Now-Next-Later Roadmap

A now-next-later roadmap groups work by confidence and focus rather than assigning a quarter to every item. Janna Bastow and Simon Cast introduced the format in 2012. In ProdPad’s description, Now contains active initiatives with enough understanding to pursue a direction, including solution experiments and delivery. Next sharpens opportunities through framing and research. Later holds broader strategic bets that can be explored before they become specific initiatives. Discovery occurs in every column. Learning can increase or reduce confidence, and an item can change, pause, or disappear rather than progressing through a fixed queue.

The format addresses a common problem: a tentative feature date can be read as a promise before the work is understood. Marty Cagan distinguishes strategic direction from credible commitments supported by discovery and scheduling. ProdPad allows meaningful external deadlines, internal strategic commitments, and milestones alongside confidence horizons. The format does not forbid dates, make all OKRs guaranteed results, or prevent a dated plan from incorporating learning. Record what each date means and who has accepted the associated commitment.

What goes in a column

Entries describe problems and intended outcomes, with candidate solutions underneath. A fictional Now entry might concern eligible data science teams waiting three weeks for a GPU environment, with a proposed target below a day without troubleshooting support. Define request and readiness events, required approvals, success criteria, and the observation window; use active-team coverage and cohort retention to understand use. A Next entry identifies the questions needed to choose a direction. A Later entry may describe a broad opportunity and exploratory research. Keep the level of detail useful for the audience and the active work within available capacity; the number of pages is not a test of whether prioritization occurred.

How it turns back into a timeline

The headings alone do not change a planning practice. Treating them as fixed quarters can conceal uncertainty; listing only features can obscure the problem each is meant to solve. Excessive work in Now can exceed capacity. Dates with no established rationale can create accidental promises, but a strategic date does not need an external origin to be legitimate. Connect genuine milestones to relevant initiatives, recording forecasts, confidence, and commitments separately from any obligation. If a prototype fails, the team might test another solution in Now, reconsider the problem in Next, broaden exploration in Later, or stop. Choose from what the evidence changed, not from a rule that failure always moves a card backward.

How the format fits with measuring what a platform delivers, the rules under which feedback moves an item between columns, and how the few genuine commitments are kept separate, is covered in Twenty-Three of Twenty-Five Shipped. Which problems deserve a place in the columns at all is the subject of The Vote Always Went to the Forty.

References:

• Janna Bastow — Now-Next-Later Roadmap, ProdPad glossary, updated 29 June 2026
• Marty Cagan — Product Roadmaps, SVPG, 7 January 2009
• Marty Cagan — The Alternative to Roadmaps, SVPG, 7 September 2015