements
A profitable, debt-free, cash-generative business — the growth is what slipped
FY2021 → FY2025as reported · $
Revenue$2.9B+18%
Operating margin20.3%+2.9pp
Net income$443M+13%
EPS$0.90+15%
Free cash flow$796M+24%
Open the full statements →As reported, FY2025 annual report (Form 10-K).
- Real profits, real cash. FY2025 revenue rose 18% to $2.90B, operating margin widened to 20%, and free cash flow hit a record $796M — a 27% margin, with zero debt and ~$1.3B of net cash.
- An intermediary balance sheet. Receivables of $3.77B and payables of $3.01B dwarf revenue because ~$13.4B of client ad spend flows across the platform; TTD keeps only the fee. This is plumbing, not leverage.
- The catch. Nearly two-thirds of that free cash flow is added-back stock compensation, and revenue growth has cooled from the mid-20s to 12% — the tension the rest of this briefing unpacks.
The setup
An 86% drawdown re-rated an elite compounder down to a value multiple
Split-adjusted close, 2016–2026.
- The multiple did the damage, not the business. Price-to-sales collapsed from ~24x at the end of 2024 to ~3.3x by mid-2026, even as revenue kept growing 18% and cash flow set records.
- Three guidance shocks, not an earnings collapse. The stock fell ~41% in February 2025 on its first revenue-guidance miss in 33 quarters, ~39% again that August, then ground lower through 2026.
- The whole debate lives in that gap. The market now prices a former hyper-grower like a structurally impaired one; whether that is an overshoot or a fair call is the question this report weighs.
The business model
A percentage fee on a $13.4B river of ad spend it never owns
Gross spend vs revenue ($M) — the river and the toll
Revenue is recognized net; the take rate rose to 21.6% in FY2025 from 20.3%.
- The master variable is the take rate. Revenue divided by gross spend — the slice TTD keeps — stepped up to 21.6% in FY2025 from 20.3%, on more data, measurement and higher platform fees.
- Capital-light and sticky. The company owns no media and carries no debt; clients contract through ongoing master agreements, and customer retention has topped 95% every year for over a decade.
- Why it matters. A rising take rate is the bull's proof of pricing power; a falling one is management's own stated warning sign. Every quarter, this is the number that decides the story.
The crux
Nine straight quarters of slowing growth — 28% down to 12%
Quarterly revenue growth, year over year
Per quarterly results; Q1 FY2026 revenue was $689M.
- The inflection was self-inflicted. In Q4 2024 TTD missed its own guidance for the first time in 33 quarters — the CEO called it "our fault," blaming a slow rollout of its Kokai platform upgrade and a reorganization.
- Cyclical or structural? Management blames macro pressure on Fortune 500 advertisers; bears see Amazon's rival platform eroding share. No print yet settles which it is — and that is the entire question.
- The next read is dated. Q2 results and the Q3 guide land August 6, 2026 — the cleanest look at whether 12% was a trough or the front edge of a permanent slowdown.
The moat
Genuine switching costs and scale — with one advantage still unproven
Moat scorecard
| Source | Evidence | Verdict |
|---|---|---|
| Switching costs | Retention >95% for a decade+ | Proven |
| Data network effect | 370+ data vendors integrated | Proven |
| Scale economics | Best margin of any independent | Proven |
| Neutrality (no media) | Buy-side only vs walled gardens | Real but narrowing |
| Identity standard (UID2) | Promoted industry-wide; still adopting | Unproven |
Per FY2025 filings.
- The evidence is the retention rate. A customer-retention rate above 95% for more than ten years is rare in any software category, and is the single strongest piece of moat evidence in the filings.
- Neutrality is the differentiator under pressure. Owning no media, TTD never bids against its own clients — a real edge over Google and Amazon, but the moat most exposed as Amazon leans into its own platform.
- UID2 is optionality, not yet an advantage. TTD wants its Unified ID 2.0 to replace the cookie; Google's 2025 decision to keep cookies in Chrome removed the near-term catalyst that would have forced adoption.
The competition
The only independent earning a real margin — dwarfed by the walled gardens
Operating margin by independent (FY2025)
TTD also earns roughly 4x the revenue of the next-largest independent.
- Scale compounds into economics. TTD's ~20% operating margin towers over Magnite (14%), Viant (4%) and a loss-making PubMatic — evidence that buy-side scale, not just size, is the advantage.
- The real threat isn't on this table. It's Amazon's platform — the one rival with the data, scale and retail-media flywheel to contest buy-side share and pressure the take rate directly.
- Leverage sits with a few agencies. Aggregated to the parent, two holding companies each represented more than 10% of gross billings in 2025 — genuine negotiating power over the take rate.
The growth engine
Connected TV and retail data are the durable tailwind
High-40s%
Video/CTV share of platform spend
>80%
Top U.S. retailer sales in its data marketplace
85%
Share of FY2025 revenue from the U.S.
370+
Third-party data vendors integrated
- TV budgets are migrating to biddable streaming. Video, which includes connected TV, has grown to a high-40s percentage of platform spend and keeps rising — the reason an 18% grower still talks of a long runway.
- Retail data is the flywheel against Amazon. The marketplace aggregates the rest of retail — over 80% of top U.S. retailer sales — against Amazon's under-15% share, a coalition Amazon cannot offer neutrally.
- International is the second leg. The U.S. was 85% of FY2025 revenue, leaving overseas markets as a largely untapped extension of the same model.
Quality of earnings
Strip out stock compensation and the "free cash flow" shrinks by two-thirds
FY2025 free cash flow, adjusted for stock comp ($M)
Reported FCF $795.7M less $490.6M of stock-based compensation.
- The 27% margin is largely non-cash. Reported free cash flow of $796M adds back $490.6M of stock-based compensation — a real cost that dilutes shareholders whether or not it touches cash.
- On owner cash, the multiple flips. Charge that dilution honestly and free cash flow is nearer $305M — so on enterprise value the business trades near 27x true cash flow, not the ~12x the headline advertises.
- One tailwind won't repeat. FY2025 operating cash was flattered by a ~$168M one-time deferred-tax lift from a July-2025 tax law, and the net-cash cushion roughly halved in a single year.
Governance
Clean books, but independent oversight collapsed in real time
38/100
Forensic risk score — "Watch", clean end of band
1
Members left on the audit committee
3
CFOs in roughly 12 months
~49.9%
Insider voting control on ~11% of economics
- The accounting itself is clean. Revenue is recognized conservatively net of media, cash conversion is genuine, there's no acquisition roll-up, and the auditor issues clean opinions. The risk is optics and control, not fabrication.
- Oversight thinned exactly when it mattered. All three audit-committee members resigned before the 2026 annual meeting, leaving one; a live class action asserts a Section 20A insider-trading claim against the CEO and CFO.
- Founder control is entrenched. Ten-vote Class B shares hand insiders ~49.9% of the vote on ~11% of the economics, and an $819M disputed CEO mega-option sits deeply underwater, its first hurdle over 4x away.
Capital allocation
$1.4B of buybacks in 2025 barely moved the share count
Buybacks vs stock-based compensation ($M)
Diluted share count fell only from 501.9M to 493.6M in FY2025.
- No dividend, almost no capex — it's all buybacks. The company spent $1.38B repurchasing 26.2M shares in 2025, yet the diluted share count fell only modestly, from 501.9M to 493.6M.
- A treadmill, mostly. Management is explicit the program is "designed to offset dilution" from employee stock; much of the cash mops up what stock comp creates rather than compounding per-share ownership.
- Cheaper shares make it more accretive. Buying back stock near $20 that traded above $100 in 2024 retires far more shares per dollar — the depressed price is the one thing working in the treadmill's favor.
What's next
The August 6 print is the next clean read on the one question that matters
Recent developments
| When | Event |
|---|---|
| Aug 6, 2026 | Q2 results and Q3 guide — revenue vs gross spend |
| Mar–Jun 2026 | Publicis pulled, then restored, TTD spend in a fee dispute |
| Apr 2026 | CEO bought ~$150M of stock at the lows |
| Apr 2026 | Launched Koa agentic-AI tools for media buying |
| Sep 21, 2026 | Deadline to cure Nasdaq audit-committee independence |
Next earnings date per the earnings-calendar feed.
- One quarter won't decide a decade, but it moves the odds. With no debt and net cash, no single print is existential — each just shifts the probability the market assigns to cyclical versus structural.
- Signals cut both ways. The CEO's ~$150M open-market purchase reads as a floor and alignment; the Publicis dispute was a live reminder that a couple of agency holding companies hold real leverage.
- The skew has shifted. Earnings-day drops have shrunk from ~35% to low single digits even as growth slid — a sign bad news is increasingly in the price ahead of August 6.
Valuation
Cheap on almost every lens — demanding on the one that counts stock comp
12–18 month price scenarios
Bear case
$14
Consensus mean
$24.42
Bull case
$40
Bull/bear targets from this report; consensus mean per ~36 analysts.
- The bull underwrites the franchise. ~3.3x sales and an 8.3% free-cash-flow yield for the only durably profitable independent DSP, still growing 18% — a ~$40 target on a re-rate to ~22x forward earnings.
- The bear charges for dilution. On stock-comp-adjusted owner cash the yield is closer to 3.2%, growth is decelerating, and a de-rated grower earns no premium — a downside case near $14 at ~2x sales.
- The report's stance is watchlist. At ~3x sales the price has already done most of the work of pricing the risks; the take rate and the next few growth prints now decide the outcome.
What to watch
A genuinely cheap, cash-rich franchise whose growth slowdown and eroding oversight leave one question unresolved
- 01Revenue re-accelerating toward the high teens with the take rate intact — the bull's confirmation
- 02A sub-10% growth print, or two quarters of revenue growing slower than gross spend — the bear's
- 03The Q2 FY2026 print and Q3 guide on August 6, 2026
- 04Whether buybacks finally shrink the share count rather than just offsetting stock-comp dilution
This distills the full report — business, numbers, moat, competition, forensics, catalysts and the bull-bear verdict — tab by tab.
Compiled from the full report · 2026-07-12 · For information, not investment advice.