# Decision-Making Framework

*This framework was generated using the Framework Builder methodology. It demonstrates the five-layer DNA architecture applied to decision-making for teams and organizations.*

Page: https://whereframeworks.com/frameworks/decision-making
Guide: https://howtoframework.com/decision-making-framework

---

## Layer 1: Principles Foundation

**Principle 1: Principles must resolve conflicts, or they are decoration.** "Quality and speed both matter" decides nothing. "Quality wins on client-facing work, speed wins on internal iteration" decides a hundred things. Write every principle as a tiebreaker.

**Principle 2: Decisions are not equal, so the process is not equal.** A meeting time and a market entry need different rigor. A framework that treats them the same will be abandoned within a month.

**Principle 3: Judge options against the requirement, never against each other.** Comparing options lets the first strong one become the benchmark for everything after it.

**Principle 4: Demonstrated evidence outweighs self-reported claims.** Weight what you can see someone do or a number you can check over what they say about themselves.

**Principle 5: Say who decides, not only who has input.** Some good decisions are opposed by a minority. Name the owner before the discussion starts.

## Layer 2: Systematic Approach

**Step 1: Classify the decision.** Rate it on reversibility, cost of being wrong and who it affects. Reversible and cheap decisions get a fast path. Costly and hard-to-undo decisions get the full path.

**Step 2: Route to a process.** Write one path per class. Example for a hiring decision: individual roles get a structured scorecard with two evaluators and a 48-hour window. Management roles add a stakeholder panel and a scenario exercise. Executive roles add reference deep-dives and a final review.

**Step 3: Gather the required inputs before the meeting.** Each path lists what must be in hand. No decision meeting is scheduled until it is.

**Step 4: Collect written judgments first.** Everyone submits a written assessment before any discussion, so the loudest voice cannot anchor the room.

**Step 5: Run the disagreement step, then decide.** One person argues the case against the leading option. The named owner decides and records the reason in one paragraph.

## Layer 3: Force Multipliers

**Multiplier 1: Assign a standing dissenter.** One person is tasked to find reasons not to proceed. This catches the enthusiasm that sets in when a team badly needs a yes.

**Multiplier 2: Write the evaluation before the discussion.** It costs ten minutes and removes most anchoring.

**Multiplier 3: Keep a one-page decision card.** The routing rules and scorecard dimensions on a single page, so nobody opens the full framework for each decision.

**Multiplier 4: Design for the hard cases first.** Test the framework on conflicting data, time pressure and missing information. If it holds there, easy cases take care of themselves.

## Layer 4: Success Metrics

**Positive indicators:**
- Outcomes match what the scorecard predicted (calibration holds)
- Decision time falls within the target window for each class
- The process was followed and the required inputs were present, on most decisions
- People can state the reason behind a past decision without looking it up

**Failure signals:**
- You track results only, so good judgment cannot be told from good luck
- Every decision is routed to the full path, or none are
- Dissent is skipped when the team is in a hurry
- The same type of decision is re-opened after it was made
- Outcomes drift from predictions and nobody revises the scorecard

## Layer 5: Implementation Guidance

**Start with one decision type.** Pick the one you make most often, such as hiring. Apply the framework to the next three instances on one team, with a coach who sits in and checks that the steps were followed. Then widen.

**Review on a schedule.** If the outcome measure for a decision type drops below your floor, revise the framework before blaming the people using it.

**Common mistake: building a consensus tool.** If the framework describes how to get everyone to agree, it is not a decision framework. Add the owner and the tiebreakers.

**Edge case: time pressure.** Use the fast path, write down what you skipped and why, and review the decision at a set date.

**Edge case: two owners disagree.** Escalate one level with both written assessments attached. Do not merge the two positions into a compromise nobody wanted.

**Edge case: new information after the decision.** Reopen only if it would have changed the outcome under the original criteria. Otherwise log it and move on.

**Common mistake: measuring outcomes without measuring process.** If you track only results, you cannot tell good judgment from good luck. Log whether the framework was followed and whether the right inputs were gathered. Those are the leading indicators of decision quality over time.

**Common mistake: designing for the average case.** Most frameworks work on straightforward decisions and fail when ambiguity arrives. Write the hard cases into the framework first: conflicting data, time pressure, incomplete information and stakeholder disagreement.

**Common mistake: skipping the disagreement step.** Teams that skip it decide faster and measurably worse. The discomfort of structured disagreement is the price of better outcomes, so keep the step even when the room is confident.
