The Trade Desk
Rating
Speculative Buy
Higher Risk / Asymmetric Reward
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
The Trade Desk remains the largest independent open-internet DSP, but Q2 2026's +3% growth against mid-teens-to-50%+ growth at Meta, Amazon Ads, Google, and AppLovin made relative share loss concrete. Amazon DSP's near-zero take-rate and spending matches, plus Google's Buyer Direct fee cap, are compressing TTD's ~20% take-rate model. Publicis resumed recommending TTD in June 2026 after the March fee dispute, removing that overhang — the binding constraint is now walled-garden competitive pressure, not agency politics.
TTD's competitive position rests on Scale, Independence from Walled Gardens, and UID2 Identity Infrastructure — all under active pressure from Amazon DSP and Google DV360:
- Scale & Open Internet Independence: The Trade Desk is still the largest independent DSP, and customer retention stayed above 95% in Q2 2026. Advertisers who want programmatic CTV, audio, display, and DOOH without buying through a seller's own tool still have few scaled alternatives. That independence thesis is intact as a product claim, but Q2 showed it is not enough to hold growth when Amazon and Google subsidize DSP fees from first-party media businesses TTD cannot match.
- UID2 — Open Identity Infrastructure: Unified ID 2.0 (UID2) remains TTD's privacy-era identity framework, with broad publisher adoption. New Q2 integrations (Databricks CustomerLake launch partner, Adobe Real-Time CDP exposure linkage, Netflix Sellers and Publishers 500+, Samsung Ads home-screen access) deepen open-internet addressability. UID2's open-source nature still limits proprietary lock-in versus Amazon's purchase graph and Google's intent data.
- Kokai AI Platform: Kokai is TTD's AI-powered campaign interface. Management spent Q2 acknowledging execution gaps and said resources will concentrate on a smaller set of high-priority initiatives through 2026–2027. Kokai lowers adoption friction but also reduces interface lock-in as AI makes DSP front-ends more interchangeable — a double-edged moat effect that matters more as Amazon and Google ship AI-native buying tools on larger proprietary datasets.
Ten Moats Verdict
TTD is approximately neutral on AI — Kokai AI enhances platform usability and advertiser ROAS, positioning TTD as a beneficiary of AI-driven ad optimization. However, AI also strengthens competitors: Google's Gemini-powered Performance Max and Meta's Advantage+ are AI-native systems with more first-party data, making the competitive pressure from walled gardens stronger, not weaker. TTD's primary moat risk in the AI era is that AI commoditises DSP interfaces (reducing learned-interface lock-in) while simultaneously enabling Amazon and Google to offer superior performance through larger proprietary datasets. The open-internet independent position remains relevant, but the margin of advantage over competitors is narrowing — Q2 2026 made that narrowing visible in the growth differential.
40.0 resilient · 53.8 vulnerable · 80/20 = 42.8 · = 43
Programmatic traders invest significant time learning TTD's platform for audience targeting, bid optimization, and reporting. Kokai AI reduces this learning curve somewhat, but institutional media-buying workflows built on TTD represent multi-year configuration investments.
Advertiser audience segments, brand safety block lists, frequency capping rules, and cross-channel attribution models are configured per-account on TTD over months. Porting this logic to Amazon DSP or DV360 requires rebuilding campaign architecture.
TTD's data marketplace provides access to third-party data segments from 100+ data providers. UID2 plus Q2 2026 partnerships (Databricks CustomerLake, Adobe RT-CDP, commerce-media signals from Booking/Marriott/Uber) keep open-internet addressability relevant, though still weaker than walled-garden first-party graphs.
Programmatic trading skills are increasingly portable across DSP platforms. Kokai's AI abstraction layer further reduces the specialization premium for TTD-specific expertise. Talent scarcity is not a meaningful moat for TTD.
TTD is primarily a single-product DSP without the multi-layer bundling of Google (search+display+video+YouTube) or Amazon (DSP+search+streaming). Some bundling through Solimar/Kokai with data and measurement, but limited.
UID2 is open-source, limiting TTD's proprietary data advantage. Amazon's purchase data and Google's search intent remain structural disadvantages; Q2's competitive fee pressure made that gap more commercially relevant as buyers optimize for media+data bundles TTD cannot subsidize.
Privacy regulations (GDPR, CCPA) created headwinds for cookie-based targeting that TTD navigated well with UID2. However, TTD does not benefit from meaningful regulatory lock-in in the traditional sense — no government mandates, certifications, or multi-year procurement cycles.
Buyer↔publisher liquidity effects still exist, and retention remains >95%, but Q2 2026's +3% growth versus mid-teens-to-50%+ growth at Amazon Ads, Meta, Google, and AppLovin is evidence the flywheel is losing relative share. Amazon's fee subsidies and Google's Buyer Direct are diverting incremental open-internet budgets that previously reinforced TTD's two-sided network.
TTD processes billions of ad auction transactions daily, sitting in the critical path of open-internet programmatic advertising. Switching means rebuilding trading infrastructure, campaign history, and attribution baselines — non-trivial, and consistent with >95% customer retention even through the Q2 miss.
TTD's reporting and attribution data does not function as a legal or financial system of record. Campaign performance data can be exported and recreated on alternative platforms. This is a notable moat weakness relative to FICO, Axon, or MSCI.
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
The Trade Desk remains the largest independent open-internet DSP, but Q2 2026's +3% growth against mid-teens-to-50%+ growth at Meta, Amazon Ads, Google, and AppLovin made relative share loss concrete. Amazon DSP's near-zero take-rate and spending matches, plus Google's Buyer Direct fee cap, are compressing TTD's ~20% take-rate model. Publicis resumed recommending TTD in June 2026 after the March fee dispute, removing that overhang — the binding constraint is now walled-garden competitive pressure, not agency politics.
Growth Score
Q2 2026 revenue was $715M (+3% YoY), missing the company's own ≥$750M guide; adj. EBITDA fell to $241M (34% margin vs 39% a year ago). H1 revenue was $1.404B (+7%). Q3 guidance of ≥$650M (−12% vs Q3 2025's ~$739M) and ~$160M adj. EBITDA (~25% margin) is pandemic-scale contraction while Meta, Amazon Ads, Google, and AppLovin still grow double digits — pointing to share and/or take-rate pressure, not just a soft ad cycle. Bright spots: CTV and audio again grew double digits; EMEA/APAC each ~+30% YTD with CTV +50%+ in both; accounts outside the top 500 grew ~50% YoY; video (incl. CTV) is a low-50% share of the mix. Publicis resumed recommending TTD in June, so the March agency overhang is cleared — the growth problem is competitive.
Valuation Score
TTD trades near ~$13.60 / ~$6.4B after a ~22% post-earnings drop — below the prior $17 bear and ~80% off the 52-week high. At ~2.2× NTM P/S on a ~$2.9B revenue base and ~16× trailing GAAP P/E on $0.84 TTM EPS, the stock sits between bear ($8) and base ($16). The de-rating has already priced a growth reset; it has not priced a durable re-acceleration. Street targets post-print cluster roughly $11–20. Cheap only works if Q3 clears the $650M floor with margin stabilization.
The Open Internet DSP Flywheel
TTD's competitive position rests on Scale, Independence from Walled Gardens, and UID2 Identity Infrastructure — all under active pressure from Amazon DSP and Google DV360:
- Scale & Open Internet Independence: The Trade Desk is still the largest independent DSP, and customer retention stayed above 95% in Q2 2026. Advertisers who want programmatic CTV, audio, display, and DOOH without buying through a seller's own tool still have few scaled alternatives. That independence thesis is intact as a product claim, but Q2 showed it is not enough to hold growth when Amazon and Google subsidize DSP fees from first-party media businesses TTD cannot match.
- UID2 — Open Identity Infrastructure: Unified ID 2.0 (UID2) remains TTD's privacy-era identity framework, with broad publisher adoption. New Q2 integrations (Databricks CustomerLake launch partner, Adobe Real-Time CDP exposure linkage, Netflix Sellers and Publishers 500+, Samsung Ads home-screen access) deepen open-internet addressability. UID2's open-source nature still limits proprietary lock-in versus Amazon's purchase graph and Google's intent data.
- Kokai AI Platform: Kokai is TTD's AI-powered campaign interface. Management spent Q2 acknowledging execution gaps and said resources will concentrate on a smaller set of high-priority initiatives through 2026–2027. Kokai lowers adoption friction but also reduces interface lock-in as AI makes DSP front-ends more interchangeable — a double-edged moat effect that matters more as Amazon and Google ship AI-native buying tools on larger proprietary datasets.
Ten Moats Verdict
TTD is approximately neutral on AI — Kokai AI enhances platform usability and advertiser ROAS, positioning TTD as a beneficiary of AI-driven ad optimization. However, AI also strengthens competitors: Google's Gemini-powered Performance Max and Meta's Advantage+ are AI-native systems with more first-party data, making the competitive pressure from walled gardens stronger, not weaker. TTD's primary moat risk in the AI era is that AI commoditises DSP interfaces (reducing learned-interface lock-in) while simultaneously enabling Amazon and Google to offer superior performance through larger proprietary datasets. The open-internet independent position remains relevant, but the margin of advantage over competitors is narrowing — Q2 2026 made that narrowing visible in the growth differential.
40.0 resilient · 53.8 vulnerable · 80/20 = 42.8 · = 43
Programmatic traders invest significant time learning TTD's platform for audience targeting, bid optimization, and reporting. Kokai AI reduces this learning curve somewhat, but institutional media-buying workflows built on TTD represent multi-year configuration investments.
Advertiser audience segments, brand safety block lists, frequency capping rules, and cross-channel attribution models are configured per-account on TTD over months. Porting this logic to Amazon DSP or DV360 requires rebuilding campaign architecture.
TTD's data marketplace provides access to third-party data segments from 100+ data providers. UID2 plus Q2 2026 partnerships (Databricks CustomerLake, Adobe RT-CDP, commerce-media signals from Booking/Marriott/Uber) keep open-internet addressability relevant, though still weaker than walled-garden first-party graphs.
Programmatic trading skills are increasingly portable across DSP platforms. Kokai's AI abstraction layer further reduces the specialization premium for TTD-specific expertise. Talent scarcity is not a meaningful moat for TTD.
TTD is primarily a single-product DSP without the multi-layer bundling of Google (search+display+video+YouTube) or Amazon (DSP+search+streaming). Some bundling through Solimar/Kokai with data and measurement, but limited.
UID2 is open-source, limiting TTD's proprietary data advantage. Amazon's purchase data and Google's search intent remain structural disadvantages; Q2's competitive fee pressure made that gap more commercially relevant as buyers optimize for media+data bundles TTD cannot subsidize.
Privacy regulations (GDPR, CCPA) created headwinds for cookie-based targeting that TTD navigated well with UID2. However, TTD does not benefit from meaningful regulatory lock-in in the traditional sense — no government mandates, certifications, or multi-year procurement cycles.
Buyer↔publisher liquidity effects still exist, and retention remains >95%, but Q2 2026's +3% growth versus mid-teens-to-50%+ growth at Amazon Ads, Meta, Google, and AppLovin is evidence the flywheel is losing relative share. Amazon's fee subsidies and Google's Buyer Direct are diverting incremental open-internet budgets that previously reinforced TTD's two-sided network.
TTD processes billions of ad auction transactions daily, sitting in the critical path of open-internet programmatic advertising. Switching means rebuilding trading infrastructure, campaign history, and attribution baselines — non-trivial, and consistent with >95% customer retention even through the Q2 miss.
TTD's reporting and attribution data does not function as a legal or financial system of record. Campaign performance data can be exported and recreated on alternative platforms. This is a notable moat weakness relative to FICO, Axon, or MSCI.
Growth Analysis
Growth Drivers
Key Risk
If Amazon DSP's near-zero take-rate / spending matches and Google's Buyer Direct fee cap force TTD into sustained price cuts or further share loss — such that Q4 2026 and H1 2027 stay at or below the Q3 −12% run-rate rather than re-accelerating — FY2027 revenue stays below FY2025's $2.896B and the multiple remains stuck near 2× P/S
Score Derivation
60.0 base + 1.3 trajectory − 4 margin − 10 risk = 47
Base 60 on a 4% midpoint — the 3–5yr blend after charging Q2's +3% print, the Q3 ≥$650M (−12%) guide, and H1's +7% into the base (prior 10–15% / FY2026 $3.28B path is dead) and decaying toward a high-single-digit open-internet recovery only if execution and take-rate stabilize. +1.3 trajectory (CTV and mid-market accelerating; large CPG/auto decelerating) − 4 margin compression (34% → Q3 ~25% guided) − 10 high key risk (Amazon/Google take-rate and share-loss spiral into 2027) ≈ 47.
Research Covering This Name
Price Scenarios (12–24 Months)
Valuation Multiples
| Trailing P/E (GAAP) | ~16× |
| Forward P/E (NTM) | ~11–14× |
| PEG Ratio | n/m |
| Price / Sales (NTM) | ~2.2× |
| EV / Adj. EBITDA | ~8–10× |
At ~2.2× sales and a mid-teens trailing P/E, TTD is optically cheap versus its 78%-ish gross-margin history and still-positive FCF / buyback capacity (~$1.49B cash and short-term investments; $269M remaining repurchase authorization). The multiple is no longer a free option: peers growing mid-teens to 50%+ command far richer prints, and TTD's Q3 guide embeds −12% revenue and ~25% adj. EBITDA margin. Valuation only becomes the bull case if the company proves the trough is Q3 rather than a new run-rate.
Approximate figures as of August 11, 2026.
Where We Are vs Targets
Loading live price…
Take-rate and share loss deepen: Q4 and H1 2027 stay at or below the Q3 −12% run-rate, FY2027 revenue falls toward ~$2.5B, and the multiple compresses to ~1.5× P/S.
- Amazon DSP spending matches and Google Buyer Direct force further TTD fee cuts or advertiser consolidation — large CPG/auto budgets keep migrating off the open-internet stack
- Q3 prints at or below the $650M floor and Q4 fails to re-accelerate; FY2026 lands near $2.7–2.8B and FY2027 declines further
- Adj. EBITDA margins stay in the mid-20s as platform and go-to-market investment continues without revenue leverage
- Multiple settles near 1.5× on a $2.5B revenue base: $2.5B × 1.5× ≈ $3.75B / ~470M shares ≈ $8
Q3 clears the $650M floor, H2 stabilizes, and FY2027 returns to mid-single-digit growth with margins recovering toward the low-30s — re-rating toward ~2.5–3× P/S on a ~$3.0B revenue base.
- CPG/auto softness proves cyclical rather than structural; CTV, audio, and mid-market (~50% growth outside top 500) carry the mix back to low-to-mid single-digit company growth by mid-2027
- Publicis recommitment (June 2026) plus Dentsu retail-data and commerce-media integrations (Booking, Marriott, Uber, etc.) stabilize agency and brand demand
- Adj. EBITDA margins recover from the Q3 ~25% trough toward low-30s as focused investment priorities (per new CFO Nate Olmstead) improve operating leverage
- Multiple re-rates to ~2.5–3× on ~$3.0B NTM revenue: $3.0B × 2.7× ≈ $8.1B / ~470M ≈ $17 — rounded to $16 as the 12–24 month central case near street's post-print cluster
Execution reset works: CTV and international keep compounding, take-rate holds near 20%, and growth re-accelerates into the low teens by 2027 — supporting a re-rating toward ~4× P/S.
- Netflix, Samsung, and broader CTV supply plus EMEA/APAC CTV +50% trajectories push video/CTV deeper as the growth engine and lift company growth back above 10% by FY2027
- Kokai and agentic integrations (Databricks, Adobe, OpenAI-era tooling) deliver measurable ROAS lift that wins back large-brand budgets from Amazon DSP despite fee competition
- Adj. EBITDA margins recover toward the high-30s / prior 40% framework as revenue re-accelerates and cost focus sticks
- Multiple expands toward ~4× on a ~$3.3B NTM revenue base: $3.3B × 4× ≈ $13.2B / ~470M ≈ $28