Marketers today describe their ideal customers in extraordinary detail, with rich personas that go beyond basic demographics to include attitudes, motivations, values, preferences and even future intentions.
But a media buyer still has to turn that persona into a handful of targetable attributes. And, ironically, the more specific the picture of an ideal customer becomes, the harder it gets to find that exact person in the media ecosystem.
Segmentation research is a major investment. Having spent the money to find out who your customer is, your buy should match the audiences you’re looking for.
Why rich customer personas get lost in media buying translation
Consider a persona we’ll call “The Experimenter,” someone who is novelty-led, receptive to new products and open to switching brands. This definition may be incredibly useful to a marketer, but there’s no “Experimenter” button in a DSP.
Instead, the persona gets retranslated at every handoff: from consumer insights to marketing, from the media brief to the agency and finally into each buying platform. “Novelty-led and open to switching” becomes “busy professional parents who want convenience,” then “parents ages 30 to 45 with household incomes above $75,000.” By the end, the actual buy might be reduced to “adults 25 to 54.”
Each team is doing its best to translate the persona with the tools it has. Even so, the attitudes and motivations that made the persona strategically valuable are the first things to disappear.
Media teams understand this challenge and optimize campaigns to the nearest match, but brand and insights teams may be less aware. The persona goes into the brief while reporting comes back as demographic reach, CPMs and performance metrics. These two audience definitions may never even get compared.
The brand team thinks the campaign is reaching the persona it developed, while the media team knows they’re only buying the closest available approximation.
Media buying focuses on the ‘what,’ but marketers already know the ‘why’
The media ecosystem is extraordinarily good at collecting signals about what people do: purchases, searches, site visits, content consumption, location and countless other observable behaviors. But those signals don’t always explain the why.
Someone researching sustainable packaging might be a true sustainability-minded shopper – or they might be researching the topic for work. A loyal brand customer will look identical online to someone actively open to switching allegiances – right up until one of them actually switches.
The missing context is already in the marketer’s hands via research. Consumer insights teams have spent decades asking people directly about their attitudes, preferences, motivations and intentions. This declared data is exactly what gives marketing personas their richness. The challenge is carrying that all the way through to the media buys.
Using consumer insights for media activation
In most cases, survey data is used for analysis but not media activation. Consumer insights and media typically operate with different teams and budgets, and research samples don’t get connected to addressable IDs.
That’s changing with consented research panels and privacy-safe identity matching. Consumers who declare relevant characteristics can become the foundation of an addressable audience. Their records can be matched to pseudonymous identifiers, modeled to find lookalikes and activated as segments in DSPs, social platforms and across CTV environments.
Media buyers no longer have to guess whether someone researching sustainable packaging is actually open to switching brands. They can activate audiences built from consumers who have definitively declared these characteristics.
The best way to learn what a consumer wants is to ask them, not to infer it from another signal.
It’s possible for precision to scale
Targeting based on declared data still faces the classic trade-off between precision and scale, of course. But a research-defined audience provides a better starting point: people who self-declare their attitudes or intentions. And modeling can expand outward when a campaign needs more reach.
Declared data also complements behavioral and transactional signals. A marketer can combine people who say they’re open to switching brands with signals showing they’re currently shopping the relevant category.
Behavioral signals determine the right timing, but declared data identifies the right people.
This dynamic also makes campaign optimization more meaningful. Algorithms learn from the audience they’re given, so beginning with more accurate data creates a better feedback loop between the ideal customer profile, reach and campaign performance.
This doesn’t require a new research project for every campaign
None of this means marketers need to commission a segmentation study before every media buy. Many declared data audiences already exist. Dynata, for example, maintains more than 3,000 targeting segments built from more than 10,000 profile variables, and its syndicated audiences are already available through LiveRamp. Custom audiences using existing variables can typically be created in three to five days.
Consider an electric vehicle brand trying to find buyers for a luxury EV. Household income shows who can afford to buy one, but declared data, including readily available syndicated audience data, can help distinguish between people who would never consider an EV and those who are persuadable, already interested or simply need a different message.
Do more with what your customers already tell you
Marketers are constantly searching for more signals, more data and more sophisticated modeling. But they may be overlooking some of the richest customer intelligence they already have.
Here’s a simple gut check: Look at the persona your organization says it wants to reach. Then look at the audience definitions you buy across media platforms.
How much of the desired persona survives?
The exciting part is the simplicity behind this approach. Instead of chasing ever-more-complex ways of inferring what consumers want, the biggest gains will come from acting on what they’ve already told you.
