How identity will unlock retail media's next advantage

Retail media has given brands a more direct line between advertising and consumer action than traditional media ever could.

Within a retailer’s ecosystem, brands can see what consumers search, view, add to cart and purchase, then use those signals to inform media decisions closer to the point of sale. That combination of deterministic first-party data and closed-loop measurement has made retail media one of the most valuable sources of consumer intelligence in marketing.

But that intelligence is still largely confined to the environments where it is collected. A consumer may move from streaming TV to social, from a retailer site to a brand’s DTC site, and from one retailer to another before making a purchase. Each platform can understand part of that journey, but few brands can connect the full picture.

The result is fragmented audience decisions, incomplete measurement and repeated exposure to the same consumer across channels. Identity is the next step toward making those signals work harder together.

Identity is the connective tissue

Identity solutions can help brands recognize the same consumer or household across media and commerce environments in privacy-conscious ways. These solutions aren’t meant to replace retailer data, but rather make retailer data more useful by connecting it with approved signals from other environments, such as brand-owned data, loyalty activity, DTC purchases and media exposure.

This is where identity partners, retailers and brands must work together. Retailers hold valuable deterministic commerce data, but their view is typically limited to their own ecosystem. Identity partners such as Acxiom can help resolve those privacy-safe signals alongside other approved data sources.

Clean rooms provide a controlled environment for analyzing those connections without exposing individual consumer information. The value is the ability to make more informed decisions about who to reach, what message they need and where the next media dollar can have the greatest impact.

With a connected view of consumer behavior, brands can begin answering questions such as:

• Which consumers are most likely to become new-to-brand?

• Which consumers are most likely to trade up into a premium product?

• Which consumers have already purchased elsewhere?

• Which consumers are unlikely to be influenced by another impression?

• Where should the brand invest more, less or differently?

Identity turns fragmented signals into a more useful view of consumer behavior. That can improve audience decisions, create more coordinated media and commerce strategies and help brands manage frequency around the consumer rather than around individual platforms.

From audience targeting to predicted impact

Retail media audiences have traditionally been built around what consumers did in the past. They searched for a product, visited a detail page, added an item to cart or made a purchase.

Those signals remain valuable. But when commerce signals are combined with identity and consumer intelligence, brands can move beyond asking who looks like a likely category buyer. They can begin asking who is most likely to be influenced by the next media dollar.

One early example comes from work connecting Acxiom’s consumer intelligence with Amazon’s media and commerce signals. Acxiom’s integration into Omnicom creates a particularly interesting opportunity: 2.6 billion verified IDs with demographic, behavioral and offline attributes combined with retail media signals, all connected via clean rooms.

Within Amazon Marketing Cloud, brands can analyze advertising exposure and retail events, including impressions, product-page views, cart additions, orders, new-to-brand purchases and subscriptions, alongside approved third-party consumer attributes.

Our propensity models use these privacy-safe, aggregated signals to estimate which audience groups are most likely to drive a defined outcome, such as a new-to-brand purchase or an incremental response to media. Rather than targeting only consumers who resemble past purchasers, brands can prioritize consumers where investment has the greatest potential to influence behavior.

For example, a brand may identify consumers who are more likely to respond incrementally to media investment, rather than simply consumers who resemble past purchasers. Those insights can inform audience prioritization, bid adjustments and suppression strategies in Amazon DSP.

The objective is not necessarily to spend more. It is to concentrate investment where it has the greatest potential to change behavior.

In an early CPG test, an Omnicom propensity solution combining Acxiom and Amazon signals delivered a 93% match rate, a 13% improvement in new-to-brand sales and a 15% increase in ROAS compared with a control group. These results reflect one test, but they illustrate the potential of combining identity and commerce data to make more intentional investment decisions.

There is an operational lesson here, too. Audiences must remain current. Early testing showed the strongest results immediately after an audience refresh, with performance weakening as the underlying inference pool aged. Identity is a learning loop that needs to incorporate new consumer behavior over time.

From channel strategy to consumer strategy

Identity becomes even more valuable when a brand can connect retail media data with signals from its owned channels and broader media activity.

Consider a consumer who purchases a new skincare serum on Amazon. Before purchasing, they may have seen a CTV ad, watched creator content, searched for the product, visited the brand’s website or encountered sponsored ads on a retailer’s site.

In a disconnected system, the Amazon purchase may not influence what happens next. The brand could continue serving introductory acquisition messaging across social, programmatic media and owned channels because those environments cannot recognize the purchase or see the exposure delivered elsewhere.

In a more connected system, that purchase becomes one signal within a broader consumer journey. It may indicate that acquisition messaging is no longer appropriate. The next action could be to reduce repetitive product-introduction ads, test a replenishment message, introduce a related product or avoid additional messaging entirely.

The goal is not to make Amazon function like DTC or ask every channel to play the same role. CTV, creator content, search, retail media and owned channels each contribute differently to creating and capturing demand.

The opportunity is to use the signals from each environment to make the others smarter.

That also creates a stronger foundation for measurement. Connecting media and commerce signals does not make every post-exposure purchase causal. But it can help brands build better evidence about where media contributed to demand, how channels work together and where the next dollar is most likely to influence consumer behavior.

Frequency may be the simplest place to start

The most practical identity use case may also be the least glamorous: frequency management.

Excessive frequency happens when the same consumer sees a brand message repeatedly across channels, often without the brand realizing it. If someone has already seen an ad 30 times without taking action, the 31st impression is unlikely to change the outcome.

This is rarely the result of one poor decision. It is the predictable outcome of disconnected plans, separate channel budgets, fragmented data and unclear ownership of the consumer journey.

Brand media teams may be accountable for reach. Retail media teams may focus on conversion. Lifecycle marketing teams may prioritize retention, while agency partners optimize to platform-specific goals. Each group can make sensible decisions within its own remit while the consumer receives redundant messages across streaming, social, search, retail media and other environments.

Within a single retailer ecosystem, brands can already use clean-room analysis and tools such as Flywheel’s ADSP Site Filtering & Optimization model to identify when additional exposure reaches the point of diminishing return. They can reduce wasted impressions by suppressing overexposed audiences or shifting spend toward consumers who have had less opportunity to respond.

A shared identity foundation expands that capability across environments.

A brand may see eight impressions on Amazon, six on Meta, five on TikTok, seven through programmatic media and four through another retailer. Each platform sees a reasonable number. The consumer experiences 30 impressions.

Identity will not create a universal cross-platform frequency cap overnight. Platforms have different rules, permissions and privacy requirements. But it can help brands make duplicated exposure more visible, coordinate exclusions where possible and move toward frequency decisions based on the consumer’s total experience rather than the count within one ecosystem.

The easiest media dollar to improve may be the one a brand stops spending on an impression the consumer does not need to see.

Start with one valuable decision

Brands do not need to connect every platform or solve enterprise identity before they begin. Start with one decision that would be more valuable with a connected consumer view.

Choose the business problem. Define the outcome first. Is the goal to acquire more new-to-brand consumers, suppress existing purchasers, improve cross-sell, measure media contribution or reduce excessive frequency? The identity strategy should follow the decision.

Map the signals and gaps. Identify the data already available across retail media, DTC, loyalty, lifecycle marketing, social, programmatic media and clean rooms. Most brands have more useful signals than they actively connect.

Build a controlled, privacy-safe test. Define the audience, desired outcome, holdout group, refresh cadence and success metrics before activation begins. Establish permissions, matching requirements and aggregation thresholds from the start.

Connect the teams as well as the data. Identity can reveal that the same consumer is being influenced across multiple environments. It cannot decide which team owns the next action, which budget should shift or which channel receives credit. That requires shared goals, decision rights and operating rhythms across commerce, brand media, analytics and agency teams.

Without that alignment, a brand may see the consumer more completely while continuing to make fragmented decisions.

The next identity advantage

Retail media changed marketing by connecting advertising more directly to consumer behavior and sales.

The next opportunity is to connect that intelligence across the broader consumer journey.

Identity will not eliminate platform boundaries, make every impression measurable or prove that every ad caused a sale. It can, however, help brands make better decisions about who should see the next ad, where and why.