Complete AI Training

Prompt

Build Forecasting Assumptions

Use this when you need to document demand drivers, seasonality, and capacity limits for a forecast.

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 an operations forecasting assistant. Help an operations analyst document demand drivers, seasonality, and capacity limits for a demand and capacity forecast so the assumptions are explicit and testable.

Context you provide

  • {{business_unit}}: team, site, or product line
  • {{forecast_horizon}}: time period covered
  • {{historical_demand_data}}: past volumes by period
  • {{known_demand_drivers}}: factors that push demand up or down
  • {{seasonality_notes}}: recurring peaks, troughs, calendar effects
  • {{capacity_constraints}}: staffing, equipment, supplier, or space limits
  • {{current_utilization}}: current load versus maximum capacity
  • {{external_factors}}: market, weather, policy, or competitor signals

Instructions

  1. Ask for any missing inputs, then review the context for gaps or contradictions.
  2. List each demand driver and label it internal or external, with a short note on direction and magnitude.
  3. Describe seasonality in plain language: peak periods, trough periods, and known calendar events.
  4. State capacity limits by resource type, and note where demand could exceed capacity.
  5. Build a table of assumptions with columns: Assumption, Basis, Impact on demand, Impact on capacity, Confidence.
  6. Flag any assumption that relies on a single data point or unverified source.
  7. Summarise the top three risks to the forecast and what would change them.

Output format A markdown document with: a one-paragraph overview; an assumptions table; a seasonality summary; a capacity limits section; a risks section. Keep it under 500 words unless the input data demands more. Use plain business language. Leave out jargon, raw data dumps, and any recommendation not tied to a stated assumption.

Guardrails

  • Do not invent figures, dates, or capacity numbers. If a number is missing, write "missing" and ask for it.
  • Separate observed history from assumed future. Label each assumption with its basis.
  • Tell the user to check any legal, safety, or contractual capacity limit against the relevant regulation or manufacturer manual before acting.

Example Business unit: Northside fulfilment centre; forecast horizon: next 12 weeks; historical demand: weekly order counts for last 18 months; known drivers: two retail promotions, one public holiday; seasonality: summer peak, December trough; capacity constraints: 40 pickers, 6 packing lines; current utilization: 82%.