How to choose a shop system without writing a 90-page spec

For a small project, no — and a one-page brief is genuinely better. Write what the shop must do, for whom, and what must not break. Ninety pages make everyone feel safe and tell a vendor nothing they will actually be measured against; most of it is copied from a template and never read again. Large, multi-system programmes are the exception. A first shop with a clear scope is not.

You stop reading the matrix. Every box can be ticked with enough effort, so 'supports customer-specific pricing' is true almost everywhere and tells you nothing about whether it is a checkbox or a three-week plugin. A feature matrix measures marketing, not effort. Compare on what a specific requirement costs in each system — which you only learn by making them build it in front of you.

Bring the three requirements your own people cannot fully explain — the pricing exception, the odd approval step, the article that behaves differently from every other article. Ask each vendor to demo exactly those, live, in their system. Not a polished scripted tour. Half a shortlist disappears in a single afternoon this way, and the half that survives has shown you something real instead of describing it.

Ask every vendor in writing: if we leave in three years, what exactly do we take with us, and in what format? Products, customers, orders, prices, media, the pricing logic itself. The answers differ enormously, and they tell you more about your future negotiating position than any pricing page does. A vendor who cannot answer that in writing has already answered it.