This article was based on the interview with Phyllis Fang, Head of Marketing at Transcend by Greg Kihlström, AI and MarTech keynote speaker for The Agile Brand with Greg Kihlström podcast. Listen to the original episode here:
It’s the central paradox of our discipline. As marketing leaders, we are tasked with delivering ever more precise, one-to-one personalization, a mandate supercharged by the capabilities of AI. The promise is a world where every interaction is relevant, every offer is timely, and every customer feels uniquely understood. Yet, at the exact same moment, the fuel for this engine—customer data—is becoming more restricted. A confluence of privacy regulations, the slow death of third-party cookies, and a healthy dose of consumer skepticism has put a fortress around the very information we need to deliver on that promise. The old playbook of broad segmentation and demographic targeting, already showing its age, is now verging on obsolete in a world where shared cultural moments are fleeting and AI can commoditize generic content in an instant.
This isn’t a problem to be delegated to the legal department or solved with a simple banner on a website. It is a core strategic challenge that sits squarely at the intersection of marketing, technology, and brand trust. The way forward requires us to re-architect not just our tech stacks, but our entire philosophy on data. The most successful brands will be those that stop viewing privacy as a compliance hurdle and start treating it as a foundational element of the customer experience—a competitive advantage that fosters the trust necessary for a meaningful value exchange. We recently explored this complex terrain with Phyllis Fang, Head of Marketing at Transcend, who brings a sharp perspective from her time at e-commerce pioneers and growth behemoths like Uber. Her insights reveal a path through the paradox, one that balances ambition with responsibility.
The Hidden Cost of “Shadow Data”
Before we can build the future of personalization, we must first address the ghosts in our current machine. Many organizations operate with vast, unseen repositories of what is often called “shadow data.” This isn’t nefarious by nature, but it represents a significant and often unacknowledged risk to both compliance and customer trust. It’s the data collected for one purpose that now lives in a siloed analytics platform, the customer list uploaded by a well-meaning regional manager, or the information lingering in a dozen disconnected MarTech tools. For marketing leaders, this ungoverned data is a minefield that directly undermines the promise of true personalization.
Phyllis Fang explains the direct impact of this phenomenon on a brand’s ability to build a transparent relationship with its customers.
“All that shadow data floating throughout an organization’s tech stack…creates unseen risks that erodes transparency. And as a marketer, as an operator, you don’t always know what data is permissible to use…customers are smart. They know when personalization is shallow and just like a dynamic field substitution, but they also can instinctively sense when brands overreach or they mishandle data and that leads to far graver consequences. Opt-outs, people just churning off of your brand or just completely blocking ads. And that just kills personalization right out the source.”
Fang’s point is critical. Consumers have a finely tuned sense for when a brand knows something it shouldn’t. It’s that uncanny valley of personalization where an ad feels a little too specific or an email references a piece of information the user doesn’t remember sharing. This is where trust evaporates. The first step, then, is not to acquire more data, but to gain visibility and control over the data you already possess. Auditing and mapping your data touchpoints isn’t just an IT exercise; it’s a marketing prerequisite for building a trustworthy foundation upon which any sophisticated personalization strategy can be built. Without it, you’re building on sand, and the tide of consumer sentiment is rising quickly.
The Value Exchange: Moving Beyond Vague Promises
For decades, the implicit value exchange for data was simple: give us your information and we’ll give you a slightly better, or at least free, experience. That bargain is no longer compelling to a savvy consumer. The promise of “more relevant ads” or a “personalized journey” has become table stakes, and frankly, is often delivered so poorly that it inspires more cynicism than loyalty. To convince a modern consumer to willingly share their data, the value offered in return must be tangible, immediate, and clearly linked to the information being requested.
The brands that excel in this new environment are those that have mastered the art of providing genuine utility. Fang highlights several examples that move beyond the kind of personalization that amounts to little more than a mail merge from 1998.
“Personalized loyalty or rewards that visibly reflect those data choices. So I think a lot of retailers do this well. Sephora’s rewards program is a great example…A lot of financial services apps will do this with credit insights or budgeting tips in exchange for seeing a user’s transaction data…And I think we can’t forget real utility by way of convenience, by way of time savings, continuity. Cross device experiences, when a user is opted in that persists across different devices or browsers, wishlist, transactions, that kind of stuff. That’s real utility and it’s not just like being marketed to in a better way.”
The key takeaway here is the shift from passive personalization (being shown a “better” ad) to active utility (the app helping you budget, the website remembering your wishlist across devices). When a customer gives Sephora their skin tone, they see visibly different and more helpful product recommendations. When they allow a fintech app to see their transactions, they get actionable advice that can save them money. This is the new contract. As marketing leaders, we must constantly ask: what tangible utility are we providing in exchange for this piece of data? If the only answer is a slightly more targeted campaign, we haven’t gone nearly far enough.
From Risk Mitigation to Growth Engine
The conversation around data privacy in most boardrooms is still dominated by the language of risk: avoiding fines, preventing breaches, and mitigating brand backlash. While these are all valid and necessary concerns, this defensive posture is also incredibly limiting. It frames privacy as a cost center and a constraint on innovation. The more forward-thinking approach, and the one that will separate the leaders from the laggards, is to reframe privacy as a strategic enabler of growth. A robust, transparent, and user-centric data governance framework doesn’t shrink your marketable audience; it expands the pool of high-intent, trusting consumers you can engage with effectively.
Fang introduces a powerful framework for this shift in mindset, contrasting the common practice of “under-permissioning” with the potential of a well-architected data strategy.
“The way I would tell marketers to start to think about this reframe is that we need to be able to show that trustworthy data pipeline is expandable usable data. So that expands your overall audience size and under permissioning is actually leading to a huge opportunity costs. You’re not able to market to an additional audience or perhaps you need to actually just do more generalized marketing to everyone, even if parts of your audience are irrelevant. That’s wasted ad spend…I would really think about it as opportunity costs and being able to like how large is your available audience to market to.”
This is the language that resonates in the C-suite. By being overly conservative—by choosing not to engage with certain segments or leverage certain data points out of an abundance of caution—we are leaving growth on the table. A modern data privacy infrastructure isn’t about locking data down; it’s about enabling its flow with surgical precision based on real-time user consent. This allows marketers to confidently activate audiences they might otherwise have avoided, to personalize with precision rather than broad strokes, and to do so in a way that increases LTV by building, rather than eroding, customer trust. The opportunity cost of not investing in this capability is immense, manifesting as wasted media spend, irrelevant messaging, and a ceiling on growth.
The New Collaborative Mandate
The era of siloed expertise is over. Successfully navigating the personalization paradox is not a task the CMO can accomplish alone. It requires an unprecedented level of collaboration across the C-suite, particularly with the CIO and CISO. The marketing leader of today and tomorrow must be as conversant in the principles of data architecture and security protocols as they are in brand strategy and campaign analytics. The success of every AI-powered initiative, every retail media network, and every seamless cross-channel journey depends on a shared understanding and a unified technical and ethical foundation. This is the new operating model for enterprise marketing.
Ultimately, the future will likely not be one of the extremes often predicted—either a world of runaway AI personalization or one where consumers lock their data down so tightly that marketing becomes impossible. Instead, we are moving toward a more nuanced, dual reality. In this world, AI will indeed power incredibly sophisticated and automated personalization, but it will operate within a framework where consumers hold the keys. They will demand, and receive, granular control over how their data is used, including the right to opt out of having their data used to train AI models. The brands that thrive will be those that not only comply with these demands but embrace them, building systems that honor user choice in real-time and turning that respect into a cornerstone of their brand identity. It’s a complex challenge, to be sure, but for those of us genuinely excited by the potential of technology to create better connections, it’s also the most interesting problem to solve.
