There was a time when personalisation in advertising was little more than a clever insert—your name in a subject line, a city-based offer, a mildly tailored banner. Today, it’s something far more sophisticated and far more loaded. AI has given brands the ability to read patterns, anticipate intent, and adapt messaging in ways that feel almost intuitive. But with that capability comes a quiet tension. The more precisely a brand can speak to an individual, the more it risks sounding like it knows too much. And in a market like India, where digital adoption is surging but trust is still being negotiated, that tension is shaping how personalisation is actually being deployed.
What’s interesting is that the smartest marketers are not chasing hyper-personalisation as aggressively as the technology allows. If anything, there’s a visible shift toward restraint. AI is being used less as a spotlight and more as a dimmer switch—adjusting tone, timing, and context rather than calling attention to itself. The focus has moved from “individual-level targeting” to “moment-level relevance.” A user browsing insurance late at night might see a different narrative than someone casually exploring options over the weekend—not because the brand knows who they are, but because it understands the situation they’re in. It’s a more observational approach, and it feels less intrusive because it doesn’t try too hard to prove its intelligence.
That shift is also being driven by a growing awareness of what crosses the line. The “creepy” factor rarely comes from personalisation itself; it comes from the way it is expressed. Consumers are far more comfortable with inferred relevance than with explicit references. There’s a difference between an ad that feels timely and one that feels like it’s been watching you. Brands that are getting this right are building internal guardrails—not just around data usage, but around tone and creative expression. They’re asking simple but important questions before campaigns go live: Would this feel natural if it appeared unexpectedly? Does it rely on information the user knowingly shared? Is the value exchange clear? Because once communication starts to feel like surveillance, no amount of creative polish can fix it.
There’s also a practical reality at play. At scale, true one-to-one personalisation is not just complex—it’s often unnecessary. AI is proving far more effective when it works with clusters, behaviours, and context rather than individuals. Cohorts built around intent signals, content consumption patterns, or even time-of-day behaviours are allowing brands to stay relevant without becoming invasive. This is particularly evident in sectors like fintech, mobility, and e-commerce, where the difference between helpful and intrusive can directly impact conversion. The approach is quieter, but often more effective. As one marketer put it, “Personalisation works best when it feels like good timing, not good stalking.”
Creatively, this has led to a different kind of thinking. Instead of building one perfect message, brands are designing flexible systems—campaigns that can stretch and adapt without losing their core idea. AI helps assemble the pieces, but the thinking still starts with a strong, human insight. A campaign might shift language, visuals, or even emotional tone depending on context, but it doesn’t abandon its central narrative. This balance is important. Over-optimisation can make communication feel fragmented, even soulless. What’s working instead is a more fluid approach, where variation exists within a clearly defined brand voice.
There’s another layer to this conversation that often goes unspoken: regulation is catching up, but consumer expectations are moving even faster. Data protection frameworks are forcing brands to be more transparent and disciplined, but audiences are already developing their own filters. They reward relevance, but they are quick to disengage when something feels off. In that sense, the “creepy line” is not fixed—it shifts with culture, context, and category. What feels acceptable in a streaming platform may feel intrusive in financial services. The margin for error varies, but the principle remains the same: respect the user’s sense of space.
Perhaps the bigger takeaway is that AI has not changed the fundamentals of advertising as much as it has exposed them. Good marketing has always been about understanding people without overwhelming them, about being present without being intrusive. AI simply magnifies both ends of that spectrum. Used thoughtfully, it can make communication feel sharper, more relevant, and more timely. Used carelessly, it can make brands feel mechanical and overly familiar in ways that unsettle rather than engage.
Looking ahead, the brands that will stand out are not the ones pushing the limits of personalisation, but the ones showing discipline in how they apply it. There’s a certain confidence in holding back, in choosing not to use every signal available. Because ultimately, personalisation is not about proving how much data you have—it’s about demonstrating how well you understand context. And sometimes, the most effective message is the one that knows when to say less.
