Why your customers don't come back, and how to find out

There are two ways to lose a customer. The first makes a noise: a complaint, a review, an argument at the counter. It is unpleasant, but it has one enormous merit — you know it happened, and you can respond.
The second makes no noise at all. The customer comes three times, then less often, then not at all. Nothing went wrong. They have no grievance. They simply stopped thinking about you — and that is by far the more common of the two.
The real reasons, in order of frequency
Shopkeepers almost always overestimate price and product. The real causes are more mundane.
Being forgotten
The leading cause, and the one nobody talks about. Your customer liked you. They meant to come back. Then their route changed, a competitor opened closer to their office, or simply six weeks went by. You did nothing wrong — you just stopped existing in their day.
It is the only cause you can fix without changing anything about your shop: you just have to reappear at the right moment.
A broken habit
A regular customer holds on to a ritual: the same coffee at the same hour. A week of holiday, a house move, a change of shift is enough, and the ritual reforms somewhere else. A habit takes weeks to build and one to lose.
A tiny disappointment
A dish below the usual standard, too long a wait, a distracted welcome on a busy day. Too little to complain about, enough not to come back. This is the most unfair category: it punishes one bad day severely, and you will never hear about it.
Price, last
It matters, obviously. But it comes after the other three, and above all it is rarely decisive on its own: people leave an expensive shop they are fond of far more slowly than a decent one they no longer think about.
The real problem: you cannot see it
A full shop gives the impression that all is well. That is misleading, because a full room does not tell you who is filling it. Thirty customers a day can be thirty regulars or thirty different strangers — the room looks exactly the same, and the future of the business is not remotely the same.
Without a way of recognising people from one visit to the next, you cannot tell the difference. That is precisely the blind spot a loyalty programme fills — long before the reward it hands out.
Three measurements you can take
None of them needs a sophisticated tool. All of them need counting.
1. The return rate
Of a hundred customers who came for the first time this month, how many came back at least once in the following two months? It is the most important number in your business, and almost nobody knows it.
2. The alert window
How long normally passes between two visits from a regular? A week for a café, six for a hairdresser. Double that: it is your alert threshold. Beyond it, the customer is on their way out, and it is the last moment a word can still do something.
3. The share of regulars
What proportion of your takings comes from customers who have been in before? The higher that share, the sounder your business — and the more a lost customer costs.
What works to bring them back
- Reappear before the alert threshold, not after. A month after someone drops off, it is almost always too late.
- Give a reason, not a discount. «Your reward is waiting» works better than «10% off this week»: one is something earned and owed, the other is one promotion among many.
- One message only. A reminder without a reply is a reply. Pushing turns a distracted customer into an irritated one.
- Acknowledge the return. A word at the counter — «it's been a while» — is worth more than any automatic message.
Where to start
Do not try to measure everything. Take the one figure that counts: how many of your customers come back. If you cannot answer that today, that is where the work is — before the card, before the reward, before the channel.
Once you can measure it, the rest becomes simple: you will finally see the effect of what you change. The guide on setting up a programme covers the decisions that come next — the mechanics, the threshold and the reward.

