← Back to library

Map and trim workflows before adding AI

Luke Pierce (@lukepierceops), an AI consultant, outlines the five-step method behind one operating system built for a warehousing tech client: map workflows with the staff who do them, remove unnecessary processes, wireframe, build and consolidate, then add AI last. He says it runs 450 active projects, structures receipts and invoices, and runs 28 automations in the background.

  • OutcomeBefore building AI into a company, map each workflow with the people who do it, cut needless steps and plan the system, then add automation on top.
  • SetupAn afternoon+
  • LevelAdvanced
  • CreditLuke Pierce (@lukepierceops)
View original
Works with
Custom agent

Published Oct 23, 2026Updated Oct 23, 2026AI dry run on sample input: passed Oct 23, 2026Source verified Oct 23, 2026

When should you use it?

  • A company runs on many disconnected tools
  • Leaders want AI but processes are messy
  • You can interview the people doing the work

Not for: Single small automations; Firms happy with their current system

What do you need, and what do you get?

You need

  • Staff interviews
  • Current tools list
  • Documents the work produces

You get

  • Workflow maps
  • List of steps to cut
  • Wireframe and build plan

How does it work, step by step?

  1. Interview the doersMeet the people who do the work, not only leaders.
  2. Map workflowsWrite each workflow step by step with inputs and outputs.
  3. Cut stepsRemove steps that add no value before automating anything.
  4. WireframePlan screens and data flow before building.
  5. Add AI lastAutomate document reading and reminders once the base is clean.

The prompt

Our own prompt, inspired by Luke Pierce (@lukepierceops) and written to work in Custom agent. Values in lime are placeholders.

Fill in the blanks (optional; the copied prompt uses your values)

Prompt2 variables
Help me map this workflow: {workflow}. Here are interview notes from the people who do it: {notes}.
List each step with input, output, owner and time taken. Mark steps that add no value and could be cut. Then suggest which remaining steps could be automated, and what data each automation would need.

Inspired by Luke Pierce (@lukepierceops): This headline is a preview of the next 5 years of AI consulting.. The prompt above is written by playbooks.bot in our own words; nothing from the original is quoted.

What can go wrong?

  • Replacing a working system can cost more than keeping it.
  • Cutting processes is not cutting people; say so early.
  • Scope creep is the common failure; freeze the wireframe.

Our take

The order of Luke Pierce's steps is the argument: AI arrives fifth. Teams that start by adding a model to a messy process usually get the same mess faster, with less visibility into why. Mapping the work with the people who actually do it is the step most often skipped, and it is where the list of things to delete comes from; every removed step is one you never have to automate, monitor or debug. Wireframing before building forces agreement on screens and data before money is spent. The client figures come from a consultant describing his own engagement, so read them as his claims rather than audited results. The method, though, is sound and cheap to trial. Pick one department, interview two or three people about a single recurring process, write it out step by step, and mark each step as keep, merge or cut. Only then ask where software, and finally AI, would remove real effort. Measure time per job before and after so the result is a number, not a feeling.

Where the time goes back: AI lands on a clean process instead of automating a messy one.

FAQ

Do I always replace existing software?
No. The poster builds on a clean existing setup when that is cheaper than replacing it.
How long does it take?
The poster puts it at about one quarter for a mid-sized company.

Source and credit

Inspired by Luke Pierce (@lukepierceops). We describe the idea in our own words, link to the original, and wrote the prompt ourselves; nothing is quoted.

Inspired by
This headline is a preview of the next 5 years of AI consulting.
Creator
Luke Pierce (@lukepierceops)
Published on
X
Original date
Jul 11, 2026
License
Not openly licensed (the poster keeps all rights), so we quote nothing and link to the original
Originally for
Custom agent
AI dry run
Oct 23, 2026: passed. Dry-run tested by our AI test agent on sample input. PASS: Gates held on desk check; no live sends, bookings or purchases.
AI dry run on sample data; not run with live connected accounts. The playbooks.bot test agent filled the prompt with synthetic inputs, produced the output and checked it against the page's stated outputs and guardrails.
Verified
Oct 23, 2026: source link resolves to the original, the credited creator published it, our excerpt or prompt matches it, and the license allows our use.

Are you the creator? Claim or correct this page.

More in Operations & Admin

All 203 Operations & Admin playbooks →

Related playbooks

Operations & AdminPrompt✓ AI dry run

Audit how you manage your agent team, not only the agents

Have your chief-of-staff agent review the last 30 days of how you delegated, clarified, interrupted, corrected, assigned, approved and followed up across your agent team, score eight management areas with evidence, list your three biggest weaknesses, and draft a seven-day improvement plan, without changing any bot, routine or instructions until you approve.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminPrompt✓ AI dry run

Brief an agent with nine fields, not a how-to

Before you hand a job to an agent, fill in nine short fields that name the result, the sources it may use, the deadline, what it may do alone, what needs your say-so, what it must never do, how to prove the work is finished, what to do when it gets stuck, and where to put the deliverable, then add a line that tells it to pick the simplest reliable method.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes
Operations & AdminPrompt✓ AI dry run

Brief your agent on the outcome, not every step

Hand a new job to your lead agent by describing the finished result instead of the steps, and have it confirm what success looks like, ask only for context it cannot find itself, pick the simplest dependable approach, state what it will do alone and what needs your approval, and check the result against your goal before calling it done.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes
Operations & AdminPrompt✓ AI dry run

Give every agent routine a clear point where it stops

Have your lead agent review every recurring job, monitor and follow-up your agents run, check each against four stop rules (what counts as done, what should hand control back to you, how many retries or days are too many, and how a stop gets recorded), then propose fixes ranked by risk, changing nothing until you approve.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminWorkflow✓ AI dry run

Scope an email triage agent after a month's trial

Get fast triage and routine drafts across several inboxes while keeping negotiations, sensitive relationships and CC choices with you.

Credit: Jarvis Specter

An afternoon+