At A Glance
Artificial intelligence can improve marketing performance when the data behind it is complete, accurate, and governed. This article explains three data investments that will improve AI’s value in marketing and shows how an AI-ready data foundation can help teams plan, activate, and measure.Advertising has entered a practical AI era, not the sci-fi version. Teams are already using models to predict outcomes, automate decisions, and tailor creative at scale. In plenty of cases, the technology does exactly what it should.
When results plateau, it’s rarely because AI “doesn’t work.” More often, it’s because the inputs aren’t built for what AI is trying to do. AI doesn’t invent customer understanding, but amplifies the foundation you feed it, good or bad. The upside is encouraging: when brands treat data readiness as a growth move (not an IT chore), AI becomes faster to activate, easier to trust, and simpler to prove.
Here are three data gaps that quietly cap AI’s value in marketing, and the investments that open up more room for measurable impact.
1. Why first-party data only gives you part of the picture
First-party data gives marketers a valuable view of customer activity, but it rarely tells the full customer story on its own. AI performs better when models can see both direct brand interactions and broader context, such as life stage, media behavior, and shifting intent.
First-party data captures how customers interact with your brand across email, websites, apps, transactions, and loyalty programs. First-party data is excellent at showing what someone did with you, but it’s less reliable at revealing who they are outside of those interactions. This narrower view can leave AI models optimizing around incomplete customer understanding.
A grocery chain may know a household buys cereal every week. But that doesn’t automatically tell you whether they’re a new parent, recently moved, in-market for a vehicle, or more likely to respond to a premium offer than a discount.
There’s another limitation to first-party data: AI tends to optimize around the patterns it can already see. When models rely primarily on first-party data, they often become highly effective at finding more customers who look like existing ones. While that can improve efficiency, it can also cause AI to overlook emerging audiences, changing consumer behaviors, or new sources of demand that aren’t yet reflected in the brand’s customer base.
AI thrives when context is available. Without it, models still produce outputs, but they’re forced to guess more often than anyone wants to admit.
Invest in the data foundation: Enrich for decisions, not vanity
To make first-party data more useful for AI in marketing, marketers should:
Brands can close these gaps by enriching first-part data with high-quality demographic, behavioral audience insights that support planning, activation, and measurement. Experian helps marketers build a more complete customer view by connecting first-party data with trusted marketing data and audience insight. When first-party data becomes more complete and consistent, AI stops optimizing around partial truths and starts driving decisions you can stand behind.
2. Why stale identity makes AI scale the wrong patterns quickly
AI can move fast, which makes identity accuracy and data freshness central to performance. If customer records are duplicated, outdated, or mislinked, AI can learn from the wrong signals and repeat those errors across campaigns.
A common example is simpler than people expect: a profile tied to the wrong household, or a record carrying outdated attributes. Even with sophisticated bidding, frequency, and creative selection, you’re still serving the wrong message to the wrong person, just more efficiently. This is where many “AI didn’t work for us” stories start. The more optimistic framing is that this is also where many turnaround stories start.
Invest in the data foundation: Make “one person” actually mean one connected view
Identity resolution gives AI a more stable base for targeting, personalization, and measurement. Marketers can improve that base by reducing fragmentation before data reaches the model. To make identity more reliable for AI in marketing, marketers should:
Accurate identity resolution is central to effective AI in marketing. When customer records are unified and current, AI models can recognize people and households more accurately, which improves targeting, personalization, and measurement. Fragmented or outdated identities can create errors, from duplicate messages to irrelevant offers.
Experian helps marketers organize, connect, and expand their data into a connected foundation, so teams can act on insights across systems, channels, and tools.
When identity and freshness hold up, AI has stable ground to stand on, and gains show up faster and more reliably.
3. How data governance makes AI scalable
Clear data governance helps AI programs scale. When teams know how data can be accessed, shared, and used, they can scale AI with greater confidence and fewer operational risks. As AI moves deeper into marketing workflows, data becomes more valuable, not only for targeting but for model training, personalization, and measurement.
That value cuts both ways. Sloppy access controls, unclear permissions, and vendor sprawl can erode customer trust and dilute business advantage. And when teams don’t feel confident in governance, they hesitate to scale. AI becomes a pilot project instead of a business capability.
The good news is that brands don’t have to choose between using data and protecting it. The winners operationalize both.
Invest in the data foundation: Treat governance as performance infrastructure
Data governance isn’t separate from AI performance and is the layer that helps leaders scale what works. To make data governance practical for AI in marketing, marketers should:
When teams understand how data can be accessed, shared, and used, they can move faster. Experian supports responsible data practices by building protection standards into activation workflows. That includes defining access by role, guardrails for partner data use, and compliance with privacy regulations. With these controls in place, brands can scale AI programs with greater trust.
The payoff: AI that’s faster, more relevant, and easier to prove
AI can raise the ceiling for advertising, but it can’t compensate for incomplete customer views, mislinked identities, or weak controls. It will accelerate what’s already happening, good or bad. The most successful brands will be the ones that make three investments: completeness, accuracy, and governance, so AI produces outcomes that can scale.
When the data foundation is ready, AI stops being a shiny layer on top of marketing and becomes a practical advantage inside it. Contact us to talk about building an AI-ready data foundation that supports stronger planning, activation, and measurement.
About the author
Jeremy Meade
VP, Data Operations & Governance, Experian
Jeremy Meade is VP, Data Operations & Governance, at Experian Marketing Services. With over 15 years of experience in marketing data, Jeremy has consistently led data product, engineering, and analytics functions. He has also played a pivotal role in spearheading the implementation of policies and procedures to ensure compliance with state privacy regulations at two industry-leading companies.
FAQs
AI-ready marketing data is complete, consistent, current, and governed enough for models to use with confidence. It connects customer activity, identity, and outcome signals so AI can support better planning, activation, and measurement decisions.
First-party data shows how customers interact with a brand, but it may not show broader context such as life stage, household changes, media behavior or shifting intent. AI needs that added context to reduce guesswork and help marketers make more relevant decisions.
Identity resolution affects AI performance by helping models recognize people and households more accurately across systems and channels. When identities are fragmented or outdated, AI can repeat the wrong patterns through duplicate messages, irrelevant offers or misread campaign signals.
Experian helps marketers build an AI-ready data foundation by connecting first-party data with trusted marketing data, audience insight, and identity resolution. Experian’s identity foundation helps teams organize, connect and expand customer data so AI can support stronger planning, activation and measurement.
Data governance matters for AI marketing programs since teams need clear rules for access, sharing and use before they can scale with confidence. Strong governance helps brands protect customer trust, reduce data sprawl, and turn AI from a pilot into a business capability.
Latest posts
Learn how brands can close the activation gap by carrying audience intelligence from identity to execution and outcomes across channels.
In our Ask the Expert series, we interview leaders from our partner organizations who are helping lead their brands to new heights in AdTech. Today’s interview is with Samantha Zhang, Senior Data Scientist, and Jim Meyer, General Manager of the DASH TV Universe Study at the Advertising Research Foundation (ARF). DASH is an annual tracking study conducted by the ARF to define and better understand TV audience behavior and household dynamics. What does DASH measure, and how does it help the industry understand TV consumption today? By capturing hundreds of individual- and household-level data points from each respondent in a rigorous and nationally projectable sample, DASH creates a comprehensive picture of U.S. consumer TV “infrastructure” – how America watches. Core elements in DASHElements that create context in DASHTV setsLocation | brand | smartness | service modes | sources DemographicsConnected devices Game consoles |video players | streaming devicesYesterday viewing Daypart | TV/device genre | Out-of-home viewingMobile devicesOwners | sharing usersShoppingOnline and in-store | Exposure to major RMNsInternet serviceModes | ISPs | connectivity by device Streaming audio Streaming TVSVOD/AVOD tiers and sharing | FAST Email accounts and apps Live TV Modes of access | including casting from devices Social media For example, DASH gathers: Data on every TV set, including brand, room location, age, “smartness,” and connection devices and modes Household connectivity and video service data, even in homes with no TV set Internet Service Providers (ISP) and TV service usage, including Multichannel Video Programming Distributors (MVPDs), virtual vMVPDs, streamers (ad-supported and premium), and Free Ad-Supported Television (FAST) channels Person-level ownership and usage of video-capable mobile devices, including smartphones, tablets, and laptops Measures of viewing and co-viewing across dayparts, devices, and services Additional modules covering shopping and retail media networks, streaming audio, social media, email, and apps Broad coverage and granularity make DASH a uniquely robust source of truth for practitioners across the industry, including measurement experts and ad programming strategists. DASH also reports regularly (and publicly) on key industry dynamics. DASH identified a growing segment of device-only viewers – now nearly 9 million households that watch TV, but do not own a TV set – and highlighted the implications of that trend for traditional ratings systems based only on households with TV sets. Households (HHs – million)2025 HHs (M) U.S. penetrationChange vs. 2024 (M)Total US134.8100%+2.7Connected TV (CTV)114.685%+2.1TV (Set)124.292.2%+1.1Device-only8.86.6%+1.6TV-Accessible133.198.7%+2.7 DASH called out the rise in app-based pay TV and proposed a new connection framework that better represents the modern TV world, in which linear and streaming overlap. DASH also defines the universes of households reachable with advertising. This graphic, for example, shows how all ad-supported linear and streaming properties in aggregate define the true scale of TV advertising. While 35 million households (and growing) are reachable only with streaming ads and 13 million (and falling) only with linear ads, most households are reachable with both, underscoring the importance of understanding the “overlap.” Who uses DASH data, and what decisions does it help inform? There are three primary users of DASH, each with its own use cases: Measurement providers, including Nielsen, use DASH to calibrate viewership data, turn household data into persons data (and vice versa) and estimate potential reached audiences–what the providers call media-related universe estimate (MRUEs)–for the calculation of ratings. Not surprisingly, measurement companies were the first to see the value that an independent TV universe study could provide. Media companies, including major broadcasters and streamers, use DASH to add context and color to their ad sales presentations – and to track the measurement providers, whose ratings play a major role in valuing ad inventory. AdTech companies, including Experian, use DASH to create high-value audience segments for activation. The recent accreditation of DASH by the Media Rating Council (MRC) and adoption by Nielsen as an input to its TV ratings have generated interest from a broad range of companies. We are actively pursuing new licensees and partners to make DASH more useful within, and even outside, the TV ecosystem. What does MRC accreditation signify, and why is it meaningful for DASH? MRC accreditation means DASH passed a rigorous audit conducted by Ernst & Young over many months, which validated our methodology, controls, and data quality. MRC accreditation establishes that DASH is an industry-standard dataset. While the service provider normally announces its own accreditation, the MRC took the unusual step of issuing its own release on DASH, announcing the accreditation of DASH for TV universe estimation and endorsing the study for broader, cross-media use. How does Experian use DASH data to build audiences? The segments combine specific TV usage habits and behaviors from DASH with Experian data on demographics, spending, and other contextual inputs to create a fuller view of consumer viewing behavior. They are designed to be valuable to advertisers in many categories and planning contexts – and to be customizable to fit advertisers’ media targets. The segments can be used to: Apply or suppress audiences to improve target coverage across a campaign Better align media and creative Reach elusive but high-value viewers, such as Ad Avoiders Drive valuable consumer behavior Achieve specific advertising objectives What are some practical use cases for DASH-based audiences? Here are some practical use cases for four different kinds of DASH segments in five different advertiser categories. Travel Co-WatchersA couples-only resort uses TV Co-Watching Households without Children to strengthen target reach and ad memory recallA big theme park destination uses TV Co-Watching Households with Children to reach families in moments of togetherness Home Entertainment TV Owners and Brand LoyalistsA premium TV manufacturer uses the overlap of Multi Brand TV Owners and Single Brand TV Loyalist Households to market its newest TV model to its most loyal consumers. Fast Food Screen Size ViewersA fast food chain with a high-impact new brand campaign uses Large Screen TV Viewers to better align the media and creativeThat same fast food chain uses Small-Screen TV Viewers to drive store traffic by increasing exposure of its retail campaign among on-the-go viewers Financial Services Cord Cutters A personal cost management app and a cash-back credit card target Streaming-First Cord Cutter Households to reach young, tech-savvy, cost-conscious consumers Thanks for the interview. Where can readers learn more about DASH? We started work on DASH seven years ago, and it’s been fun to watch it “grow up.” Our partnership with Experian is a big step toward putting DASH to work for advertisers and agencies. To learn more, visit our site at https://theARF.org/DASH or contact us at DASH@theARF.org. Contact us About our experts Samantha Zhang, Senior Data Scientist at ARF Samantha Zhang is a Senior Data Scientist at the Advertising Research Foundation working on the DASH TV Universe Study, with additional research spanning areas including attention measurement, digital privacy, and artificial intelligence. Jim Meyer, General Manager, DASH, at ARF Jim Meyer is general manager and co-founder of the ARF DASH TV Universe Study and managing partner of Golden Square, LLC, which advises media and research technology companies on growth strategy and development. Latest posts
Reach auto buyers based on how they shop in a changing market with Experian Audiences
Marketing data and identityReach buyers based on how they shop in a changing market. Experian Audiences help target switchers, align to budgets, and match messaging to inventory.