The Amazon Overlap
The Amazon Overlap
The competitive fear most often attached to The Trade Desk is Amazon, whose advertising business reached $68.6 billion in 2025, up 22%, roughly 24 times TTD's revenue. But the two businesses meet in a narrow place. By Amazon's own disclosure and management's decomposition, about 90%-plus of that revenue is sponsored listings and 97%-99% is owned-and-operated inventory; only a small fraction is open-internet buying of the kind TTD does. The direct overlap is small today. The risk is directional — Amazon turning its demand-side platform outward — and one independent peer already calls it a threat.
The scale gap, and what it measures
TTD names two rivals above all others in its filings: "demand-side platform providers, some of which are smaller, privately held companies and others are divisions of large, well-established companies such as Google and Amazon" [1]. Amazon is the one that has dominated the narrative, because its advertising line has compounded fast: $37.7 billion in 2022 to $68.6 billion in 2025 [2].
Source: Amazon FY2025 Form 10-K, disaggregated net sales [3].
The headline comparison flatters the fear, because it sets two different things side by side. Amazon's $68.6 billion is gross advertising revenue earned selling ads on its own properties — mostly the sponsored product placements shoppers see on retail search results. TTD's $2.9 billion is not gross ad sales at all; it is the platform fee TTD charges — on the order of a fifth — on roughly $13 billion of client spend that flows across the open internet [4]. The only part of Amazon's business that competes for the same dollars TTD wins is the slice of its DSP that buys inventory Amazon does not own.
Sizing the overlap
Management has put numbers on that slice. On the third-quarter 2025 call, CEO Jeff Green estimated that of Amazon's roughly $70 billion, "approximately 90% of this revenue comes from sponsored listings, likely more than 95%," which he frames as competition with Google Search rather than with TTD; Prime Video contributes "a couple of billion dollars at most"; and Amazon's DSP "is primarily focused on buying ads for Prime Video, with minimal investment geared towards the open Internet." His conclusion: "close to 97% to 99% of Amazon's advertising efforts center around monetizing their owned and operated inventory, with only a small portion dedicated to the open Internet" [5].
Source: management estimate (Q3 FY2025 call) applied to reported 2025 advertising revenue of $68.6B; splits approximate [6] [7].
Within that DSP / other bucket, the open-internet decisioning that actually competes with TTD is, on management's telling, "a small fraction." Two things support the estimate independently of TTD's incentive to make it. Amazon does not break out its DSP, but its disclosed advertising has always been dominated by sponsored products tied to retail search, and its structural motive points inward: Green notes that "being overly focused on the open internet and anything other than Prime Video could jeopardize their core business," and that Amazon "often compete[s] with advertisers," which discourages the neutral, publisher-friendly posture an open-internet DSP requires [8]. His forward view is that Amazon builds "tools for buying owned and operated inventory," resembling Facebook's model, rather than "a DSP as we understand it" [9].
The independent-DSP read
Green's dismissal comes from an interested party. A more useful check is the only other pure-play independent DSP of any scale, Viant Technology — a company that competes with TTD but has no reason to talk Amazon up. Viant's revenue, $344 million in 2025 against TTD's $2.9 billion, makes clear how thin the independent field is: TTD is roughly eight times its nearest same-model rival [10].
Source: TTD FY2025 Form 10-K [11]; Viant FY2025 Form 10-K [12].
On the present-scale question, Viant corroborates TTD. Asked in late 2025 about Amazon's widely reported 0% DSP-fee promotion, Viant's CEO reported no rise in competitive intensity: "Most of their revenue is sponsored listings. The DSP, I would guess, is a very small portion. And we do not see them… in the competitive bake-off processes at the finish line." He added, drily, that "the Amazon DSP marketing team deserves a trophy for how much coverage they have been able to get" [13].
The same executive, one quarter later, is the strongest voice against the comfortable read — and it is worth more precisely because it works against his own book. Pressed on Amazon's new deals to serve ads into Netflix and Roku, Viant's CEO reversed the tone: "I do not want to say they are not a threat. They are a threat. They are subsidizing their products. They are doing the bundling strategy that Google executed… I do not want to discount Amazon as a competitor in the space like some others have" [14]. The distinction he draws is where the analysis lands: Amazon's data advantage is real but bounded — "they know a lot about customers of Amazon… very little in all the other retailers like Walmart, CVS" — so Amazon's DSP fits an advertiser selling through Amazon, and fits far worse a brand that does not [15].
Pricing, and the shape of the real risk
Amazon's current DSP footprint is small; pricing is the sharper risk. TTD earns roughly a fifth of the spend it directs. Amazon has been reported offering its DSP at a 0% fee — a subsidy a retail-and-cloud giant can carry indefinitely and a standalone platform cannot [16]. TTD's own filings name the pattern without naming Amazon: some competitors "differentiate themselves… primarily on the basis of artificially low prices, which are enabled by inherent conflicts of interest and a lack of objectivity," and walled-garden inventory owners "may exclusively sell their own inventory directly to advertisers, which prevents us from competing with them entirely for such inventory" [17].
Watch item: Amazon's competitive pressure on TTD would show up first as pricing — a subsidized DSP fee against TTD's ~21% take — and only later as share, if Amazon opens Prime Video and third-party inventory (Netflix, Roku) to outside demand. Neither has dented TTD's take rate or retention through 2025.
The counterweight is the same neutrality that anchors The Independence Moat. Green's argument for why publishers keep partnering with TTD rather than Amazon is that content owners "such as Disney, Peacock, or Paramount" will not route demand through "a company that is also vying for ad placements against them" [18]. Viant makes the sell-side version of the same warning: if Amazon "runs the same playbook as Google," it will grade its own Prime Video inventory as outperforming rival apps it also sells — the conflict that a neutral buyer does not carry [19].
The read
On the evidence, Amazon is a real competitor whose direct overlap with TTD's open-internet buying is small today — a low-single-digit slice of a $68.6 billion business that is otherwise retail-search advertising and Prime Video inventory. TTD remains the independent DSP of scale, roughly eight times its nearest same-model peer, and its CTV growth has continued to outpace Amazon's advertising growth. The strongest fact against that read is that the overlap is a moving target: Amazon has begun serving ads into Netflix and Roku, has the balance sheet to subsidize a 0% DSP fee, and is judged "a threat" by the one independent peer with the least incentive to say so.
What would change the read is observable in the same filings and calls the report has leaned on throughout (What to Watch): Amazon's DSP scaling its outward, third-party CTV deals beyond a rounding error; a subsidized-pricing regime that pulls on TTD's take rate rather than leaving it near 21%; or TTD's CTV growth falling behind Amazon's ad growth. Absent those, the Amazon overlap is a risk to underwrite, not a share loss to mark.