For most of the last half century, repair was the part of the business you tried to make disappear. It was labour-intensive, hard to standardise, and it competed directly with the thing you actually wanted to sell. The rational move was to price a repair just high enough that a customer would buy new instead, and then to design the product so the question rarely came up.

That logic is now failing in several places at once, and not for the reasons its critics predicted. Regulation has played a part. So has the cost of raw materials. But the decisive change has been quieter: repair turns out to be one of the few remaining ways to hold a relationship with a customer over a decade rather than a quarter.

The margin nobody was counting

Consider the arithmetic from the manufacturer's side. A first sale is expensive — marketing, distribution, discounting, returns. A repair, by contrast, arrives with the customer already acquired, already committed and already in possession of the product. The acquisition cost is zero. The conversation is warm. And the customer who has just had something mended is, by a considerable margin, the likeliest person to buy the next one.

We stopped treating the service desk as a cost centre the day we worked out it was the highest-margin square metre in the building.

A retail operations director

The businesses succeeding at this share a structural quirk: they sell products expensive enough to be worth mending. Below a certain price point the economics still collapse, because the labour cost of a skilled hour will not shrink to meet a low-cost import. This is the uncomfortable centre of the repair argument. It works beautifully at the top of the market and barely at all at the bottom, which is precisely where the volume — and the waste — sits.

Designing for the second owner

What is genuinely new is a small number of firms designing for resale from the outset. If a product is expected to be sold twice, the calculus changes: fasteners instead of adhesive, published parts lists, serial numbers that survive a refurbishment. The first buyer benefits from a higher residual value; the manufacturer captures a slice of the second transaction rather than watching it happen on a marketplace it does not control.

  1. Design the disassembly sequence before the assembly line, not after it.
  2. Publish the parts catalogue — the independent repairer is a distribution channel, not a competitor.
  3. Price the repair against the cost of losing the customer, not against the cost of the part.

None of this amounts to a movement yet. It is a handful of companies, mostly mid-sized, mostly privately held — which is to say, mostly able to take a view longer than the next reporting period. But the direction of travel is clear enough that the sceptical question has changed. It is no longer whether repair can be made to pay. It is which category it reaches next.