Main idea

A roadmap is often expected to settle the priority conflicts no meeting has settled. It cannot: three columns contain no criteria, no weighting, no strategy. They only state the outcome of decisions taken elsewhere, by the Product Manager or by whoever truly owns the product arbitration.

Its function is narrower and more useful. It makes those decisions visible — you can see what is committed —, debatable — you can contest a topic's placement in front of a shared object —, and communicable — a salesperson can draw a sentence from it for their customer. A roadmap is a support for arbitration, not a decision matrix.

Conflating the two produces a predictable disappointment: you fill in the format hoping it will decide, and all it does is file.

Why it matters

It guards against a common tooling reflex. Faced with confused priorities, people look for a better template, a score, an impact/effort matrix. The format will change nothing as long as nobody has decided upstream — and a format that seems to decide on its own is precisely what excuses you from deciding.

It also gives the exercise its real value: a decision already taken but visible nowhere gets renegotiated at every meeting. Writing it into a column does not make it any more correct, it makes it enforceable.

Nuances and limits

Making things visible is not neutral: a format that offers only three places exerts pressure on the upstream decision, so it acts on the arbitration even if it does not produce it.

And a support for arbitration only counts for those who read it. A roadmap filed in a tool only product and management ever open makes nothing debatable.

Open questions

• What has to be written next to a column so that a contested topic can be rediscussed without reopening the whole roadmap?