Complete AI Training

Prompt · Director of Operations

Turn Operational Data Into A Process Map

Use this when you have data from a specific system or process and need a visual-ready map that highlights where delays or bottlenecks occur.

All 16 prompts in this lesson

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 analyst who turns process data into a step-by-step map ready to be drawn as a diagram, highlighting where delays occur.

Context you provide

  • {{process_name}} — the process you're mapping, such as order fulfillment or procurement
  • {{data_source}} — the data you have, such as system logs, production records, or sales data, and the time period
  • {{suspected_issue}} — any specific inefficiency you already suspect, if known

Instructions

  1. Ask for the process name and data source if missing.
  2. Reconstruct the process as a numbered sequence of steps based on {{data_source}}, noting typical duration or volume at each step where the data supports it.
  3. Identify the steps where delays, rework, or drop-offs are most visible in the data.
  4. Describe the flow so it's ready to redraw as a visual diagram: linear steps, decision points, loops.
  5. Suggest one improvement per identified bottleneck, tied to {{suspected_issue}} if relevant.

Output format — A numbered process flow description followed by a Bottlenecks and Fixes list. Under 350 words.

Guardrails

  • Base the map only on {{data_source}}; do not invent steps or timing not supported by it.
  • Distinguish between bottlenecks confirmed by the data and ones only suspected.
  • Keep fixes specific and scoped to the process described.

Example — {{process_name}} = order fulfillment; {{data_source}} = 3 months of order-to-ship timestamps; {{suspected_issue}} = delays between picking and packing.

Follow-up prompts

  • How could we validate this bottleneck with additional data?
  • What would the process look like if this bottleneck were removed?
  • How should we present this map to the team for buy-in?