In a MediaBrief exclusive, Anish Roy, Director of Analytics, MiQ, speaks about how a strong foundation in statistics drives MiQ’s approach to solving complex challenges in programmatic advertising, and why data science is the engine powering innovation across identity resolution, real-time bidding, and creative optimization. Roy also shares how MiQ is evolving its analytics to meet the demands of emerging channels like CTV, DOOH, and retail media; how the company is balancing personalization with privacy in a consent-first era; and what excites him most about the future of advertising analytics.
Director of Analytics at MiQ, Anish Roy brings a deep academic and professional grounding in statistics to one of the most data-driven industries in the world—programmatic advertising. And for him, that foundation is not just academic theory—it’s the backbone of how MiQ addresses some of the sector’s most complex challenges.
“My academic and professional foundation in statistics has been absolutely instrumental in navigating the complexities of programmatic advertising,” Roy said. “It instills a rigorous, first-principles approach to problem-solving, emphasizing causality over correlation, robust experimental design, and a deep understanding of data distributions and potential biases.”
This methodical approach, Roy added, enables MiQ to build more accurate predictive models, design insightful measurement frameworks, and extract actionable intelligence from vast datasets. “It allows us to systematically deconstruct challenges like audience identification, campaign optimization, and incremental lift measurement with a level of statistical rigor that drives truly reliable outcomes for our clients,” he explained.
Anish Roy on MiQ’s analytics advantage
Operating at the convergence of data, science, and storytelling, MiQ’s engine of innovation is fueled squarely by data science. And according to Roy, it’s not just important—it’s essential. “Data science is not just critical; it’s the very engine of innovation in today’s programmatic media landscape,” Roy asserted.
In identity resolution, for instance, he explained how MiQ uses machine learning and multifaceted identity signals to piece together a cohesive, privacy-compliant view of consumers across a fragmented ecosystem. In real-time bidding, data science drives instantaneous, high-stakes decisions to maximize campaign efficiency across billions of impressions per second.
“And for creative optimization,” Roy continued, “it moves beyond simple A/B testing to dynamic, AI-driven personalization, ensuring the right message reaches the right person at the optimal moment via the right medium.”
In short, he said, “Data science enables MiQ to transform raw data signals into intelligent actions, delivering superior performance and measurable ROI.”
Evolving analytics to meet the demands of emerging channels
As newer channels like connected TV (CTV), digital out-of-home (DOOH), and retail media reshape the landscape, MiQ is doubling down on analytics innovation to keep pace. “The emergence of CTV, DOOH, and retail media presents exciting new frontiers for analytics, each with unique data characteristics and measurement challenges, as well as having a unified view across watching, browsing and buying,” Roy said.
MiQ, he explained, is responding by investing in unique data signals, refining identity resolution capabilities, and developing bespoke measurement models tailored to the specific consumption patterns of these channels.
“We’re building robust attribution frameworks that connect these new touchpoints back to unified campaign performance,” Roy added. “Our focus is on creating flexible, scalable analytical solutions that provide a holistic view of the customer journey, enabling brands to seamlessly engage across all emerging platforms.”
How MiQ balances personalization with privacy
With privacy regulations becoming both more complex and more central to advertising strategy, Roy emphasized that MiQ is fully aligned with a consent-based, ethical approach to personalization. “MiQ is deeply committed to leading a privacy-first approach to programmatic,” he said.
That commitment means using advanced contextual intelligence, privacy-enhancing technologies like differential privacy and federated learning, and focusing on anonymized, aggregated insights.
“We are building sophisticated models that derive value from signals that respect user consent, enabling highly effective audience engagement while rigorously upholding ethical data practices,” Roy explained. “As a company operating across multiple countries, we inherently maintain compliance with regional privacy regulations such as GDPR and CCPA, due to the very nature of our business.”
Anish Roy on the future of analytics in advertising
Looking to the future, Roy is excited by the convergence of two powerful forces: the evolution of AI and the maturation of causal inference methods in advertising analytics. “What excites me most about the future of analytics in advertising is the continued maturation of causal inference methodologies and the increasing application of AI,” he said. “Moving beyond correlation to truly understand what drives advertising effectiveness is transformative.”
While on the subject of emerging technologies and frameworks MiQ is particularly bullish on, he pointed to several areas of current innovation within MiQ’s analytics teams: leveraging generative AI for intelligent campaign planning, applying reinforcement learning for real-time strategy optimization, exploring whether attention or outcomes provide better measurement of ad effectiveness, and creating synthetic data to train models while preserving privacy.
“These innovations,” Roy said, “promise not just better performance, but greater transparency and strategic confidence for brands.”
Anish Roy on the real magic of AI
Despite the attention focused on generative AI in creative spaces, Roy believes that one of AI’s most powerful contributions lies under the surface. “While the spotlight often falls on the visible outputs of Gen AI like creative generation, I believe the real magic of AI lies in its power in automated feature engineering and pattern detection within vast structured and unstructured datasets,” he said.
This foundational layer of AI enables MiQ to surface subtle, high-impact predictors and detect issues or opportunities long before they become apparent. “This predictive, proactive capability is where AI truly elevates operational efficiency and strategic agility, allowing our human teams to focus on higher-order problem-solving and client strategy,” Roy explained.
When asked to sum up MiQ’s analytics advantage in a single sentence for a brand manager, Roy was clear and confident: “MiQ’s analytics superpower is our ability to transform fragmented advertising data signals into precise, actionable intelligence that drives business outcomes and sustainable competitive advantage for your brand.”
