Dean de la Pena on how Resonate’s AI transforms martech with privacy-first identity, deep psychographics, and adaptable data stacks for compliant personalization at scale.
Dean de la Peña, Vice President of Identity, Data Strategy, and SaaS at Resonate, brings over 16 years of experience scaling technology and AI companies to transform how brands understand and engage with their audiences. From helping lead Applied Predictive Technologies to a $600M acquisition by Mastercard to his current role at Resonate, he has built a career at the intersection of innovation, strategy, and impact.
In this conversation, he explores how AI reshapes identity resolution in a privacy-first era, the technical and ethical challenges of balancing personalization with compliance, and what it takes to build a future-ready martech data stack. He also shares how Resonate’s advanced predictive AI, rAI, is powering real-time, consent-driven insights on over 250 million consumer profiles—helping marketers move beyond basic demographics to truly actionable psychographics, values, and intent signals.
His insights cut through the noise on AI, privacy, and data strategy, offering a roadmap for marketers who want to deliver personalization at scale—without compromising trust.
Full interview;
How is AI fundamentally reshaping identity resolution in a privacy-first world? What are the core technical challenges in ensuring its accuracy, compliance, and sustained utility?
Privacy-oriented standards, regulations, and consumer preferences challenge a traditional identity approach. A responsible, ethical, and compliant approach to linking online consumers and their identifiers must center on pseudonymized, opt-in data points – of which there are fewer than ever.
Sorting through this limited data to create proper identity resolution can still be done, but it requires cutting-edge predictive AI models to connect the dots effectively. These non-linear, deep-learning models can be costly to train and tune over time to find valuable connections on a smaller base – an investment of tens or even hundreds of millions of dollars – but the ability to deliver up-to-date and deep consumer insights beyond basic demographics is priceless. Key information like political preferences, core values, spending intent, and behavioral motivations goes far beyond basic demographics and helps marketers truly personalize campaigns and drive much better profitability and engagement across channels.
AI-powered identity resolution can be practical and compliant because the data used is both pseudonymized and consented.
What are your critical technical considerations for data governance, model interpretability, and scalable operationalization to deliver actionable insights for AI and predictive models in customer analytics?
Strong data governance starts with data handling. Practices such as data minimization, hashing, and pseudonymizing help brands meet privacy and compliance standards and reduce data bias. This strengthens trust in the ethical quality of the inputs and gives marketers greater confidence in the insights.
Model interpretability starts with a reminder to focus on the end goal – engagement and return on ad spend – and begs the question, “Is interpretability truly necessary?” The most predictive, accurate modeling techniques are typically less interpretable but create the strongest consumer engagement and most profitable response to digital marketing. When necessary to support strategy and decision-making, techniques like gradient boosting interpretation and Shapley additive explanations can provide the intuition and interpretability behind the outcomes that drive true personalization and marketing excellence.
Finally, AI needs to drive outcomes directly to have a real impact at scale. Connecting AI-driven data directly to efficient, generative AI-based activation platforms enables marketers to immediately translate campaigns into engagement by the consumers who are readiest to engage with campaign content.
What core architectural principles and technologies define a resilient, future-ready data stack for modern martech? How does Resonate balance diverse data pipelines, scalability, and strict compliance?
A future-ready data stack for modern martech is built to flex with the pace of consumer change. It must support AI-powered real-time data to optimize the marketing engine. Suppose marketers can’t integrate the freshest behavioral shifts and psychographic signals into their predictive modeling. In that case, they may miss their window of opportunity to reach a consumer at the precise moment they’re ready to make a buying decision.
At Resonate, our goal is always to deliver scalable, trustworthy insights while upholding the highest standards of data responsibility.
How do we do this? It starts with strict compliance as an inviolable constraint. While our data is human-centric, it’s also privacy-first by being opt-in, pseudonymized, and exclusive of personally identifiable information—even before we translate it into human-unreadable, embedded vectors that simultaneously maximize privacy and prepare the data for AI processing. From that point, rAI, Resonate’s advanced predictive AI, powers the deep, continuously updated intent-driven consumer data insights from 15,000 attributes on 250 million consumer profiles that drive industry-best marketing results for our clients.
To enhance scalability, we’ve integrated natural language prompts into our rAI-powered Audience Builder so that marketers can create sophisticated audience segments in seconds. And increasingly, generative AI within our Ignite activation platform is driving directly to the most engaged audiences for any set of content without difficult decisions on the part of the marketer. There’s no need for time-intensive manual searches through thousands of attributes or deep technical skills, making activation much faster, scalable, and effective. Marketers can reach out when consumers are ready to buy, not days (or worse, for many other data offerings, weeks or months) late.
Beyond technology, what fundamental strategic shifts must organizations embrace to effectively leverage the next evolution of identity resolution, balancing personalization with robust privacy and consumer trust?
The winners of the next stage of identity resolution will be the brands that see privacy and personalization not as at odds, but as intrinsic to the same goal. Data from our 2025 State of the Consumer Report shows that consumers will continue to expect personalized experiences while demanding stronger privacy protections. Contrary to popular belief, privacy-focused consumers aren’t a small group with a niche concern: they comprise over 32% of the population across all generations. If brands aren’t already prioritizing the combination of personalization and privacy, they must start now or risk losing the trust of their customer base. This requires a mindset shift for many companies.
Assessing the data they use to personalize marketing is another foundational element to enact change. Only cutting-edge predictive AI, centered on real-time behavioral and psychographic cues gathered through transparent, consent-driven, pseudonymized data, can deliver on both privacy and personalization.
As a product and strategy leader, how do you consistently translate cutting-edge AI and data innovations into clear, measurable customer value, especially within complex, regulated sectors?
Translating AI and data innovation into value boils down to keeping sight of the core objective: a greater marketing return that stems from stronger personalization and better consumer engagement.
It’s easy to be distracted by trying to answer the wrong question or by trying to achieve perfection. My clients build much stronger, longer-term, and more valuable relationships with their customers by knowing who is likeliest to engage, purchase, or churn with 80% confidence – along with their preferences and motivations – than having 95% confidence about simple demographics like age and gender. The latter is handy, sure; but the former is game-changing, especially in this era of increased privacy constraints and heightened responsibility.
This is hardest in regulated sectors that limit the data available for use. It is where best-in-class predictive models that can operate on a restricted data infrastructure come in. Our ability to provide insights to healthcare-focused or financial services firms is unparalleled, given our combination of model transparency and our predictive AI, rAI’s, ability to accurately glean regulation-friendly intent and preference insights despite constraints.
At the end of the day, it boils down to results. Our clients often see a 30%+ increase in marketing ROAS and consumer engagement. Focusing on ensuring that we’re consistently driving real, measurable benefit is at least half the battle.
In a rapidly evolving martech landscape, what is your strategic vision for building data stacks that are not just future-ready but inherently adaptable and resilient to unforeseen market disruptions, changing regulations, and shifting privacy paradigms?
It starts and ends with speed combined with a focus on consumer behavior now. Aging data infrastructure and old-school thinking often combine to rely on data and insights that are weeks and months old. In a more stable 2015, that flew – but in 2025, with constantly changing headlines and shifting economic and geopolitical uncertainties, it’s no exaggeration to say that engagement and purchase patterns change daily. Just look at consumers loading up on pre-tariff French wine or German-manufactured cars.
If we can accurately track today’s behavior and align it to preferences, motivations, values, and intent, and have developed the flexibility to adjust our approach to shifting constraints and limited information, we’re future-proof. Thanks to rAI, we have done both.
