Epismo
Get started

v4 Published snapshot

Decision to Action

Turn a consequential question into a decision and an executable plan. Start proportionally, investigate only what could change the choice, use specialist skills when the decision calls for them, then make one recommendation with a clear path to act.

Use when: there is a real choice or problem to resolve, more than one viable move exists, or the result needs a deliberate plan.

Do not use for: simple factual lookup, straightforward drafting, trivial reversible choices, or work where the requested action is already unambiguous.

Working convention: keep intermediate work in runtime-private scratch state and return a concise decision brief and action plan in the conversation. Do not create tickets, files, Cases, or Records unless the requester explicitly asks for durable collaboration or an export.

Back to playbook

Starting inputs

question *

What decision, choice, or problem do you need to move forward?

string

Steps

  1. 1

    Classify the decision and set the bar

    Start with the question. State the decision to make, the desired outcome, the decision owner, and the time or risk boundary only to the degree needed to choose a useful depth.

    Classify it as direct, analytical, or strategic. A direct, reversible choice can move quickly. An analytical choice needs targeted evidence. A strategic or product-direction choice needs explicit alternatives and assumption testing.

    Use the grilling skill only when the question remains materially ambiguous after the supplied context is read and the missing answer could change the decision. Do not turn a clear question into an interview.

    Gate: the decision to make and the proportional path to reach it are clear.

  2. 2

    Choose the decision method and decisive questions

    Identify the viable options, the criteria that distinguish them, and only the unknowns capable of changing the recommended move. Set a bounded research plan, evidence standard, and stop condition.

    For a new product idea, product direction, or question of whether something is worth building, use the office-hours skill to test the problem, user demand, current workaround, narrowest viable wedge, and key assumptions. Do not use it for an already-scoped implementation choice.

    Default to supplied context, repository evidence, and public sources. Internal connected sources, external contact, spending, publishing, permission changes, delegation, and durable Epismo state require explicit authorization.

    Gate: every investigation has a decision-changing purpose and a clear stopping point.

  3. 3

    Gather the evidence that can change the choice

    Investigate the decisive questions to the agreed depth. Prefer primary sources, official documentation, original data, and direct observation. Use independent research lanes only when questions are genuinely independent and the stakes justify it.

    For each material finding, record the source, what it establishes, its limitations, and how it affects an option. Treat unsupported claims as unknowns rather than filling gaps with confidence.

    Gate: the decision criteria have evidence, an explicit uncertainty, or a stated reason that further research is not worth the cost.

  4. 4

    Decide and pressure-test the recommendation

    Choose a provisional recommendation, then make the strongest case against it. Test the best alternative, hidden assumptions, stale or contradictory evidence, downside scenarios, reversibility, and what would make the recommendation fail.

    When choosing among multiple initiatives, options, or assumptions, use the prioritization skill to select a decision framework that fits the choice. Use its scores as a transparent input, not as a replacement for judgment.

    Use verification-before-completion when the recommendation relies on claimed completed work, a built artifact, or test evidence. It is a verification aid, not a substitute for product or strategic judgment.

    If unresolved uncertainty could reverse the decision, either obtain the missing evidence or elevate one precise decision blocker to the owner.

    Gate: the recommendation survives the best available counterargument, or its decision blocker is explicit.

  5. 5

    Make the decision clear

    Deliver one recommendation, not an unranked pros-and-cons list. State what to do, why it wins against the strongest alternative, the evidence and confidence behind it, material risks, and the exact conditions that would change the choice.

    When the evidence does not justify a final irreversible decision, recommend the smallest bounded pilot that can resolve it, with a decision rule and owner. If a consequential choice remains outside granted authority, present the prepared options and their consequences for the owner to choose.

    Gate: a decision owner can accept, modify, or reject the recommendation without reopening the full investigation.

  6. 6

    Create the executable action plan

    Translate the chosen move into a concise plan for the actual domain, whether it is product, operations, research, commercial work, policy, or software. Define the next actions, expected outcome, owner for each consequential action, dependencies, sequence, checkpoints, and the evidence that will show progress or trigger a change of course. Keep the first action small enough to start.

    Separate safe preparation from actions requiring approval. Do not send, publish, buy, sign, delete, deploy, change permissions, contact external parties, or create durable Epismo state without the required authorization.

    Return the decision brief and plan in the conversation. Create tickets, files, a Case, or a Record only when explicitly requested.

    Done means: a decision or precise blocker is clear, and the next action has an owner, boundary, and success signal.