Complete AI Training

AI agent for process improvement analysts

Current-State Process Map Agent

An accurate map of the real process with waits and rework

Current-State Process Map Agent: what goes in, what the agent does and what you get

What it does

People describe how a process should work, not how it really works, so interview maps miss loops and waits. This agent reads system event logs, such as ticket or order status changes, and builds a map of the actual paths, how often each happens and where time is spent waiting. It checks the log covers the full process and enough cases. If steps are missing, it asks for other data sources and joins them. Cases still open at the end of the period are kept out of time measures. It compares the real map with the documented one and highlights loops, rework and long waits. It drafts a summary for a workshop. The process analyst approves before sharing.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Project starts 2 USES A TOOL Load event logs 3 CHECKS THE RESULT Does the log cover the full process and enoughcases? If not: ask for other data sources and load them. Backto step 2. 4 DOES Remove open cases from time measures 5 DOES Build paths and calculate waits 6 DOES Compare with documented process 7 DOES Draft workshop summary of loops and waits 8 YOU APPROVE Analyst approves summary 9 RESULT Process map and summary
Read the steps as a list
  1. Project starts
  2. Load event logs
  3. Does the log cover the full process and enough cases?If not: ask for other data sources and load them. Back to step 2.
  4. Remove open cases from time measures
  5. Build paths and calculate waits
  6. Compare with documented process
  7. Draft workshop summary of loops and waits
  8. Analyst approves summaryThe agent waits here for your OK.
  9. Process map and summary

How it decides

It builds the paths from case events and ranks steps by wait time and rework.

  • Open cases out of time measures
  • Rank by wait and rework
  • Minimum case count

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Minimum cases
  • Time period
  • Steps to merge

What keeps you in control

It always asks you first

  • Sharing the map

Hard limits

  • Never names individuals

It stops when

  • Done
  • Stop: logs unavailable

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensThe first purchase approval log had only system approvals and no goods receipt. The coverage check failed, so the agent asked for the warehouse receiving log and joined it by PO number. Across 4,100 closed cases, 27% looped back to the requester for missing cost codes. The longest wait was 3.2 days at finance review. The analyst approved the summary for the workshop.

More agents for process improvement analysts