The Trade Desk is betting that AI will be a growth engine — not an existential threat — but the company's fate now hinges on an unmistakable tension: can Kokai and a push into CTV and retail media offset the structural supply loss that AI‑powered search and summarized discovery are already imposing on the open web?
The Trade Desk (TTD) built its reputation as the independent demand‑side platform (DSP) for advertisers who want transparency, granular control, and an alternative to the vertically integrated ad stacks run by major platform owners. That positioning became strategic doctrine as advertising budgets shifted from legacy display and search into programmatic formats and connected TV (CTV). Over the last two years the company has responded to a new, faster moving threat: AI is changing how people discover information and where attention — and therefore advertising budgets — concentrate.
In June 2023 The Trade Desk unveiled Kokai, its most ambitious AI initiative, which it presents as a distributed deep‑learning layer running across the programmatic buying stack. Kokai is marketed as a copilot for buyers: advertisers set goals and guardrails, and Kokai automates impression scoring, bid optimization, budget pacing and dynamic audience creation at scale. Management claims rapid adoption — roughly two‑thirds of clients had migrated to Kokai by early 2025, with full migration targeted by year‑end — and points to case studies that show measurable performance benefits for adopters. Those claims are central to The Trade Desk’s thesis that AI can deepen client stickiness rather than hollow out its business.At the same time, major platform owners have embedded generative and conversational AI features into search and discovery. When AI overviews give users direct answers, publisher referral traffic and link clicks can fall sharply — which reduces the pool of ad‑supported impressions on the open web and creates the prospect of attention consolidating inside a few AI ecosystems that can serve ads directly in their interfaces. That trend is the precise structural risk that could squeeze independent DSPs.
From a practical standpoint, Kokai and the company’s curated supply play are meaningful competitive assets. But their value will be determined not just by technology, but by advertiser behavior, CPM dynamics, and how quickly integrated AI discovery experiences are monetized. These are measurable, high‑signal variables over the next 12–24 months: the balance between Kokai adoption and AI search monetization will determine whether AI is an accelerator or a headwind for The Trade Desk.
The core question is deceptively simple: will advertisers prefer an independent, transparent copilot across a diversified set of channels, or will they surrender discovery and monetization to a smaller set of platform owners who control the AI experiences users increasingly use? The Trade Desk has placed an intelligent, coherent bet with Kokai and channel diversification. Whether that bet pays off depends on how fast AI discovery is monetized, how publishers adapt, and whether advertisers can empirically verify that Kokai‑driven outcomes match the company’s claims. For now, the company is better positioned than many peers — but the balance between structural disruption and product differentiation remains finely poised.
Source: The Globe and Mail AI at The Trade Desk: Risk or Opportunity?
The Trade Desk (TTD) built its reputation as the independent demand‑side platform (DSP) for advertisers who want transparency, granular control, and an alternative to the vertically integrated ad stacks run by major platform owners. That positioning became strategic doctrine as advertising budgets shifted from legacy display and search into programmatic formats and connected TV (CTV). Over the last two years the company has responded to a new, faster moving threat: AI is changing how people discover information and where attention — and therefore advertising budgets — concentrate.In June 2023 The Trade Desk unveiled Kokai, its most ambitious AI initiative, which it presents as a distributed deep‑learning layer running across the programmatic buying stack. Kokai is marketed as a copilot for buyers: advertisers set goals and guardrails, and Kokai automates impression scoring, bid optimization, budget pacing and dynamic audience creation at scale. Management claims rapid adoption — roughly two‑thirds of clients had migrated to Kokai by early 2025, with full migration targeted by year‑end — and points to case studies that show measurable performance benefits for adopters. Those claims are central to The Trade Desk’s thesis that AI can deepen client stickiness rather than hollow out its business.At the same time, major platform owners have embedded generative and conversational AI features into search and discovery. When AI overviews give users direct answers, publisher referral traffic and link clicks can fall sharply — which reduces the pool of ad‑supported impressions on the open web and creates the prospect of attention consolidating inside a few AI ecosystems that can serve ads directly in their interfaces. That trend is the precise structural risk that could squeeze independent DSPs.
How Kokai works: technology, claims and limitations
Distributed deep learning across the buying stack
Kokai is described as a distributed deep‑learning architecture that evaluates vast numbers of signals in milliseconds to decide whether to bid on an impression and at what price. The Trade Desk’s materials state Kokai processes the equivalent of more than 13 million ad impressions per second, applying thousands of signals for impression scoring, predictive clearing, dynamic audience generation, and budget allocation. These capabilities underpin features such as Koa Audiences, the TV Quality Index, and the Retail Sales Index, which are intended to bring transparent measurement to opaque channels like CTV and retail media. The company frames Kokai as automation for micro‑decisions while leaving strategic choices to human teams.What Kokai actually delivers today (and how to read company claims)
The Trade Desk’s engineering feat is nontrivial: real‑time scoring at scale requires low‑latency inference, robust feature engineering, and careful orchestration of model updates and privacy boundaries. Company case studies and investor presentations show early performance uplifts and faster spend growth from Kokai adopters, but independent verification of specific percentage improvements is limited because most metrics are presented as anonymized or vendor‑sourced case studies. Investors and advertisers should treat the 13‑million‑per‑second figure and performance percentages as company‑reported claims that are plausible given modern infrastructure — but that merit independent A/B testing and holdout experiments before wholesale budget migration.Generative creative plug‑ins: from placements to language
Kokai is also being extended into the creative workflow. The Trade Desk has integrated third‑party generative creative tools — notably partnerships named in its ecosystem such as Rembrand, Spaceback and Bunny Studio — to enable virtual product placements, dynamic creative optimization, and localized ad generation across formats and languages. The goal is to capture more of the ad value chain: not just where ads are bought, but how ads are made and personalized at scale. If these creative integrations scale, Kokai moves from a pure optimization engine toward a consolidated creative + media platform. That diversification could increase per‑client revenue and raise switching costs, but execution is complex and dependent on third‑party partners and content rights.The risk: AI search, “zero‑click” discovery, and inventory compression
What is shrinking and why it matters
AI‑augmented search interfaces — voice assistants, generative "overviews," and chatbot responses — are changing user behavior. Multiple industry analyses show that when a generative overview appears, click‑through rates to publisher pages fall significantly: sessions with AI overviews produce link clicks at roughly 8% of cases versus historical averages nearer to double that. The practical consequence for programmatic buyers is fewer referral‑driven pageviews and therefore fewer ad impressions in open‑web inventory that depends on search discovery. For publishers and advertisers who rely on scale from organic search, this is material.Beyond fewer clicks, platforms are beginning to monetize these AI experiences natively. Google has moved to include ads in AI Overviews and its “AI Mode”; Microsoft has been testing ad formats inside Copilot. When advertiser budgets can be placed directly into those AI discovery surfaces with low friction, demand becomes concentrated in the hands of platform owners who control both discovery and monetization. That reduces the bargaining leverage and addressable inventory for independent DSPs over time.Why not all impressions disappear — but why scale still matters
It is important to be precise: a drop in referral clicks does not translate one‑to‑one into a total loss of ad impressions. Many publishers still have direct audiences through apps, subscriptions, social distribution and direct navigation. However, the mix and quality of the remaining inventory changes. For campaigns that target audiences discovered primarily via search referrals or that rely on scale across long‑tail publishers, reach and efficiency can be hit disproportionately. That shift can push advertisers either to bid more aggressively for the scarce, high‑quality open impressions (raising CPMs) or to reallocate budgets to the platform owners’ ad surfaces — both outcomes erode the independent DSP value proposition.The Trade Desk’s strategic response: hedging with channels and curated supply
Curated supply and transparency plays
The Trade Desk is leaning into curated, premium inventory and transparency products designed to protect clients from supply volatility. Initiatives such as OpenPath and the Sellers & Publishers 500+ list are explicit attempts to direct buyers toward brand‑safe, measurable placements that advertisers will continue to value even if general search discovery shrinks. These tactics aim to preserve margins and maintain reachable audiences where programmatic buying still makes sense. The emphasis on measurement transparency — via the TV Quality Index and Retail Sales Index — is central to the pitch: if advertisers can trust the buying signals and outcomes, they may prefer an independent DSP to an opaque platform that hides bid dynamics.Prioritizing CTV and retail media
Two channels stand out as structural hedges against AI search disruption:- Connected TV (CTV): CTV audiences are discovered through viewing behavior and platform ecosystems that are typically outside traditional search referral flows. Programmatic CTV is growing and The Trade Desk already has a strong footprint in programmatic video. That channel offers scale and brand reach that are less likely to be cannibalized by search summaries.
- Retail media: Retail networks (first‑party commerce data) present ad opportunities tied to purchase intent within retailer environments. These placements are driven by point‑of‑sale behavior and in‑store discovery rather than search referrals, making them relatively insulated from search overview effects. The Trade Desk’s Retail Sales Index is explicitly designed to measure outcomes in this channel.
Extending into creative: capturing more of the funnel
By integrating generative creative partners into Kokai, The Trade Desk is attempting to capture a larger share of advertiser workflows. Virtual product placement, localized creative generation, and automated video edits increase the company's total addressable market (TAM) beyond media buying to include creative production and personalization. This can raise client switching costs and increase average revenue per user if advertisers embrace an integrated stack. Execution risk is nontrivial: creative quality, rights management, and brand safety must be managed tightly.For advertisers: pragmatic steps and control measures
Advertisers should treat Kokai and similar automation platforms as powerful tools that require disciplined measurement and governance.- Run controlled holdout tests and A/B experiments before migrating full budgets into automation. Track lift relative to manual or legacy strategies.
- Prioritize multi‑channel measurement frameworks that include incrementality testing, not just last‑click attribution. Use holdouts to determine whether Kokai’s reported uplifts translate to genuine acquisition or only reallocation within existing audiences.
- Preserve a creative + media workflow that can flex between open‑web, CTV, retail media, and platform‑native formats. Generative creative can accelerate scale, but human review and brand governance remain essential.
- Monitor inventory quality and CPM changes closely; if CPMs for curated open‑web inventory spike without commensurate transparency or outcomes, reconsider allocation.
For investors: scenarios, indicators and what to watch
AI is a double‑edged sword for The Trade Desk. The platform’s success depends on two outcomes happening simultaneously: Kokai delivering sustained performance improvements and The Trade Desk successfully shifting the revenue mix toward channels that are less exposed to AI search disruption.Key indicators to monitor quarterly
- Kokai adoption and spend share — the percentage of client spend running through Kokai and whether Kokai adopters increase retention and expand budgets. Rapid, persistent adoption combined with demonstrable revenue uplift would be a strong positive.
- Open‑web supply and CPM trends — are premium open‑web CPMs increasing because supply is compressed, or are advertisers shifting budgets away from open web toward platform‑owned surfaces? Significant structural price inflation without proportional performance gains would signal stress.
- Monetization of AI discovery surfaces — the pace at which Google, Microsoft and other AI providers insert ads (and the formats they use) will determine how quickly advertiser budgets migrate to those ecosystems. Evidence of rapid monetization with healthy advertiser ROI favors the platforms.
- Client churn and agency feedback on Kokai — anecdotal and industry reports of agency resistance to automation should be treated seriously; operational frictions can slow migration and create churn risk.
- Retail media and CTV growth rates — are these channels capturing sufficient incremental spend to offset open‑web declines? Rising share here would validate The Trade Desk’s diversification strategy.
Potential scenarios
- Bull case (AI as accelerant): Kokai becomes the industry standard copilot for open‑internet buying, delivering measurable ROAS gains that increase client stickiness while Trade Desk captures disproportionate share in CTV and retail media. The company’s combined creative + buying stack raises ARPU and expands margins.
- Balanced case (AI as partial threat, product offsets): Open‑web inventory compresses, but premium curated supply, CTV and retail media offset enough budget to keep revenue growth positive though perhaps at lower margins. Kokai adoption stabilizes client churn and supports steady expansion.
- Bear case (AI consolidates demand in platforms): Monetization inside AI discovery surfaces and competitive advantages held by large platform owners draw significant advertising budgets away from independent DSPs. Open‑web inventory shrinks materially and CPM dynamics or client reallocation reduces Trade Desk’s TAM, compressing growth prospects.
Strengths, limits and regulatory tail risks
Notable strengths
- Product engineering and early adoption: Kokai is a meaningful technical investment that has attracted a large share of client spend quickly, which is an important validation of the platform approach.
- Channel diversification: Heavy emphasis on CTV and retail media reduces single‑channel dependency and matches where advertisers are increasing spend.
- Transparency narrative: Promises of better measurement for opaque channels — TV Quality Index, Retail Sales Index — resonate with advertisers demanding auditable outcomes.
Important limits and risks
- Inventory exposure: The structural shift in discovery patterns is out of product teams’ direct control; platform owners who embed both discovery and monetization retain a fundamental advantage. If AI summaries permanently reduce open‑web supply or if platforms monetize these surfaces at scale, independent DSPs face an uphill battle.
- Execution complexity: Integrating generative creative, measurement indices, and curated supply at scale while maintaining brand safety and legal compliance is operationally demanding. Any missteps could slow adoption.
- Regulatory and legal uncertainty: Antitrust scrutiny, disputes over content licensing, and evolving rules on model training data could reshape economics for AI search products and the broader ad market in unpredictable ways. These are non‑linear risks that could materially affect outcomes.
Verdict: risk‑adjusted view for readers and market participants
The Trade Desk’s response to AI is both coherent and defensible: build a better product, make AI a core differentiator, and shift the business mix toward channels that are less dependent on open‑web referral traffic. That strategy buys the company time — and may even create new growth levers — but it does not eliminate the structural threat posed by platform owners who control discovery and who can insert ads inside AI interfaces.From a practical standpoint, Kokai and the company’s curated supply play are meaningful competitive assets. But their value will be determined not just by technology, but by advertiser behavior, CPM dynamics, and how quickly integrated AI discovery experiences are monetized. These are measurable, high‑signal variables over the next 12–24 months: the balance between Kokai adoption and AI search monetization will determine whether AI is an accelerator or a headwind for The Trade Desk.
Actionable checklist: what advertisers and investors should track now
- Verify Kokai performance with controlled holdouts before migrating the majority of spend.
- Monitor premium open‑web CPMs and reach metrics monthly to detect supply compression early.
- Watch the speed and format of ads inside AI discovery products from Google and Microsoft — and whether those placements deliver ROI comparable to open‑web buys.
- Evaluate The Trade Desk’s revenue mix: rising share from CTV and retail media offsets open‑web risk meaningfully.
- Track regulatory developments that could change licensing economics or force new disclosure requirements for AI model training and monetization.
The core question is deceptively simple: will advertisers prefer an independent, transparent copilot across a diversified set of channels, or will they surrender discovery and monetization to a smaller set of platform owners who control the AI experiences users increasingly use? The Trade Desk has placed an intelligent, coherent bet with Kokai and channel diversification. Whether that bet pays off depends on how fast AI discovery is monetized, how publishers adapt, and whether advertisers can empirically verify that Kokai‑driven outcomes match the company’s claims. For now, the company is better positioned than many peers — but the balance between structural disruption and product differentiation remains finely poised.
Source: The Globe and Mail AI at The Trade Desk: Risk or Opportunity?