Skip to content

How to shape a product strategy

Diátaxis: how-to. Shape the path from a product vision — the second product altitude. For the model, see The intent tree; for the altitude above, Frame a product vision; for how the pieces fit at engagement start, Shaping a new engagement.

You have a product vision that survived its market-existence test, and now you need the path to realize it: which problem, for which segment, in what order, and why now. That path is a product-strategy intent. Install the product-engineering pack, then work the moves below.

1. Decompose the vision, or frame the strategy directly

If you came from a vision, invoke decompose-intent on it. A surviving product-vision intent produces a product-strategy child that inherits the vision's outcome context and carries a Parent intent: back-link.

If you're entering at the strategy altitude with no framed vision above it, invoke frame-intent and answer the altitude as product-strategy. When the work really needs a vision above it, decompose-intent and frame-intent both carry a missing-parent offer — they'll propose framing the parent and hanging the strategy beneath it. It's an offer, never a block; accept it or proceed.

2. Write the strategy as challenge, policy, actions

A product-strategy intent is the same artifact as any other — Outcome, Opportunity, Assumptions, Decomposition — but the opportunity reads as a path. Name four things:

  • The central challenge — a diagnosis of the one obstacle that, left unaddressed, sinks the vision. Not a list of problems; the crux.
  • The guiding policy — the overall approach you'll take to that challenge.
  • The coherent actions — the few moves that deliver on the policy and reinforce each other rather than scatter effort.
  • The problem/segment sequence — which problem for which segment, in what order, and why now.

Keep the outcome honest: a steerable input, the lagging result it drives, a guardrail that must not get worse. Don't bolt a quantified target onto a path you've already chosen — name the input you can actually steer.

3. De-risk at the product level

Invoke de-risk-intent. A strategy bet is still market-existence in kind — will this path reach a market that pays — not feature desirability. Predeclare the kill condition before you probe, take the survive/kill verdict, and fold what you learn back into the strategy. A killed strategy reshapes the vision above it; it doesn't spawn capabilities that inherit a dead bet.

4. Decompose toward capabilities

Invoke decompose-intent on the survived strategy. It produces the next level down — capability intents, each of which re-enters the loop (frame → de-risk → decompose) until the leaf is a core brief your delivery loop can build. Don't skip the capability rung: it's where the architectural and adoption bets live, distinct from the market bet you tested up top.

Before you commit to how the capabilities get built, shape the architecture concept against this strategy — the strategy's bets are its inputs.

See also