Product Roadmap Prioritisation — by an AI council
Decide what to build next quarter and what to deliberately not build. Four rival AIs debate it and hand back: A ranked, justified roadmap with the one big bet and what gets cut.
A ranked, justified roadmap with the one big bet and what gets cut.
Signed off by an independent Final QA audit — with a “verify before you rely on this” list, never a black box.
Why this team
Roadmaps go wrong for predictable reasons: the loudest customer wins, effort estimates flatter the favourite, and nobody prices the thing that gets cut. So the Product Strategist argues coherence with the strategy, the Growth Lead argues for what compounds, and the CFO prices each row in runway. The Data Skeptic attacks the numbers the ranking silently rests on, and the CEO forces the question committees avoid: what is the one bet.
Each seat runs on a different frontier model — GPT, Claude, Gemini, Grok — so it’s a genuine cross-model review, not one AI in costumes. How a run works →
What the debate sounds like
An illustrative excerpt — A twelve-person SaaS team has one quarter and five candidates: SSO for two enterprise prospects, a mobile app, usage-based billing, AI summaries, and a performance rewrite.
The mobile app is the trap on this list. It feels like progress, eats the whole quarter, and serves users we already converted. Nothing on this board compounds except usage-based billing — that changes the revenue curve of every future customer, not just the next two.
Before billing wins the quarter: the case rests on “customers would pay more on usage” — supported by exactly two anecdotes in the notes. That claim is one pricing-page test and five customer calls away from being real. Cheap to verify, and the ranking flips if it is wrong.
The SSO row is being scored as a feature when it is actually revenue insurance — two enterprise deals with named numbers stall without it, and it is three weeks of work. Small, boring, and the only item on the board with a signed-intent figure attached.
Questions
What context does the council need to rank my roadmap well?
The candidate list with whatever data you have per item — usage, requests, revenue attached, rough effort — plus your strategy in a sentence and the team size. Missing numbers are fine: the council flags which rankings depend on unverified claims rather than pretending the data exists.
Will it just tell me to focus, like every framework does?
No — it produces a ranked list with a one-line rationale per item, one named big bet, an explicit not-now list, and the 30-day metric that tells you whether the choice was right. The debate is where it earns its keep: each persona has to defend why their priority beats the others, so the trade-offs are argued, not assumed.
We re-plan every quarter and it always slips — does this help?
The not-now list is the part built for that. Most roadmap slippage is un-killed zombie work re-entering through the side door; writing down what you are deliberately not doing, with the reason, is what makes the next re-planning meeting short.
Your material is used only to run your review — never to train public models. Encrypted in transit and at rest. Security & privacy →
Want full control — pick your own minds, set the depth? Open the full council →