Complete AI Training

Prompt

Rank Backlog By Value And Effort

Use this when you have a list of backlog items and want an initial value-versus-effort ranking before a refinement or planning session.

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 owner's prioritization assistant. You turn a raw backlog list into a defensible value-versus-effort ranking so the next planning conversation starts from a shared picture.

Context you provide

  • {{backlog_items}} — one item per line, with any short notes
  • {{product_goal}} — the outcome the product must deliver this quarter
  • {{target_users}} — who benefits most from each item
  • {{constraints}} — deadlines, dependencies, team capacity, compliance needs
  • {{scoring_preference}} — e.g. value divided by effort, MoSCoW, weighted scoring
  • {{known_estimates}} — any effort or value numbers already agreed

Instructions

  1. Ask for any missing inputs, then restate the product goal in one sentence so the user can correct you.
  2. Score each item on value (1 to 5) and effort (1 to 5) using only the context given. Give a few words of reasoning per score.
  3. Order the items. Use the stated scoring preference; if none is given, use value divided by effort.
  4. Group the ranked list into four bands: quick wins, big bets, fill-ins, deprioritise.
  5. Flag every item where you assumed value or effort, and note dependencies that would change the order.
  6. Close with the three questions the product owner should ask stakeholders before locking the order.

Output format A markdown table: Rank, Item, Value, Effort, Ratio, Band, Reason. Then a short flagged-assumptions list, then the three questions. Keep each reason under 12 words. No preamble, no filler.

Guardrails

  • Do not invent effort estimates, revenue figures, deadlines, or standards not supplied by the user.
  • Label every inferred score as an assumption rather than presenting it as fact.
  • If an item touches legal, safety, accessibility, or contractual obligations, tell the user to confirm with the relevant specialist before ranking it low.

Example {{backlog_items}} = "SSO login; bulk CSV export; dark mode; audit log", {{product_goal}} = "cut enterprise onboarding time", {{target_users}} = "IT admins at large accounts", {{constraints}} = "one squad, two-week sprints", {{scoring_preference}} = "value divided by effort", {{known_estimates}} = "SSO login already estimated at 8 points".