How brands can personalize ads without invading privacy

Marketers face a defining question: How to deliver personalized experiences in an environment that demands stronger privacy protections? For years, digital advertising operated on a simple premise: The more you know about someone, the better you can reach them. But that equation is being rewritten by the fading availability of third-party data and the accelerating capabilities of AI. The evolution of traditional personalization Historically, personalization has depended on identity tracking, but this model is becoming increasingly fragile. Third-party cookies are disappearing, Apple’s privacy updates have transformed the mobile landscape and consumers are no longer passive. According to a 2025 IAB study, 70% of Americans are at least somewhat familiar with U.S. data privacy laws, and among those, more than 70% believe these laws improve their data privacy, safety and control. This growing awareness is reinforced by evolving regulations like GDPR and CPRA, which continue to reshape what constitutes ethical data use. Meanwhile, AI is unlocking new capabilities that, without intentional design, risk reinforcing the same overreaches of the past. A new possibility: AI that respects privacy What if AI could enable personalization without compromising privacy? That future is already taking shape. Forward-looking brands and platforms are beginning to use AI in ways that reduce reliance on personal identifiers and instead focus on patterns, context, and declared intent. Here are some approaches: Companies are already leveraging these strategies. Instacart, for example, is using AI to make its first-party data work harder without crossing privacy lines. Its new Universal Campaigns use machine learning to adjust ad formats and placements based on performance automatically, so no personal tracking is required. AI-generated landing pages help brands deliver relevant product showcases in seconds. By focusing on cart context and declared intent, Instacart personalizes discovery experiences at the moment, not over time. Pinterest’s recommendation system offers a strong parallel to privacy-first personalization in advertising. Rather than relying on personal data, it uses contextual signals from user interactions, such as the type of content being engaged with, to make predictions. This shift has helped Pinterest maintain high relevance without building deep personal profiles, resulting in improved engagement and stronger user trust. It shows how brands can personalize at scale by focusing on behavior in the moment, not identity over time. Unilever’s beauty brands offer another strong parallel. Tools such as BeautyHub PRO and Dove’s Scalp + Hair Therapist use AI to deliver personalized product recommendations based on real-time inputs such as selfies or quiz responses. These experiences increase conversion and engagement without relying on long-term identity or behavioral tracking. Like Pinterest, Unilever shows that personalization can thrive when built on moment-based context and declared intent. These methods are not only more privacy-compliant but are often more resilient in a world with reduced data access. Personalization does not need to depend on knowing everything about someone. It can thrive by understanding the right signals at the right moment. A tool kit for privacy-first personalization To evolve, marketers need to move from thinking about data as something to acquire to something to protect and activate responsibly. That shift requires practical steps: Redefine relevance Explore opportunities beyond identity-based targeting. Invest in models that use behavioral signals, intent, and context rather than personal histories. Evaluate your AI partners Choose partners that protect user privacy by keeping personal data out of their systems entirely, whether by processing it on the user’s device, or working with anonymous patterns instead of personal profiles. Prioritize self-reported data Allow users to share their interests directly. Then, use AI to activate those preferences across channels, with transparency about how the data is used. Build cross-functional literacy Ensure teams across legal, marketing, and product understand the implications of privacy-first strategies. Personalization is no longer just a marketing function. It touches every part of the business. Track trust as a performance signal Metrics like opt-in rates, churn, and customer satisfaction are already standard, but they can also serve as early indicators of how privacy-first strategies are landing. While not always directly attributable, shifts in these numbers can reflect how much users trust the experience you’re creating. AI is not inherently invasive. It can be used to rebuild trust rather than erode it. The difference lies in how it is applied. As regulation grows and consumer expectations evolve, the brands that succeed will not be the ones with the most data. They will be the ones who design for relevance without crossing boundaries, using AI as a tool for ethical intelligence rather than indiscriminate tracking. Personalization is not going away. But what it looks like is changing. The opportunity now is to create advertising that is both intelligent and respectful, delivering value without compromise.

How brands can personalize ads without invading privacy

Featured Stories