Decision Quality in practice.

Fictional teaching example. The company, constraints, judgments, and proposed actions below are invented to explain the method. This is not a customer case study, a product screenshot, or evidence of measured business results.

1. Frame: what choice are we actually making?

The company initially asks, “Which support provider should we hire?” That wording already assumes outsourcing is the answer. The team reframes it: “How should we provide reliable customer support over the next 12 months while retaining specialist knowledge and adapting to uncertain demand?”

The operations director owns the choice. Customer operations, finance, technology, and procurement contribute evidence. Before comparing options, the team treats customer-data handling and minimum service coverage as requirements every feasible option must meet.

Decision record: the choice, time horizon, owner, contributors, requirements, and boundaries. Changing the support model is in scope; changing the company's products is outside this decision.

2. Alternatives: what meaningful choices do we have?

• Internal team: recruit and train additional staff, keeping delivery and specialist knowledge inside the company.
• External partner: contract most routine support to a provider, retaining a small internal escalation team.
• Staged hybrid: trial a partner for a limited set of routine requests while keeping complex work internal.

These alternatives differ in commitment, control, and flexibility. The team also considers retaining the current arrangement, but excludes it from the feasible shortlist because it cannot meet the assumed minimum coverage requirement. That exclusion and its supporting evidence belong in the decision record.

3. Information: what could change our preference?

The team identifies three important uncertainties: future request volume, recruitment lead time, and the partner's ability to resolve requests without escalation. It separates measured current demand from forecasts and supplier claims.

Rather than gather every available statistic, the team asks what information could change its choice. It proposes checking recruitment availability, examining representative support requests, and testing the partner on an agreed sample before making a larger commitment.

Information plan: for each important uncertainty, name the source, owner, date needed, and decision it could affect. A trial is useful only if the knowledge it provides can justify its time, cost, and operational exposure.

4. Values: what trade-offs are we willing to make?

After checking the mandatory requirements, the team compares service continuity, specialist knowledge, flexibility, and total effort and cost. It defines what better and worse performance means for each criterion before assigning scores.

Its initial qualitative comparison is:

• Internal team: direct control and knowledge retention, but exposure to hiring delays and a larger staffing commitment.
• External partner: access to additional capacity, but reliance on onboarding, contract terms, and escalation quality.
• Staged hybrid: an opportunity to learn before expanding, but extra coordination and a longer transition.

These are hypotheses to validate, not universal facts about sourcing. If the team uses weighted scores, it must explain the scales and the trade-offs represented by the weights. A weighted total is a comparison under stated assumptions, not a probability that an option will succeed.

5. Reasoning: when would our recommendation change?

The staged hybrid initially looks attractive because demand is uncertain and a trial could test the partner's capability. The team then challenges that preference against three scenarios.

• Demand stays low: would a trial add coordination effort without enough benefit?
• Demand rises sharply: would a staged transition leave the business short of capacity?
• The partner needs frequent escalation: would the hybrid overload internal specialists?

The team proposes a limited trial only if the partner meets the data-handling requirements, internal capacity can support the trial, and the results can be assessed before a larger commitment is needed. Otherwise, it must revisit the alternatives. The recommendation is conditional because the evidence is incomplete.

6. Commitment: what happens after the decision?

The operations director owns the trial decision. Procurement confirms the proposed commercial and exit terms; technology checks data access; customer operations defines the eligible requests and escalation process.

Before starting, the team must agree measurable acceptance criteria for response time, resolution quality, escalation workload, and cost, and establish the current baseline. At an agreed review point, management can expand, revise, or stop the trial. A breach of a mandatory requirement triggers an earlier review.

Commitment record: actions, owners, deadlines, conditions, review evidence, and the person authorized to make the next choice. “Proceed with a trial” is incomplete until those responsibilities are clear.

What should management challenge?

A reviewer can use these questions before discussing the recommendation:

1. Does the frame describe the business choice, or assume a preferred answer?
2. Were credible alternatives considered and exclusions explained?
3. Which claims are supported, and which remain assumptions?
4. Are mandatory requirements separated from trade-offs?
5. What change in demand, capacity, or performance would reverse the recommendation?
6. Who owns the conditions, actions, and next decision?

How this connects to DQNavigator workshops

In DQ for Teams, participants practice developing connected evidence like this in a shared workspace. In DQ for Management, participants assess a defined package independently and discuss strengths, gaps, and differences in judgment.

The exercise illustrates the six Decision Quality requirements. A clear decision process makes the basis for action easier to examine; it does not guarantee the outcome.

Published by Project Victor Co., Ltd., developer of DQNavigator. Read about the methodology and authorship.