Personalisation now means the individual, not the segment

Look at the last email your favourite brand sent you. It probably arrived in a batch of a hundred thousand, wore your first name in the subject line, and offered you the exact thing you bought a fortnight ago.

That is a segment doing its best. And its best is no longer enough.

McKinsey found that 71% of consumers now expect a company to treat them as an individual, and 76% get frustrated when it doesn't happen. Personalisation used to be a bonus you could earn goodwill with. It has quietly become the baseline people measure you against, and most brands are still answering it with a segment.

Here is the version that stings. Somewhere in your list is a customer whose second jar of protein runs out this Friday. She sits in the same segment as a man who bought once in March and never came back. On Tuesday they both get the same email, the same 15% off, the same hero image. For her the offer is late and slightly insulting. For him it's wasted. The segment can't tell them apart, because a segment was never built to.

The segment was always a compromise. We grouped people because we couldn't afford to look at them one at a time. Reading every customer, remembering what each of them did, working out the one useful next thing for each, that was too much to do by hand. So we drew boxes. Women, 25 to 34, bought skincare, opened in the last 90 days. Good enough, because good enough was all there was.

That constraint is the thing that's dying, not personalisation. The unit of marketing is shrinking from the group to the person, and it's worth understanding before your competitors get there first.

What is a segment of one in personalisation?

A segment of one is personalisation aimed at a single person rather than a group they happen to resemble. Instead of "women, 25 to 34, skincare buyers," it's "this person, second bottle due Friday, reads on Sunday nights, ignored the last three sends." The segment describes a type of customer. A segment of one describes a customer.

A segment tells you what someone is like. A segment of one tells you what to do next.

The commercial case is not soft. The same McKinsey research found that companies which grow faster drive 40% more of their revenue from personalisation than the ones that don't, and that getting it right lifts revenue by 5 to 15%. That gap isn't a rounding error. It's the difference between a brand that speaks to people and one that broadcasts at them.

Why do customer segments stop working?

Segments stop working because inside any box, people are on different days. Two customers can match every filter you own and still need opposite messages, because one is about to reorder and the other is about to leave. The attributes are identical. The moment is the exact reverse.

The proof shows up in the sends. Acoustic's 2026 benchmark, drawn from 12.2 million mailings across 519 senders, found that emails triggered by a person's own behaviour earned 2.2 times the clicks of scheduled batch sends, 2.36% against 1.07%. Same brands. Same lists. The only real difference is that a triggered message arrives on the customer's day and a batch arrives on yours.

A segment freezes people at the moment you built it. A person keeps moving. By the time your Tuesday send goes out, half the segment has changed their mind and you never noticed.

How do you personalise for one customer at a time?

Start with moments, not attributes. Attributes describe who someone is. Moments describe where they are right now, and the moment is what tells you what to say. Pick the handful that actually matter for your product: the first order, the refill window, the browse that didn't become a purchase, the quiet lapse before someone churns. Building flows around a customer's own lifecycle rather than your campaign calendar is most of the work.

Then, for one real customer, answer three plain questions. Where are they in their own story with you. What is the single most useful thing to say to them next. When do they actually open and read. You already hold the data to answer all three. It's sitting in your platform, unused, because the segment never asked.

You can do this on Monday without buying anything. Take one moment, the refill window. Work out each customer's likely reorder date from their own purchase history, not from a calendar. Write one honest message for the person about to run out, not for a demographic. Trigger it off their date. That single flow, built for the person, will usually beat the newsletter you spent all week designing. If you send through Klaviyo, this is a flow you can build today with data you already have.

Does personalising for the individual break at scale?

No, and that's the part that has genuinely changed. Treating a hundred thousand people as individuals was impossible by hand, so we didn't. Now a marketer sets the logic once, per moment, and it runs for each person from their own data. The repetition is carried for you. The judgement, the taste, the decision about what's worth saying, stays with the human who understands the brand.

That's the real shift behind the numbers. A team of three can now do what used to need a room of fifty, because the work that scaled badly, doing the same thoughtful thing for every single customer, is the work software is finally good at. The marketer stops writing one email for everyone and starts setting the rules that give everyone their own.

Two customers can match every filter you own and still need opposite messages.

How do you know the personalisation actually worked?

Hold a slice back. Take a random share of the customers who would get the individual treatment and keep sending them the old batch instead. That group is your control group. Compare what the two earn per customer over the next 90 days, and you have an answer you can trust.

This is the step most teams skip, and it's the only one that tells the truth. Without a control group you'll credit the personal approach for sales that would have happened anyway, and quietly fool yourself. With one, you find out whether treating people as individuals actually moved money or just felt more modern. It's the same discipline we build on at PilotX, where four agents work every customer on their own timeline against a control group you set, so the lift is modelled first and checked against a real baseline, never assumed.

You don't need any of that to start. If you want to see where your own calendar is already leaving money in the gaps between customers, the free Revenue Leak Audit walks your funnel and shows you the moments no one is speaking to. Pick one of them. Build the flow for the person, not the segment, and let next week tell you the rest.