There is a version of personalisation that converts.
An email that arrives with the product the buyer was looking at three days ago, at the moment they are ready to make the decision. A recommendation that matches what the buyer has bought before, not what the algorithm assumes they might buy based on a demographic category. A message that uses the buyer’s name and references their specific situation – because the business actually knows the situation, not because a merge field was populated.
And there is a version of personalisation that unsettles.
The ad that appears on a platform the buyer has never connected to the purchase they were browsing on a different device. The email that reveals knowledge the buyer did not consciously share. The retargeting that follows a buyer from platform to platform so persistently that it stops feeling helpful and starts feeling watched.
The marketing is the same in both cases – targeted, data-driven, designed to reach the right person at the right moment with the right message. The experience is completely different. And the experience is what determines whether the personalisation builds trust or destroys it.
What makes personalisation feel like service
Personalisation feels like service when the data being used is data the buyer gave in the context of a commercial relationship – purchase history, explicitly stated preferences, interactions with the business’s own content.
When the business uses that data to make the buyer’s experience easier, more relevant, or more useful – to surface the right product, to remember the right preference, to avoid asking the same question twice – the buyer experiences the data use as the business paying attention.
That is what converts. Not the sophistication of the targeting. The feeling that the business knows them well enough to be useful.
What makes personalisation feel like surveillance
Personalisation feels like surveillance when the data being used comes from outside the buyer’s conscious relationship with the business – third-party data, cross-device tracking, inferred attributes the buyer did not explicitly share.
Or when the business demonstrates knowledge in a way that is more specific than the context warrants – naming a concern the buyer had in a private search query, referencing a location the buyer did not consciously share, knowing something about the buyer that they had no reason to believe the business knew.
The reaction is not neutral. A buyer who feels tracked becomes a buyer who disengages, blocks, or develops a negative association with the brand that is difficult to reverse. The data that was supposed to produce a better conversion produces a worse relationship.
The practical principle
The personalisation that serves the buyer uses data the buyer gave in the context of the relationship, and uses it to make the relationship more useful.
The personalisation that unsettles the buyer uses data from outside that relationship – and uses it to demonstrate knowledge the buyer did not expect the business to have.
The former scales. The latter has a ceiling imposed by the moment the buyer notices – and once noticed, the trust cost is high.
Why this matters for how marketing data is used
Most discussions of personalisation focus on capability – what can be tracked, what can be inferred, how precisely audiences can be targeted. The capability is real and continues to expand.
The constraint is not capability. It is consent – not in the legal compliance sense, but in the experiential sense. Does the buyer’s experience of receiving this personalised message feel like service or surveillance?
The data strategy that produces the best commercial outcomes is the one that stays clearly within the service side of that line – using confirmed buyer understanding rather than inferred buyer surveillance.
Find out whether your buyer understanding is confirmed well enough to personalise without overstepping
