As marketers and publishers look to improve targeting capabilities, they have faced accuracy challenges with traditional IDs and strategies, as well as with ID-less environments. In response, many have turned toward alternative IDs and solutions as they seek to connect fragmented ecosystems.

For this new State of the Industry report, Digiday and IntentIQ polled 91 brands, agencies, publishers and retailers to learn how their audience targeting and identity resolution strategies are evolving. Only 21% of survey respondents reported being “very confident” about their ability to accurately identify and reach their target audiences across digital channels.

The report explores how stakeholders are identifying and reaching audiences, as well as their priorities in terms of data and transparency. While organizations still lean heavily on traditional IDs — including third-party cookies and device identifiers — there is growing adoption of alternative IDs.

“The main challenge is fragmentation — across identifiers, environments and data sources. As cookies and mobile IDs disappear, agencies and brands are losing the ability to connect data across platforms,” said Tamir Shub, vp of business development at IntentIQ. “Publishers, meanwhile, are trying to preserve and activate their first-party data without compromising privacy or scale.

“Without a unified or consistent identity spine, marketers can’t accurately cap frequency, measure reach or attribute outcomes,” he said. “Solving for identity in a privacy-safe, interoperable way is now foundational to maintaining both performance and accountability.”

While most industry stakeholders express some level of confidence in their ability to identify and reach audiences, addressability and targeting remain fraught with platform-specific challenges, data fragmentation and uneven first-party data availability.

As reported by Digiday, though many digital marketers prefer to trust the precision of the first-party data, some are turning to a blended approach — employing both deterministic (authenticated information that can be used to directly refer to an audience or customer) and probabilistic (data used to infer information about an audience) data in their audience targeting — in the hope they can enjoy both reliability and reach.

Nearly all respondents (92%) feel at least “somewhat confident” about their identifying and targeting abilities across digital channels, including 21% who are “very confident.”

However, the picture becomes more nuanced upon further examination into specific channels and platforms across the digital landscape. When asked on which platforms they find it most difficult to maintain audience addressability, respondents reported the most challenges with mobile in-app (41%), followed by CTV (40%) and walled gardens, such as Amazon or Meta (30%).

Regarding specific browsers, Safari was the most challenging (22%), followed by Firefox (12%) and Chrome (8%). Of course, Apple and Mozilla removed third-party cookies from their browsers several years ago, while Google famously made a U-turn on cookies and continues to use the targeting technology. Chrome also remains the dominant browser globally, across both desktop and mobile browsers.

“The biggest challenge now in terms of audience targeting and addressability is the lack of identifiers, aka cookies, in the cookie-restricted environment,” said Jasper Liu, programmatic yield manager at Daily Mail. “On the app level, Apple’s App Tracking Transparency has heavily restricted the user information that can be shared between the buy-side and the sell-side.”

Among respondents, the leading challenges impacting audience discoverability and targeting include data privacy compliance complexity (56%), fragmented platform ecosystems (55%), lack of interoperability between ID solutions (47%), technical integration challenges (47%), difficulty scaling audiences across platforms (45%) and insufficient or inaccurate first-party data (45%).

Taking a closer look at first-party data, among non-publisher respondents, only 9% reported that “most publishers” have enough first-party data to provide scale for ad targeting.

However, there is reason for optimism as 48% find that “primarily larger publishers” have sufficient first-party data, and 35% believe that “publishers are improving” in this area.

“There are conversations about growing authenticated users — not only for us, but across all publishers — to drive incremental revenue through first-party data activation,” Liu said. “The challenge for now is that the login rates are currently very low across major news publishers and their apps. So, for the buy-side to deterministically identify users is very difficult right now.”

Marketers and publishers are using multipronged identity strategies to navigate the complexities of the digital media landscape.

Despite the limitations in access and accuracy, more than three-quarters (76%) of respondents are using first-party data segmentation as an audience discovery strategy. Other popular strategies include the use of publisher-declared audiences (57%), contextual targeting (55%), lookalike modeling (46%) and identity-based solutions (44%).

“What we are currently seeing is more of a consent-based approach, where publishers and brands are being very intentional about what data they’re collecting, what data they’re sharing and, I think most importantly, why they are collecting it,” Liu said.

To acquire more first-party data for scalable ad targeting, he recommended publishers and brands grow authenticated touchpoints. These could include app logins, newsletter signups, user profiles, comment systems or even loyalty programs. Rather than keeping a user out, like a hard paywall, these are opportunities to give a user a reason to identify themselves. “Every logged-in user really helps you grow the relationship with the audience,” Liu said. 

Data enrichment (64%) and measurement and attribution (63%) are the primary reasons respondents are implementing identity solutions, including alternative IDs. Other motivating factors include revenue optimization (44%) and privacy concerns (36%).

“Alternative IDs allow marketers and publishers to re-establish continuity across devices, browsers and channels,” Shub said. “When built on deterministic data, these IDs let brands reliably link exposure and conversion data in a privacy-compliant way — powering enrichment, analytics and attribution at scale.

“When you can recognize and reach the same person across environments, you reduce waste and increase effectiveness,” he said. “Brands using our identity framework typically see meaningful improvements in match rates and attribution accuracy, which translate into double-digit lift in return on ad spend in addition to measurable gains in addressable reach.”

Although 46% of respondents cite the decline of third-party cookies as a reason for implementing an identity solution, marketers and publishers continue to rely on traditional identifiers. While third-party cookies remain a liability in some ways, two-thirds of respondents (67%) use third-party cookies. More than half use IP address and user agent string (53%) and platform or walled garden IDs (53%).

However, those traditional identifiers are also being used alongside alternative identifiers, which continue to gain traction. Half of respondents (50%) are currently investing in or experimenting with first-party IDs, while more than 3 in 10 (32%) use retailer-owned IDs. Notably, 58% use other alternative ID solutions (such as deterministic or cookie-less options) and only one-quarter of respondents (25%) do not use alternative IDs.

“Alternative IDs are permission-based in the GDPR region, so the user has opted into this relationship with the publisher or the brand. That’s how identity is translated — whether it’s a hashed email or a probabilistic-based ID — to the buy-side,” Liu said. “The good news is that there is no raw personal data transferred into the buy-stream. Buyers can still target in a relevant, performant way without directly exposing user-level data. This is happening within the industry right now, and alternative IDs play a role.” 

Google’s decision to continue using third-party cookies has had a varying impact on organizations and their approaches to third-party cookies.

Fifty-six percent of respondents report a limited impact, though they may be reprioritizing or reallocating some resources, while another third (33%) have seen no impact or are proceeding as before. Only 11% are fully pausing their migration to alternative IDs.

Excluding those who are fully pausing migration to alternative IDs, respondents are primarily shifting focus to first-party data readiness (46%) and are prioritizing high-performing IDs and logins (46%) since Google’s pivot. Forty percent continue to test alternative IDs but admit urgency has dropped, while 35% are building flexible, ID-agnostic stacks.

With marketers and publishers increasingly under pressure to deliver results, revenue lift measurement (67%) is the primary way respondents are measuring the effectiveness of their identity solutions, including alternative IDs. Other notable KPIs include A/B testing (50%) and CPM uplift tracking (50%).

“We’ve been testing multiple ID solutions in the market, and what really matters to us is the metrics,” Liu said. “We usually see more uplift when we have a specific identity approach or data to be targeted in the private marketplace. Programmatic direct deals, with a pre-negotiated approach with the buyer, are generally able to see a 20–40% lift when both parties agree on an ID.”

Respondents who have adopted alternative IDs are seeing performance gains. When examining the performance of alternative IDs in particular, the majority of respondents (63%) reported that these have improved ad performance. Only 2% reported that ad performance has decreased with the use of alternative IDs.

Among the respondents who have found that alternative IDs have positively impacted ad performance, 39% found them most effective on AVOD/streaming platforms. This is notable in part because 40% of all respondents reported challenges in maintaining audience addressability on CTV (see chart 2).

“For publishers, alternative IDs drive higher yield when they are accurately identifying and matching all available demand platform IDs tied to the same user, clustering them into a single universal ID and turning them into addressable demand for SSPs, DSPs and, ultimately, reaching the buyer themself,” said IntentIQ’s Shub. “For advertisers, they drive more efficient targeting and more confident optimization.”

Identifiers alone are not the only way marketers and publishers are tackling their addressability and targeting challenges.

When asked what would improve transparency and efficiency in their media buying or selling processes, 64% of respondents cited standardized reporting across platforms and unified frameworks. This underscores the impact of today’s fragmented environments on marketing efforts, as well as the need for a more holistic view of audiences.

“Transparency starts with knowing how an identity framework is constructed and validated,” Shub said. “Our identity graph is continuously tested against deterministic truth sets and signals to ensure it remains accurate day in and day out. Transparency extends to privacy practices as well. We operate within strict privacy frameworks and ensure that data activation happens without exposing personally identifiable information.”

Other capabilities and offerings that respondents believe would improve the buying and selling processes include insight into how IDs translate to audience attributes (60%), better understanding of identity signal quality (55%) and better support for cookie-less and ID-less environments (53%).

As Shub explained, solving for these cookie-less and ID-less environments is beneficial for various stakeholders. “ID-less environments are one of the largest untapped opportunities for marketers,” he said. “When you can accurately recognize and activate these audiences, both buyers and sellers unlock high-value reach that has been untenable until now.”

Regarding the transparency and reporting from partners that would improve targeting and measurement, respondents are most interested in audience definitions and segmentation logic (68%), frequency and reach metrics across platforms (66%) and clarification of data ownership and permissions (62%). More than one-third (36%) believe signal loss reporting would improve targeting and measurement abilities.

“Everything has to be measured, because identity isn’t just something you turn on and just forget about it,” said Liu. “You have to test metrics like auction pressure, regular CPM, etc. And we have to treat this approach like ongoing optimization efforts. So as a publisher — whether you set up a dashboard, automate the data pipelines or set up alerts so when something goes wrong, you can look into that issue in real time — those kinds of combined approaches really help you to stay in the market, stay within the flow and know what’s going to happen.”

While there have been ups and downs, pivots and pauses when it comes to identity’s role in targeting and measuring ads, respondents have seemingly reached a turning point in the broader shift away from third-party cookies.

While nearly three-quarters of respondents (74%) are mainly using traditional IDs this year for targeting, only 35% expect to be doing the same in 2026. Instead, half expect to be targeting with a more even split of traditional IDs and alternative IDs in 2026, more than doubling from 24% this year. The share of respondents expecting to use mainly alternative IDs is also expected to jump dramatically, from 2% in 2025 to 15% in 2026.

With the adoption of alternative IDs on the rise, more publishers, marketers and other stakeholders are turning to identity resolution partners. 

“The best identity partners should be able to demonstrate three things clearly: deterministic grounding, interoperability and proof of performance,” Shub said. “Ask how often the solution is tested, what data signals feed it and whether the identity layer is actively maintained. Static graphs degrade quickly — partners who test and recalibrate daily are the ones you can trust.”

Liu echoed these sentiments, particularly emphasizing the importance of interoperability.

“The question really comes down to, can this ID coexist with other IDs? If possible, then yes, we are increasing our scale, our metrics, but not undermining other parts of our identity frameworks,” he said. “As we invest in IDs and identity solutions, we also look into how they fit into our overall privacy framework, especially in GDPR regions. IntentIQ has the ability to support multiple GDPR regions without extensive work from the publisher side.

“No single, universal ID will win in the long-term; instead a portfolio of IDs and model-based signals will,” Liu continued. “Contextual will also become more AI-driven and somatic, rather than something really manual to identify audience segments.”

The digital advertising ecosystem will continue to fragment and evolve, creating new identity and addressability challenges for brands, agencies, publishers and retailers. 

“By 2026, identity will be fully embedded into the fabric of activation and measurement,” Shub said. “The distinction between ID solutions and media platforms will blur — identity resolution will become a core utility powering attribution, frequency management and optimization across every channel. Identity will be more than a tool, it will be an integral part of every advertising strategy, across both the buy and sell sides. 

However, as traditional identifiers decline in value and alternative IDs unlock better ad performance — and revenue potential — industry players should prepare for a significant shift in the identity landscape next year.

“We’ll also see the convergence of AI and identity, where learning models use verified identity graphs to enhance targeting precision while maintaining privacy controls,” he said. “The winners will be those who combine deterministic accuracy with adaptive intelligence — delivering both performance and accountability in a privacy-first ecosystem.”

Parnter insights from IntentIQ