The Video Advertising Bureau today published a guide answering six questions about identity resolution in video advertising, the third installment in a five-part measurement series that names three specific Universal ID products and recommends marketers question vendors on match quality before signing contracts.

The document, titled "What's the Deal with Identity?", arrived on July 29, 2026, distributed to VAB members and what the trade organization calls qualified marketers. It follows two earlier entries covering viewership data collection and engagement, with two further installments on outcomes and what VAB terms "what's next" still to come. According to VAB, identity is the mechanism that links raw viewership data to audience profiles, engagement signals and campaign outcomes across devices and platforms.

What identity means in VAB's framework

The guide defines identity as the ability to recognize an individual viewer's or household's characteristics, viewing behaviors, purchase behaviors and other lifestyle traits. That definition frames identity not as a single technology but as a category spanning several tools: personal identifiers such as names, addresses and email addresses; digital identifiers including IP addresses, device IDs and mobile advertising IDs; and behavioral signals gathered through loyalty programs or app logins.

According to the guide, that information can be used to build more complete customer profiles, construct audience segments of varying size, and find matches between datasets collected at different touchpoints. The report positions identity as glue rather than as an endpoint, connecting exposure data - who saw an ad - to device data - where they saw it - and outcome data - what they did afterward. Without that connective layer, a marketer might know a household streamed a program but have no way to tie that viewing event to a purchase recorded somewhere else.

Marketers currently face more measurement choices than in prior years, and identity solutions built around third-party cookies have been losing ground for years as browsers restrict tracking, a shift that has already reshaped how the industry builds identity graphs; VAB's own guide leans heavily on alternatives designed for a cookie-scarce environment.

Why VAB says identity matters

The guide lists five reasons marketers should care about identity infrastructure: it ties ad exposure data to individuals or households, provides what it calls a singular view of an audience across platforms, functions as a key element in generating insights about audiences, increases a marketer's ability to manage campaigns from activation through measurement, and supports efforts to deliver a seamless customer experience. None of these five points comes with a supporting statistic in the guide itself. They read as functional claims about what identity infrastructure is supposed to accomplish, not as measured outcomes from a specific deployment - a distinction worth holding onto, since much of the wider industry conversation around identity in 2026 has centered on whether those promises hold up once implementations move from slide decks to live campaigns.

Identity graphs and identity resolution, defined separately

The report draws a specific distinction between two terms often used interchangeably in casual industry conversation. An identity graph, according to the guide, is a database of common identifiers matched with privacy-protected customer profiles. Identity resolution, by contrast, is described as the process of matching common identifiers across devices and interactions into that single profile. One is the structure; the other is the mechanism that builds and maintains it. The guide ties these two components to four marketer functions: resolving identity by assembling multiple signals into one profile, enriching audience data by layering in additional consumer information, activating relevant ads on the right platforms, and measuring where and how many times a campaign reached that audience.

Marketers often hold audience information across multiple internal databases that were never designed to talk to each other, according to the guide - a CRM system here, a loyalty program there, an ad server logging exposure somewhere else. Identity graphs and resolution processes are the mechanism VAB describes for stitching those fragments into something usable, though the report does not specify what accuracy rate marketers should expect from that stitching process, an omission that becomes relevant given what independent measurement research has found elsewhere in the market this year.

Privacy gets three specific recommendations

Rather than treating privacy as an afterthought, the guide places it as the third of six questions and offers three concrete actions. It recommends complying with consumer protection laws and regulations at both the national and state level, noting that staying current with a patchwork of jurisdiction-specific rules lets a marketer protect a customer's identity and personal information while continuing to serve relevant advertising. For a reference point on that patchwork, the guide points directly to iapp.org, the website maintained by the International Association of Privacy Professionals.

Second, the report recommends establishing first-party relationships directly with customers, arguing that gathering consent while collecting information allows a marketer to identify consumers accurately throughout a campaign's full lifecycle. Third, it recommends what the industry calls a "privacy by design" approach - embedding privacy at every step of the marketing process rather than retrofitting compliance after launch. That framing treats privacy as a design principle rather than a constraint bolted on afterward, though its translation into practice is a separate question regulators have not been shy about raising this year.

Data clean rooms: benefits, mechanics, and a regulatory caveat the guide omits

The fourth question addresses data clean rooms, which the guide defines as a secure software environment enabling data collaboration between two or more parties, allowing sensitive customer data to be shared or matched for joint analysis in a manner it describes as privacy compliant. In a simplified example, an advertiser contributes first-party customer data, a publisher contributes first-party ad exposure data, and the two are matched inside the clean room to produce a combined, anonymized dataset feeding measurement, attribution, or broader audience analytics. According to the guide, clean rooms offer four benefits: a secure and privacy-compliant way to share and match data, marketer control over data and personally identifiable information, support for advanced audience-based targeting, and greater attribution capability.

What the guide does not mention is that the premise underlying clean rooms - that they are inherently privacy-preserving by design - has drawn direct regulatory pushback. The Federal Trade Commission published guidance in November 2024 warning that data clean rooms "are not rooms, do not clean data, and have complicated implications for user privacy," and that most clean room services are not privacy-preserving by default despite the name. The FTC's technical staff noted that data disclosure through clean room systems can carry privacy risks comparable to older methods such as tracking pixels, and that the protections a clean room actually provides depend on the specific governance rules the participating parties agree to, not on the underlying technology alone.

That regulatory caution has not slowed clean room adoption in practice. New York Times Advertising ran its first-ever clean room collaboration in the fourth quarter of 2025, using Decentriq's infrastructure to match first-party data with a global wealth management client, and reported a 61 percent increase in international click-through rate along with a 20 percent uplift in a metric the publisher calls Quality Visit Rate - all without third-party cookies. That gap between the FTC's warning and the results publishers now report illustrates why the guide's benefit list, though accurate as a description of intended function, omits a caveat already shaping how at least one regulator frames the category.

Three questions for vendors, framed around match quality

The fifth section moves from concepts to a practical checklist, framed as three questions marketers should put directly to their advertising partners. The first asks how a vendor validates the quality of matches within its dataset, since quality matters at least as much as quantity and determines the accuracy of both targeting and measurement. Suggested follow-ups ask how a vendor calculates a match and what standards of accuracy it applies, defining a match as occurring when an individual has been identified across datasets using one or multiple identifiers.

The second question asks how often a vendor refreshes its identity solution, since people move, change phone numbers, and open new email addresses, and a stale dataset degrades in accuracy as those changes accumulate uncorrected. The third asks what security measures a vendor has in place, given that many identity solutions integrate across multiple data providers and that the security of those partnerships signals how well a vendor protects data generally.

The emphasis on match-rate quality is not incidental framing. It lands directly on a debate that has intensified across the identity infrastructure market throughout 2026, as independent measurement firms have begun publishing figures that quantify exactly how often supposedly deterministic identity matches turn out to be less reliable than advertisers assumed.

What independent match-rate research has found

VAB's guide recommends that marketers ask vendors about match accuracy, but it does not supply a benchmark for what an acceptable rate looks like. Other research published this year offers texture. Identity resolution vendor Adstra and independent agency InterMedia Advertising found in research first reported July 15, 2026, that just 23 percent of residential IP addresses reached their intended geographic target, while device-level connected television identifiers proved 71 percent consistent month over month, a 24-point gap against IP-based targeting. A companion Truthset report broke the problem down by linkage type: email-to-postal matching reached 51 percent accuracy, the strongest pairing measured, while IP-to-email came in at 16 percent and IP-to-postal at 13 percent. Demographic segments fared unevenly too, with gender proving 61 percent accurate against age segments that collapsed to 13 percent.

Separately, an Adstra study found that IP-based connected television targeting fails roughly three times in four, a failure rate tied directly to identity resolution quality rather than media-buying decisions. Those figures give concrete shape to the abstract match-quality question VAB's guide raises but leaves unquantified, and help explain why the third recommended vendor question carries more weight in 2026 than in prior years, when fewer independent benchmarks existed to check a vendor's own claims against.

Universal IDs: three named products and a maturing category

The sixth and final question addresses Universal IDs, which the guide defines as unique identifiers allowing publishers and advertisers to recognize and authenticate users across publishers, platforms and devices, producing what it calls a holistic view of audiences. According to the guide, Universal IDs have existed in the marketplace for several years, but their adoption and interoperability are increasing in ways that will play a larger role in audience targeting and measurement going forward. The report names three specific products: UID 2.0, associated with The Trade Desk; OpenID; and Ramp ID, associated with LiveRamp. It lists four functions Universal IDs can serve: eliminating dependency on third-party cookies, enabling seamless partnership between advertisers and both publishers and data providers to match first-party data on personally identifiable information securely, achieving greater precision in targeting, and increasing the accuracy of cross-channel measurement.

UID 2.0 in particular has drawn substantial coverage this year as its role in the post-cookie stack has expanded. The Trade Desk consolidated UID2 documentation into a single developer portal called OpenTTD, launched March 4, 2026,bringing together UID2, its European counterpart EUID, and a consent mechanism called OpenPass, reflecting an infrastructure category moving from point integrations toward a formal ecosystem.

Adoption has not been without friction. A major national connected television publisher spent three months sending improperly encrypted UID2 tokens through bid requests earlier this year, an error The Trade Desk - the protocol's sole administrator - never detected on its own, according to reporting on the case. The publisher eventually corrected the error internally, and found that adopting UID2 had made no measurable difference to its advertising revenue - an episode sitting uneasily alongside VAB's framing of Universal IDs as accuracy tools, since it shows an implementation running broken for months without anyone, including the protocol's own administrator, noticing.

On a more constructive note for the identifier's adoption trajectory, Truthset announced an integration with UID2 on May 7, 2026, giving media buyers the ability to match independently validated audience segments against UID2 identifiers at the individual record level across connected television, mobile and web inventory, without relying on third-party cookies or probabilistic signals such as IP addresses. Gabe Richman, general manager of Global Identity at The Trade Desk, said at the time that "as the industry moves beyond cookies, accuracy and accountability in audience data matter more than ever."

Best practices, condensed to four principles

The guide closes its main content with four best-practice recommendations: integrating identity into marketing strategy throughout planning, activation and measurement rather than treating it as a bolt-on layer; emphasizing a quality-over-quantity approach to ensure accuracy, echoing the match-rate question raised earlier; taking a privacy-first approach to measurement and partnering with vendors who share that orientation; and using emerging technology to keep data secure while maximizing its utilization.

Context within VAB's broader measurement push

The identity guide does not exist in isolation. It is the third installment in a five-part series VAB has been publishing across 2026, following earlier guides on viewership data collection and engagement, with outcomes and a forward-looking installment still to come. VAB expanded its separate Advanced Measurement Solutions Directory to 20 vendors on July 21, 2026, organizing that catalog around the same four categories referenced in the "What's the Deal with" series: viewership data collection, identity, engagement and outcomes. That update arrived at a moment reporting characterized as one in which the television industry is questioning who is allowed to grade its own advertising and which figures can be trusted.

VAB has also been active on addressable television this year, distributing an updated guide on June 9, 2026, reporting that 92 percent of pay television households are now addressable-enabled, and separately building a standalone newsletter around a statistic showing authenticated television viewers generate 42 percent more revenue than unauthenticated viewers, a figure originally sourced from Experian research. Authentication of that kind depends on the same underlying identity infrastructure the "What's the Deal with Identity?" guide describes, linking the trade organization's revenue arguments for addressable and authenticated inventory back to the identity-resolution mechanics it is now explaining in more detail.

The report was created by three named VAB staff members: Jason Wiese, executive vice president of strategic insights and measurement; Benjamin Vandegrift, senior vice president of measurement strategy and innovation; and Rohan Gosalia, associate director of measurement intelligence.

As a trade organization whose membership consists of premium multiscreen television providers and distributors, VAB has a structural interest in identity solutions that support confident measurement of television advertising. The guide's framing - that identity is foundational to proving where and how many times an audience was reached - aligns with arguments the organization has made repeatedly elsewhere in its 2026 schedule, including the addressable television and authenticated-viewer figures cited above. That does not make the guide's content inaccurate, but it is a lens worth applying to a document produced by an interested party rather than an independent standards body.

Timeline

Summary

Who: The Video Advertising Bureau, a trade organization representing premium multiscreen television providers and distributors, published the guide. It was created by three named staff members: Jason Wiese, executive vice president of strategic insights and measurement; Benjamin Vandegrift, senior vice president of measurement strategy and innovation; and Rohan Gosalia, associate director of measurement intelligence.

What: VAB published "What's the Deal with Identity?", a 15-page guide answering six questions about identity resolution in video advertising measurement - covering identifier types, identity graphs and resolution, privacy considerations, data clean rooms, vendor evaluation questions, and Universal IDs, naming UID 2.0, OpenID and Ramp ID as specific examples of the latter.

When: The guide was published and distributed on July 29, 2026, as the third installment in VAB's five-part "What's the Deal with" measurement series, following earlier guides on viewership data collection and engagement.

Where: VAB is headquartered at 220 East 42nd Street in New York, and the guide was distributed to VAB members and qualified marketers, with promotion through an email sent from the organization's Insights team.

Why: The guide addresses a persistent gap in how marketers evaluate identity vendors at a moment when third-party cookie deprecation has pushed the industry toward alternative identifiers, and when independent research published earlier in 2026 has quantified significant accuracy gaps in identity matching - findings that give concrete weight to the vendor-evaluation questions the guide recommends marketers ask before committing budget to an identity solution.