Hightouch yesterday announced a partnership with The Trade Desk that connects the demand-side platform's raw impression data directly to brands' own customer records stored in their data warehouses, removing the multi-day lag that has long made log-level measurement a slow and fragmented exercise.

The product is called Exposure Log Matching. It launched on June 1, 2026, and targets the structural gap between bidstream identifiers - UID2s, mobile advertising IDs, and cookies - and the customer and household records that live inside a brand's own data infrastructure. The announcement marks a significant step in how programmatic measurement infrastructure can be built without handing control of that infrastructure to a third-party identity vendor.

What the product does

The Trade Desk operates a data product called the Raw Event Data Stream, or REDS. According to Hightouch, REDS delivers detailed context on every consented impression, click, video view, and conversion to the customer's data warehouse. Each log entry includes metadata on the campaign, publisher, and even environmental signals such as weather conditions at the time an ad was served.

Those logs are rich. But to use them for real measurement, a team has to connect the bidstream identifiers back to actual customer records. A UID2 token or a mobile advertising ID is not the same as a customer ID, a household ID, or any internal identifier that a brand uses across its own systems. Without that translation, the logs describe ad events but cannot explain which customers were exposed, how often, or what happened next.

Exposure Log Matching handles that translation inside the warehouse. The mechanism is a feature Hightouch calls Match Booster. According to Hightouch, Match Booster resolves exposure logs directly to the brand's own primary key - whether that is a customer ID, household ID, or another pseudonymous identifier used across the business. The process starts with consented first-party signals and falls back to consented third-party data only where gaps remain. The resolved exposure logs are then delivered into the customer's data warehouse, ready to be joined directly to consented customer records for analysis and activation.

The result is that exposure data arrives keyed to identifiers the brand already owns and uses, rather than to identifiers owned by an external vendor.

The problem with traditional workflows

Before this approach existed, brands that wanted to use log-level exposure data had to route it through third-party identity providers. That traditional workflow involves file transfers, multi-day turnaround times, and a matching process that relies entirely on external data sources. According to Hightouch, the result is unresolved records and inaccurate reporting. The bridge between ad exposure and customer data is controlled by a vendor whose commercial interest lies in owning that bridge.

The fragmentation compounds quickly. Even after days of processing, the match rate using only third-party identity is incomplete. Records that do not resolve are simply dropped from analysis. That means measurement is not only slow but structurally biased toward the portion of a brand's audience that happens to be addressable through a particular vendor's graph.

Hightouch positions Exposure Log Matching as a warehouse-native solution that combines first and third-party data to improve match rates without creating identity dependency. The brand retains the primary key. The vendor provides matching capability as a service, not as a proprietary lock.

What analysts and AI agents can do with it

Once exposure logs are resolved back to warehouse-native customer identifiers, the use cases expand considerably. According to Hightouch, analysts can understand performance by customer segment, household, region, SKU, lifetime value tier, and membership status. Media networks can combine onsite and offsite ad exposures to measure omnichannel attribution and surface dashboards to their advertisers.

The AI agent use case is perhaps the most forward-looking. According to Hightouch, AI agents can traverse exposures, customers, and financial outcomes in seconds to uncover hidden patterns, provide detailed campaign recommendations, and act autonomously to adjust bids, budgets, and creative. The critical condition is that the data must be quickly available and connected to the context layer in the data warehouse. Exposure logs that take three to five days to resolve and arrive in a format not natively joinable to customer records are of limited use to any agent operating in near real time.

This is not a theoretical concern. Agentic AI systems attracted $1.1 billion in equity investment during 2024, according to McKinsey data reported by PPC Land, with job postings for agentic roles increasing 985 percent from 2023 to 2024. The operational infrastructure those agents rely on - fast, joined, warehouse-native data - is exactly what Exposure Log Matching is designed to provide.

The identity ownership question

The timing is notable. Identity infrastructure is under active consolidation right now. Publicis Groupe agreed to acquire LiveRamp for $2.5 billion in May 2026, a move that immediately raised questions for holding companies and brands that depend on LiveRamp's matching and clean room capabilities. If identity infrastructure becomes a subsidiary of a competing agency group, brands that relied on that infrastructure for neutral measurement will need to reconsider their options.

The Hightouch announcement lands in that context directly. According to Hightouch, brands and media networks are reevaluating the identity and measurement infrastructure they depend on - they want neutral alternatives, control over their context layer, and measurement that is not locked into someone else's proprietary graph.

The Trade Desk launched Galileo in January 2023, its first-party data matching platform, designed to let advertisers activate CRM, CDP, and clean room data through the platform. Galileo works in concert with UID2 and includes direct integrations with Adobe, Amazon Web Services, Habu, InfoSum, LiveRamp, Salesforce, and Snowflake. Exposure Log Matching represents a distinct layer: rather than helping brands push data into The Trade Desk for activation, it helps brands pull the resulting exposure data back out and connect it to their own records. The two capabilities address opposite ends of the data flow.

The Trade Desk has been expanding its ecosystem infrastructure through OpenTTD, launched on March 4, 2026, a unified portal giving data providers, publishers, and brands API access to programmatic advertising tools. Exposure Log Matching operates within that ecosystem model.

UID2 oversight has also drawn scrutiny. PPC Land reported in May 2026 that broken UID2 tokens from a major CTV publisher went undetected for months, exposing gaps in how bidstream identity signals are validated at the point of supply. That episode highlighted a structural tension: when identity signals embedded in bid requests are incorrect or fabricated, the downstream measurement built on those signals is corrupted from the start. A warehouse-native matching approach that resolves exposure logs after the fact against a brand's own consented first-party records offers one response to that problem - the brand's own data acts as a verification layer rather than taking bidstream signals on trust.

Hightouch's position in the market

Hightouch is a composable customer data platform provider. Founded in 2018 and headquartered in San Francisco, the company raised $172.4 million across multiple funding rounds. In January 2026, Gartner named Hightouch a Leader in the Magic Quadrant for Customer Data Platforms, its first appearance in that ranking.

The company's core architecture is warehouse-native - it activates data directly from cloud data warehouses such as Snowflake and Databricks, without replicating data into a proprietary system. That architecture is what makes Exposure Log Matching technically coherent: the resolved exposure logs do not need to travel through Hightouch's systems after matching; they land in the warehouse the brand already operates.

Hightouch also offers ID Express, a separate product for one-click UID2 activation to The Trade Desk, and a Trade Desk Conversions API integration. Exposure Log Matching is the third leg of that integration story - not just pushing audiences and conversions to The Trade Desk, but retrieving exposure data back and making it usable.

The Gartner recognition followed a period in which the CDP market shifted notably toward composable architecture. Acxiom and ActionIQ had already partnered to offer a composable CDP approach earlier, and LiveRamp deployed AI agents for audience building and cross-media measurement in March 2026. The common thread across these moves is that brands want data infrastructure they control, not infrastructure operated on their behalf by a party with its own interests.

What the data actually looks like

The exposure log output visible in Hightouch's documentation gives a sense of the raw material involved. Each log entry contains a log entry timestamp, an impression ID, an audience ID, a creative ID, an ad format, a frequency count, a deal ID, a site domain, a referrer, categories, a user identifier, and an hour-of-week signal. A sample drawn from the documentation shows records across formats including 300x250 and 360x640, across exchanges including Index Exchange and Smaato, and across site domains ranging from reddit.com to washingtonpost.com and mlb.com.

That granularity is what makes log-level data qualitatively different from aggregate reporting. Frequency by user can be calculated precisely. Publisher-level performance can be isolated. Hour-of-week patterns in exposure can be cross-referenced against purchase or engagement behavior in the customer record. None of those analyses are possible from campaign-level summaries.

The measurement architecture shift

The broader dynamic here reflects a structural shift in how programmatic measurement is being architected. For most of the programmatic era, measurement was a post-campaign exercise built on aggregated reporting surfaces provided by the DSP or MMP. The advertiser received a summary. The underlying log data, if accessible at all, required significant engineering effort to use.

The data warehouse changed that equation. As cloud warehouses became standard infrastructure for most large advertisers, the possibility of routing raw event data directly into that warehouse became practical. The Trade Desk's data marketplace overhaul with AI scoring, announced in September 2025, reflected the same underlying logic: programmatic data has more value when it is accessible and processable, not when it is locked inside a platform's own reporting layer.

Comscore's integration with The Trade Desk for audio measurement, announced in January 2026, addressed a parallel problem: connecting content-level signals from outside the DSP to campaigns running inside it. Each of these integrations pushes in the same direction - toward programmatic data that is joinable across sources rather than siloed inside proprietary reporting surfaces.

Hightouch's Exposure Log Matching partnership extends that logic to the customer record itself. Rather than stopping at campaign-level or segment-level attribution, the product resolves all the way to the individual customer or household ID that a brand uses across its own systems. The financial outcome, the CRM history, and the ad exposure can sit in the same warehouse table, joinable by a primary key the brand owns.

What this means for media networks

Hightouch notes one specific use case for media networks: combining onsite and offsite ad exposures to measure omnichannel attribution and surface dashboards to their advertisers. That use case is worth unpacking. A retail media network, for example, runs ads both on its own properties and through offsite programmatic channels including The Trade Desk. Measuring whether an exposure on either channel contributed to a sale requires linking the ad event back to a customer identity that can then be matched against purchase records. Without warehouse-native identity resolution, that linkage either does not happen or happens slowly through a third party that charges for the service and retains a copy of the match.

With Exposure Log Matching, the retail media network retains both the match and the resolved record. The exposure log for an offsite impression delivered through The Trade Desk arrives in the warehouse keyed to the same household ID the network uses for its own onsite measurement. That creates a unified attribution surface the network can use for its own reporting and for the dashboards it offers to brand advertisers buying media on the network.

Timeline

  • January 6, 2023 - The Trade Desk launches Galileo, a first-party data matching platform working in concert with UID2, covering CRM, CDP, and clean room provider integrations.
  • September 29, 2025 - The Trade Desk announces Audience Unlimited, an overhaul of its data marketplace with AI-powered segment scoring and simplified pricing at 3.3% and 4.4% rates.
  • January 8, 2026 - Comscore announces content-level audio measurement integration with The Trade Desk, combining programmatic audio targeting and campaign measurement on the DSP.
  • January 26, 2026 - Gartner names Hightouch a Leader in the Magic Quadrant for Customer Data Platforms, its first inclusion in the ranking.
  • March 4, 2026 - The Trade Desk launches OpenTTD, a unified portal giving data providers, publishers, and brands API access to programmatic advertising tools across the open internet.
  • March 3, 2026 - LiveRamp deploys AI agents for audience building and cross-media measurement, extending identity infrastructure into agentic marketing workflows.
  • May 2026 - Publicis Groupe agrees to acquire LiveRamp for $2.5 billion, raising questions for brands that depend on LiveRamp for neutral identity infrastructure.
  • May 2026 - PPC Land reports on UID2 oversight gaps following a CTV publisher's broken tokens going undetected for months, exposing structural weaknesses in bidstream identity validation.
  • June 1, 2026 - Hightouch announces Exposure Log Matching in partnership with The Trade Desk, enabling warehouse-native resolution of REDS exposure logs directly to first-party customer and household IDs.

Summary

Who: Hightouch, a composable customer data platform provider, in partnership with The Trade Desk, the programmatic demand-side platform.

What: Exposure Log Matching - a product that resolves The Trade Desk's Raw Event Data Stream exposure logs directly to a brand's own first-party customer or household IDs inside the data warehouse, using a combination of consented first-party and third-party signals through a feature called Match Booster.

When: Announced on June 1, 2026.

Where: The resolution process and the resulting data operate inside the brand's own cloud data warehouse. No data needs to travel to Hightouch or any other intermediary after the matching is complete.

Why: Brands and media networks have historically needed third-party identity providers to translate bidstream identifiers - UID2s, mobile advertising IDs, cookies - back to their own customer records. That process takes multiple days, produces incomplete matches, and leaves the measurement bridge in the hands of a vendor with proprietary interests. As AI agents require faster, joined, warehouse-native data to operate effectively, and as identity infrastructure faces consolidation through events such as Publicis's acquisition of LiveRamp, brands are seeking measurement approaches that keep control of the context layer with the brand rather than with a third party.