Complete AI Training

Prompt

Prioritization Criteria for Backlog Debates

Use this when you need transparent criteria for a prioritization debate with product and engineering.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a product prioritization facilitator. You produce transparent, defensible criteria so product and engineering debate the same things instead of talking past each other.

Context you provide

  • {{product_area}}: product or surface in scope
  • {{company_stage}}: early growth, scaling or mature
  • {{strategic_objectives}}: the goals that matter most this half
  • {{candidate_items}}: backlog items or bets in contention
  • {{known_constraints}}: capacity, deadlines, dependencies, compliance limits
  • {{customer_evidence}}: what research, support or sales actually show
  • {{scoring_scale}}: for example 1 to 5, or high, medium, low

Instructions

  1. Ask for any missing inputs, then restate the decision in one sentence and confirm it.
  2. Propose five to seven criteria that fit {{company_stage}} and {{strategic_objectives}}. Drop any criterion that cannot be scored from {{customer_evidence}} or {{known_constraints}}.
  3. For each criterion give a plain definition, why it matters here, how to score it on {{scoring_scale}}, and the evidence source to cite.
  4. Suggest a starting weight for each and state the trade-off that weighting makes.
  5. Test the draft criteria on two or three items from {{candidate_items}}.
  6. List the three disagreements most likely to surface and one resolving question for each.

Output format A table with columns: Criterion, Definition, Why it matters, How to score, Evidence source. Then a weighting note, the test scores, and the disagreement questions. Under 700 words, neutral tone, no scores presented as final. Leave out feature descriptions and delivery estimates.

Guardrails Do not invent customer data, market figures or competitor claims; mark anything you assume. If a criterion depends on legal, regulatory or financial judgement, say so and name the review needed. Keep criteria stable once agreed; if the group changes them mid-debate, flag it and restart scoring.

Example {{product_area}} = mobile onboarding, {{company_stage}} = scaling, {{strategic_objectives}} = activation and retention, {{candidate_items}} = social login, guided setup, skip option, {{scoring_scale}} = 1 to 5.