Google has abandoned its plans to deprecate third-party cookies, but marketers should continue to invest in consumer data and identity resolution. Having access to audience data helps marketers more accurately serve messages to in-market consumers and tailor their campaign messaging for maximum effectiveness. It also helps them measure outcomes, identifying which channels or campaigns moved the needle with consumers.
And the implications of identity resolution go beyond just measurement. Ideally, a more targeted campaign leads to more personalized recommendations, simplifying the customer journey and improving the customer experience.
Even with third-party cookies in play, the ways that marketers can collect and access customer data are changing. Consumers have become more informed about how and why companies collect their data, prompting many to opt out of sharing any data at all. Increasing statewide privacy laws are also limiting marketers access to data.
- Want to learn more about ad measurement and other marketing trends? Sign up for the ĢAV Daily newsletter.
To ensure a holistic view of the customer, marketers should explore privacy-preserving ad measurement tactics and privacy-safe identity solutions, like universal IDs, seller-defined audiences, or .
But not every digital channel is the same when it comes to data and measurement. Marketers must understand the identifiers and measurement methods used across each channel and find a way to stitch them together for a more robust picture of their audience.
This guide explores the facets of identity resolution; how it impacts marketing performance; and what measurement looks like across digital channels, including display, out-of-home (OOH), retail media, and podcasting. It also outlines the emerging identity solutions that marketers may consider as they battle signal loss.
What identity resolution is and how marketers measure advertising
Customer data enables advertisers to target their campaigns at the right audience and measure how those campaigns performed.
Understanding the different types of consumer data
There are four types of consumer data, each named according to how the data is collected.
- Zero-party data is data provided voluntarily and directly by consumers. Examples include customer surveys, polls, quizzes, or feedback forms.
- First-party data is collected directly from consumers via brand’s own channels, like their website or app. Examples include browsing behaviors; add-to-cart, conversion, and cart abandonment rates; purchase history; or personal and demographic information.
- Second-party data is another organization’s first-party data, which can be purchased or shared via partnership.
- Third-party data is data aggregated from various websites and apps by providers without direct consumer relationships. For example, browsing history, interactions with websites, or purchases.
To illustrate, cookies stored by a website a consumer is visiting are considered first-party cookies, while those stored by a different entity are considered third- party cookies.
Currently, there’s no shortage of customer data available to marketers, which can be collected and stored by a to form a single customer profile. Available data sources include:
- Demographic (gender, age, marital status, profession, location, income)
- Behavioral (website and app visits, email opens, social engagement, foot traffic)
- Attitudinal (survey data, Net Promoter Score, social listening data, customer service data)
- Transactional (transaction frequency, average spend per transaction, spend across demographics or geographies)
- Digital identifiers like an IP address or a device ID
Because there are so many different sources of data available, marketers may consider investing in identity resolution, which is the process of connecting disparate sources of consumer data together to create a single identifier for an individual.
The effect of privacy legislation on identity resolution
In 2022, the on looming privacy regulations, warning that the ad industry wasn’t prepared to deal with its impact. As individual state laws are passed, they change how businesses can collect and use consumer data, which may make it more difficult for advertisers to use that data for targeting purposes.
- As of July 2024, 20 states have passed comprehensive data privacy laws in the US, per White & Case LLP.
- A drafted federal privacy measure, the American Privacy Rights Act (APRA), was introduced in April 2024. If passed, the APRA would eliminate the need for the current patchwork of state-specific privacy laws and rival the EU’s General Data Protection Regulation (GDPR) to become “one of the leading global privacy standards,” according to international law firm Pillsbury Winthrop Shaw Pittman LLP.
- However, the APRA was put on hold when the markup for the bill was canceled in June 2024, leaving the fate of the bill uncertain.
Why Google walked back plans to deprecate third-party cookies in Chrome
In July 2024, Google announced it would no longer be getting rid of third-party cookies in Chrome, abandoning the plan it’s had for deprecation since 2020.
- Instead, the company said it would introduce a “new experience” in Chrome that would enable users to make “informed choices” about their web browsing data which could be adjusted at any time.
- Google did not provide additional details on what that experience would look like or when it would launch.
- The company confirmed it would continue to make its Privacy Sandbox APIs available and invest in them to further improve privacy and utility.
Google’s decision to abandon cookies most likely stemmed from negative feedback from industry regulators like the UK’s Competition and Markets Authority (CMA) and other ad players.
In April 2024, the CMA released its Q1 2024 update report on implementation of the Privacy Sandbox commitment, which outlined the regulator’s many concerns about third-party cookie depreciation and Google’s Privacy Sandbox. This led to Google delaying cookie deprecation through 2025.
- After an eight-week test, Criteo concluded the Privacy Sandbox falls short of Google’s goal of limiting revenue lost to 5%. Instead, Criteo believes if third-party cookies were deprecated today, publishers would lose an average of 60% of their revenue from Chrome
- The Interactive Advertising Bureau (IAB) Tech Lab said that the current Privacy Sandbox would “restrict the digital media industry’s ability to deliver relevant, effective advertising” and “throttle” smaller brands’ ability to compete.
The emergence of alternative identity solutions
An is a product or service that helps identify individuals or households across digital channels like web browsers, mobile apps, connected TV (CTV), or other devices, per the IAB’s Identity Solutions Guidance report.
Identity resolution isn’t new, but changes in digital advertising (like Apple’s AppTrackingTransparency framework and mounting privacy laws) have accelerated advertisers’ pivot away from traditional identifiers (including cookies or mobile ad IDs) to alternative identity solutions.
Identity solutions leverage deterministic data, probabilistic data, or a mix of both.
Deterministic data is known info, like an individual’s name or email address.
- Deterministic solutions use this type of data to match user profiles with a high degree of accuracy, ensuring advertisers can target more specific audiences.
- Much of this data lies within walled gardens (provided by consumers to create logins or user profiles), making it hard to scale outside of those environments.
Probabilistic data consists of temporary information that changes often, like IP addresses, device details, or time stamps, which is then analyzed and used to infer a user’s identity or behavior.
- Marketers can use this data to scale campaigns more effectively, but it is less precise than deterministic data.
- A mix of deterministic and probabilistic data is necessary for an identity solution that addresses the full customer journey.
Probabilistic and deterministic can also refer to how data is matched together. Deterministic matching links data sources or sets together based on a common identifier, usually an email address. Probabilistic matching uses different data sources, sets, and algorithms to identify the same user across different devices and applications.
The following list includes current identifiers and identity solutions on the market, including universal IDs, Google’s Privacy Sandbox solutions, and other privacy-preserving tools.
Universal IDs
Universal identity solutions are a privacy-compliant way for advertisers to keep track of consumer behavior across the internet. They use a unique and persistent data signal (like an email address or a phone number) to create an anonymous identifier, which can be encrypted and shared with advertisers, publishers, and ad tech platforms.
Many universal ID solutions currently exist on the market, both deterministic and probabilistic, including:
- The Trade Desk’s Unified ID 2.0 (UID2) uses a single piece of deterministic data to create an encrypted, anonymous identifier.
- ID5 ID utilizes various signals, such as hashed email addresses, page URLs, IP addresses, and timestamps, to generate a unique user identity.
- LiveRamp’s RampID LiveRamp’s AbiliTec identity resolution technology transforms consumer data, such as email addresses, names, or postal addresses, into AbiliTec IDs.
- Lotame’s Panorama ID utilizes a combination of deterministic and probabilistic data to link consumer behavior across web, mobile, and CTV platforms. This identifier connects device identifiers, behaviors, and emails without depending on any single signal, such as cookies.
- LiveIntent HIRO addressability solution utilizes LiveIntent’s database of logged-in users, publisher relationships, and identity graph to enable audience addressability across the web.
- Yahoo ConnectID utilizes Yahoo’s Identity Graph, which is derived from logged-in users across Yahoo Mail and other owned media sites. Yahoo ConnectID can be integrated with Yahoo’s other identity solutions, including audience and measurement tools.
- Criteo ID consists of three components: an identity graph, shopper browsing behavior, and transaction history to create a user identity.
- Neustar Fabrik ID replaces audience and subscriber IDs with a FabrickID in real-time. This ID automatically substitutes personally identifiable information (PII) with an alias identifier, safeguarding both publishers’ data and consumers’ privacy.
- Prebid SharedID creates unique first-party identifiers for publishers to recognize users and associate interests with them.
Non-ID solutions
Solving for identity will likely require advertisers to adopt additional solutions outside of universal IDs, such as:
Google Privacy Sandbox
Google’s Privacy Sandbox provides marketers with ways to phase out their use of third-party cookies while still being able to target ads to users across the web.
Currently, the Sandbox consists of nine APIs which fall into four categories based on their purpose:
- Fighting spam and fraud on the web (Private State Tokens API)
- Showing relevant content and ads (Topics API and Protected Audience API)
- Measuring digital signals (Attribution Reporting API)
- Strengthening cross-site privacy boundaries (Related Website Sets API, Shared Storage API, CHIPS API, Fenced Frames API, Federated Credential Management API)
The most important APIs for marketers to start with are the Topics API, Protected Audience API, and Attribution Reporting API, as those directly relate to ad targeting and measurement.
Amazon Ad Relevance
Amazon has debuted Ad Relevance, its own cookieless tracking alternative. Ad Relevance is a service designed to help advertisers target online audiences without relying on third-party cookies or the vast ID solutions that have emerged in recent years. This offering operates within Amazon’s DSP (Demand Side Platform). It uses signals such as browsing, shopping and viewing behavior across Amazon properties to predict user behavior and then offers media buyers ad opportunities.
Cohort-based solutions
Cohort solutions group consumers by interest, creating more contextually relevant (but less personalized) campaigns.
Google’s Topics API and Protected Audience API are both cohort-based solutions, as is the IAB Tech Lab’s seller- defined audiences (SDA) framework.
Data clean rooms
A data clean room is a secure digital environment where multiple parties can commingle their first-party data to produce audience and campaign insights. Data clean rooms add another layer of privacy for marketers and can be used in conjunction with other identity solutions.
Adopting an identity solution
Before , marketers should identify gaps in organizational data and provide more insight into what kind of solution (or solutions) are right, get organizational buy-in from leadership, and continue to test and learn what works and what doesn’t. In addition, marketers should reconsider how they’re approaching measurement and keep privacy at the core of any identity solutions they implement.
Because many of these solutions are still in their early stages, marketers must leave room for trial and error. In addition, identity solutions can be quite costly and require the appropriate expertise and skills to apply them correctly.
According to Winterberry Group, in the United States alone, $27 billion will be spent in 2024 on data, data services, and data infrastructure to support an estimated $270 billion in U.S. media buying across data-driven channels.
How identity resolution and measurement is changing across channels
Each digital channel is facing its own set of challenges when it comes to identity resolution and management.
Digital display is caught between the past and the future
The dust is settling from Apple’s AppTrackingTransparency. and mobile advertisers are updating their measurement strategies with new tools and frameworks, according to ĢAV’s report.
Web-based advertising has gotten a bit of a reprieve, as third-party cookies aren’t going away anytime soon.
- But marketers should still be building out their first-party data strategy, exploring how identity solutions can help improve targeting and measurement, and focusing on data governance, which ensures customer data is being securely, legally, functionally, and efficiently managed.
- This may become especially important if a federal privacy law is ever passed.
Lack of measurement standards threatens retail media growth
While retail media is one of the fastest-growing ad channels we track, measurement challenges could jeopardize its future.
- The in September 2023 that aim to help the industry achieve comprehensive standardization, enabling the retail media channel to continue to grow.
- Agencies and ad tech vendors are also launching tools and solutions to help advertisers navigate the headache of executing and optimizing strategies across retail media platforms, per ĢAV’s report.
- For example, in August 2023, global media agency , a “first-of-its-kind” retail media budget allocation and optimization tool. This tool is part of IPG Mediabrands’ Unified Retail Media Solution, a dedicated business unit that helps brands manage their investment performance across all .
DOOH measurement is catching up
Having solved for the exposure piece of OOH measurement, advertisers’ next big hurdle is making OOH metrics compatible with other digital channels, per ĢAV’s Ad Measurement Trends H2 2023 report.
- Programmatic will account for 26.4% of US ad spending in 2024, according to ĢAV’s June 2024 forecast.
- While technical standards have evolved to facilitate tracing alongside other digital media, there’s still no methodological standardization for omnichannel measurement.
Podcast measurement is struggling to break free of legacy tech
Over two-thirds (66.9%) of the in 2024, including podcasts, per an ĢAV August 2023 forecast. But between 2021 and 2025, will remain essentially flat at about 2% of total media spending, according to ĢAV’s Ad Measurement Trends H2 2023 report.
- This is probably due to a , per the same report.
- Because most podcast content is delivered via download, it makes it difficult to measure exposures and outcomes. This is muddled even further by the transition from primarily embedded ads to dynamic insertion.
- It’s in the best interests of to move away from the download-based model, otherwise advertisers could begin to move away from the channel.
This article has been updated. Original was posted January 23, 2024.
