Roadmapping — Examples
Example 1 — A roadmap that is a feature list with dates
A product team publishes the roadmap below. Rewrite it so it commits to outcomes and forecasts the work, then sort it into now, next and later.
Q2 Returns portal shipped 12 June
Q2 Faster checkout shipped 30 June
Q3 Multi-currency 15 August
Q3 Dashboard redesign 30 September
Q4 Mobile app December
Solution
Committed cut order-to-shipment from 4 days to 2 by the end of Q3.
Now automate the packing slip; remove the manual stock check.
Next same-day courier option, if the volume supports it.
Later fewer support contacts per hundred orders (theme).
Not in it multi-currency, dashboard redesign, mobile app.
The committed line names a change the business can hold the team to, while the work underneath is a forecast that can move; the feature list promised five deliveries on five dates written when the least was known. Sorting into now, next and later keeps dates only where they can be kept, and the explicit exclusion line stops stakeholders reading absence as an oversight. The thing that usually goes wrong is an outcome so vague it commits to nothing — a theme must come with a measurable result.
Example 2 — Themes that name a feature
The roadmap’s theme column contains feature names, so the roadmap dies whenever the first idea is wrong. Rewrite each as a theme that answers a question, then add the line that says what the document is not.
Themes
returns portal
pricing page rewrite
admin dashboard v2
Solution
Themes (each with a measurable outcome)
reduce the cost of returns
help buyers choose a plan without sales contact
let support change an account without an engineer
This roadmap is not a contract, not a delivery schedule and not a
prioritisation list. It is revised quarterly with evidence.
A theme survives its first idea being wrong — “reduce the cost of returns” can be met by a portal, a policy or a courier change, while “returns portal” is spent the moment it is chosen. Saying what the roadmap is not on the document itself removes the three readings that cause the most friction, and the fixed quarterly cadence keeps revision from being driven by whoever escalates last. The trap is leaving the disclaimer out because the team already knows; the reader who treats it as a contract is the reader who never heard.