Prototype Decision Radar

Prototype Decision Radar — Quick Decision Aid

Use this tool to quickly choose a prototype approach that gives you the learning you need without overbuilding. The radar maps four learning objectives against common constraints so teams can align on what to build, how fast, and what to measure.

When to use

• You need to validate a specific assumption (value, usability, performance, or integration) before committing engineering or operational effort.
• Stakeholders disagree on fidelity, scope, or acceptable risk for a prototype.
• You want a repeatable way to choose prototype type and success criteria for experiments and MVPs.

Core dimensions

• Learning objective — What do we need to learn? (Concept, Usability, Performance, Integration)
• Time — How quickly do we need results? (hours, days, weeks)
• Cost — How much are we willing to spend in effort or cash?
• Fidelity — How closely must the prototype resemble the final experience?
• Risk — Safety, regulatory, brand, operational or customer risk if prototype fails or is seen broadly.

Decision matrix (compact)

Example success criteria (make these measurable)

• Concept: 200 signed interest forms in two weeks; 25% click-to-signup on the landing page.
• Usability: 80% task completion in moderated test; median task time under X minutes.
• Performance: 95th percentile response time under 500ms at 1,000 concurrent users in benchmark.
• Integration: End-to-end workflow completes without manual intervention in 95% of runs during a pilot week.

Printable Decision Radar template (how to use)

1. Draw a radar with five axes (Learning Objective, Time, Cost, Fidelity, Risk). Alternatively, use the attached grid to record scores.
2. For each candidate prototype, score each axis from 1–5 (1 = minimal, 5 = maximal). Higher Risk means more caution required.
3. Plot scores for 2–4 prototype options on the same radar. The option that best matches your prioritized objectives (for example, high on Learning Objective and low on Cost) is usually the best first experiment.
4. Capture a short hypothesis, primary metric, success threshold, and intended audience for the prototype.

Quick facilitation script (10–20 minutes)

1. Frame the hunger: “What do we need to learn by the end of this prototype?” Ask participants to pick one primary learning objective.
2. List candidate prototypes (keep to 3–4). Example: landing page, clickable mock, Wizard-of-Oz, partial integration pilot.
3. For each candidate, quickly agree scores for Time, Cost, Fidelity, Risk (1–5). One facilitator records; avoid debates longer than 90 seconds per cell.
4. Plot results and discuss alignment: which option delivers the prioritized learning with acceptable cost/time/risk trade-offs?
5. Agree on a single prototype to run, one lead owner, one primary metric, and a deadline for the experiment.

Common mistakes to avoid

• Confusing internal neatness for external value—validate with customers or representative users.
• Overbuilding fidelity before validating the riskiest assumption.
• Using vanity metrics (e.g., traffic without conversion) as the primary success criteria.
• Neglecting labeling and communication — prototypes that could be mistaken for production need clear disclaimers and retirement plans.

Sample scenarios (very short)

Scenario A (New feature idea): Landing page (Concept) — score: Learning 5, Time 5, Cost 5, Fidelity 1, Risk 1. Action: Run a paid acquisition test with sign-up form and measure conversion.

Scenario B (Workflow integration): Staged pilot (Integration) — score: Learning 5, Time 2, Cost 2, Fidelity 5, Risk 4. Action: Run in a single customer site with rollback plan and monitoring.

Next steps & templates

• Copy this template into your experiment tracker and attach the prototype plan: hypothesis, metric, success threshold, owner, and end date.
• Consider turning this into an interactive picker that records choices and stores them with the experiment record.

Notes for operators

Label prototypes clearly and document how they will be retired or hardened if validated. For high-risk prototypes, include safety, legal, and compliance checks before exposing real customers or critical systems.