The playbook

Most firms don't have an AI problem. They have an AI operating problem.

A field guide that takes your firm from scattered personal use to AI running real work reliably. You don't read it front to back: you find where you actually are and start there.

The maturity ladder

Where is your firm today?

Stage 15 articles
Personal leverage
Win back hours, starting this week.
You're here if…
  • A few people use AI, in their own way, when they remember to
  • Nothing is written down or shared
  • The value is real but scattered and personal
Read Stage 1
Stage 26 articles
First business win
Bring AI into one real workflow.
You're here if…
  • Individuals are fluent, but no process runs on AI yet
  • You want a shared quality bar, not personal hacks
  • You're asking "what should we automate first?"
Read Stage 2
Stage 36 articles
Expansion
Let automation and agents carry more.
You're here if…
  • One workflow works; you want the next five
  • You're weighing your first agent, and want it done safely
  • Things run unattended and you need guardrails
Read Stage 3
Stage 46 articles
Operating engine
Make AI a dependable part of how the firm runs.
You're here if…
  • AI is in several workflows but nobody owns the whole picture
  • Clients, insurers, or regulators are starting to ask
  • You want it to compound, not sprawl
Read Stage 4
This is not a ladder you owe anyone a climb. Many firms operate well at Stage 2 or 3 indefinitely, and each stage says so at its close.
Stage 1 · Personal leverage

Win back hours, starting this week

Which recurring tasks actually drain your week? You have the basics; the value so far is real but scattered. That is the norm, not a failing, and this altitude is where it stops being scattered.

The articles

Most people meet AI as a curiosity: a poem on demand, a quick answer to a trivia question. The working version is less flashy and worth far more. The rule of thumb: if the work is made of language, a model can carry part of it today. Reading, drafting, comparing, extracting, summarizing. What stays yours is the judgment: decisions, advice, the final word. The model prepares those, gathering and comparing so you decide with better material, but it does not make them. Most professional jobs are a braid of the two, which is why the gains land everywhere once you look for them.

If the work is made of language, part of it can be handed over.

A model is strongest at volume: the reading and comparing that costs a person hours, done without tiring on page two hundred. It is weakest where a single exact fact or figure must be right without checking. Both of those facts shape how you use it, and the next article turns them into working habits.

What you're building
A week, marked up
Your own task list is the first deliverable: each item marked for what it is, not what it feels like.
Sample output — yours will look like this
Read the new supplier agreements, pull the deadlines
the PDFs in the shared drive, into one table for checking
Hand overlanguage work: extracting and tabling
Answer the enquiry emails in the shared inbox
the Outlook inbox, replies in the firm's tone
Hand over the draftsthe model drafts, you read and send
Decide whether to re-price the Meyer engagement
the model compares this year's hours against the fee, you make the call
Stays with youjudgment; the model prepares it
Reconcile the bank export against the ledger
two exports that never quite agree, matched line by line, variances explained
Hand overcomparing is language work too
Write the Friday client update
drafted in your voice from the week's notes and sent messages
Hand over the draftyour review puts your name on it
Approve the final proposal
the last read before it carries the firm's name
Stays with youthe final word is the job

The tasks are one person's week at one firm. The marking is the part to copy: language work gets handed over, judgment stays, and most items turn out to be a braid of both.

The personal version of the exercise Stage 2 runs for the whole firm. What you mark "hand over" this week becomes your delegation practice; what a whole team keeps marking becomes the firm's first workflow candidate.
The moves
01

Separate the language work from the judgment

Go through what you actually did this week and sort it. The test is what the work is made of, not how important it feels.

Language work can be handed over

Reading, drafting, comparing, extracting, summarizing. If you could hand the task to a capable temp with written instructions and check the result, a model can carry it.

Judgment stays, prepared

Decisions, advice, the final word. The model gathers, drafts and compares so you decide with better material. It prepares the decision; it does not make it.

02

See the six patterns across one firm

The same tool lands on every desk differently. Six patterns cover most of what firms find in the first month.

1The core work. A lawyer hands over a stack of agreements and gets back every deadline, obligation and unusual clause, matched and tabled for checking.
2Intake and admin. Messy enquiry emails become clean, structured entries for the case or project system, drafted and ready to file.
3Finance. Two exports that never quite agree get reconciled line by line, with the variances explained in plain language.
4Communications. One announcement gets drafted in the firm's tone, then reshaped for the newsletter, the website and a client email.
5Analysis. A raw spreadsheet export comes back as the finding: what moved, what did not, and what is worth a closer look.
6Research. A week of industry and regulatory news becomes a briefing note, with the two developments that matter on top.

These are patterns, not products. None of them needs special software: the mainstream AI tools can do a version of all six today, in a browser, from a plain-language instruction. Your own week supplies the real ones.

03

Mark up one real week

Keep your task list for one full week and mark each item as it happens. Marking real work beats brainstorming about work, for the same reason the later stages record processes instead of remembering them.

Three marks are enough

Hand over, hand over the draft, stays with me. Most knowledge work turns out to be the middle category: the model produces, you certify.

Note what repeats

The items that show up every week are worth the most, because whatever you learn about them pays back weekly. They are also the seeds of the library this stage ends with.

04

Hand over the first task this week

Pick one recurring item from the marked-up week and delegate it properly. The weekly update is a good first choice because the material already exists and you can judge the result at a glance.

Give it a real instruction

Gather your notes and sent messages from the week, and ask for a summary of what was achieved, in your voice, addressed to the person who actually reads it.

Review, fix, send

The first draft will be close. Your corrections are not overhead; they are what makes next week's draft better, and they are the beginning of the review habit the next article builds.

This needs no setup, no connected systems, and no permission from anyone. The account you already have is enough, and free plans are enough to feel the gain.

Takeaway

If the work is made of language, a model can carry part of it today, across every desk in the firm. Start with one recurring task this week; the account you already have is enough.

The difference between a party trick and a working tool is how you talk to it. A question asks the model to produce an answer from what it already knows, and you get something generic, because it knows nothing about your situation beyond the sentence you typed. A brief gives it the situation, the material and the standard to hit, and you get something usable. Most disappointing first experiences with AI are questions that should have been briefs.

Brief it like a colleague, review it like a junior's draft.

Any of the mainstream tools works, in a browser, and a free account is enough to learn on. The tool is not the skill. The skill is the loop this article teaches: instruct, review, improve.

The mechanism
One task, run through the loop
The same follow-up email, asked for twice. The difference is everything this article has to teach.
The question
"Write a follow-up email to a client."
Generic by necessity. The model knows nothing about this client, this thread, or what the follow-up needs to achieve.
The brief — instruct
"Here is the email thread and the proposal we sent. Draft a follow-up that answers their two objections, in my tone, under 150 words."
The situation, the material, the standard to hit. The model now works from your facts instead of guessing them.
The read — review
The draft comes back well structured. The day rate it quotes is from the old proposal.
Read it the way you would a junior colleague's draft: the structure is usually sound; the specific facts, names and numbers are where you check.
The correction — improve
"The rate is from the old version. Use the one in the attached proposal, and tighten the middle paragraph."
The second pass with real feedback is where the quality jumps. This brief worked; keep it. The last article of this stage turns it into an asset.
The loop is the skill, and it transfers: the same instruct, review, improve carries the whole-task delegation in 1.4, and it is the loop the Stage 2 build runs on a schedule.
The moves
01

Instruct with the full brief

State the task, hand over the material, name the audience and the standard to hit. What you would tell a colleague taking over the task is exactly what the model needs.

Attach the material, don't describe it

The thread, the notes, the document. The model works far better from your actual material than from your summary of it.

02

Review like a junior's draft

The structure is usually sound. The specific facts, names and numbers are where you check.

Make "look it up" your reflex

When a fact matters, have the model search and cite its source rather than answer from memory, and open the source yourself when the stakes are real.

AI executes, you certify. Nothing the model produces goes to a client, a colleague or a decision without your read. That rule is what makes delegating safe, here and at every later stage.

03

Improve, and keep what worked

Say what is wrong and ask for the revision. The second pass with real feedback is where the quality jumps.

Feedback beats retyping

"The rate is wrong, use the attached" gets a better result than silently fixing it yourself, because the correction improves this draft and teaches the next brief.

Keep the briefs that worked

A brief that produced a usable draft is worth saving as written. The last article of this stage turns those into a library.

04

Run two session habits

Two habits keep the loop honest, and both are about what the model knows.

Give it your context deliberately

Within a chat it knows only what you put there. Paste the relevant background instead of assuming it remembers your business. The next article makes this permanent.

Start fresh when a thread drifts

A long chat full of corrections drags old misunderstandings forward. Take the good output and open a clean session.

Takeaway

Brief it like a colleague, review it like a junior's draft, and make "look it up" your reflex. The loop is the skill; the tool is just where you run it.

A model that knows nothing about your business starts every chat as a stranger. This article shows you how to give it your world once. Context is what it knows before the chat starts: standing instructions, project spaces, memory. Connections are what it can see and do, live, inside the access you granted: your calendar, your files, your inbox. Set up what it should always know, connect what it may see, and keep both under your control.

Give it your world once, and whole categories of asking disappear.

Control is the other half of the setup. You stay accountable for what the system reads and does. Narrow scopes, per-use approvals, and an occasional look at what memory has stored keep that accountability real rather than theoretical.

The mechanism
Three layers, each under your control
What the model knows and what it can reach, from standing knowledge to live systems.
Layer 1 · It always knows
Standing instructions
Who you are, what the firm does, how you like output. Written once, applied to every chat.
Your control · you can read it and edit it at any time; fix what has gone stale.
Layer 2 · It knows inside the space
Project spaces
One space per ongoing matter, loaded with its documents, so every chat inside it starts informed.
Your control · you decide what goes in, and the space holds only what you loaded.
Layer 3 · It can reach, live
Connections
The systems the work needs: calendar, files, inbox. It reads or acts inside the access you granted, and nowhere else.
Your control · granted one at a time, smallest access that does the job, revoked when the work stops needing it.
Each layer reaches further than the one above it, and each stays a decision you make rather than a default you inherit. The Stage 2 build runs on exactly these: the saved instruction is layer one, the loaded documents are layer two, the systems it reads and writes are layer three.
The moves
01

Write the standing instructions

Who you are, what the firm does, how you like output. Written once, applied everywhere.

Keep it short and current

A page at most: the firm's services, your role, the tone rules, the formats you want by default. Stale instructions quietly degrade every chat, so re-read them when something feels off.

02

Build one project space per matter

One space per ongoing engagement, loaded with its documents, so every chat inside it starts informed.

Load the documents, not descriptions of them

The engagement letter, the notes, the drafts. A space that holds the real material turns "let me explain the background" into a sentence you stop writing.

03

Connect the narrowest access that does the job

Connect the systems the work needs, one at a time, each with the smallest access that does the job.

Grant access like a key card

To named rooms, not the whole building. Approve actions per use rather than choosing "always allow", and disconnect what you stop using.

Let the work pull the connection

Connect the calendar when the Friday summary needs it, not because connecting things feels like progress. Every connection you don't have is one you don't have to think about.

04

Run the three checks on your setup

A short standing review of your own setup, the personal version of what Stage 4 runs for the whole firm.

What does it know?
Read the stored memory and standing instructions; fix what has gone stale.
What can it reach?
List the connections; disconnect what the work no longer needs.
What can it do alone?
At this stage, nothing irreversible without your approval.
05

Know what happens to what you paste

One professional habit before client work goes in: know what your tool does with what you paste.

The data settings tell you whether conversations are used for training and how long they are kept. Check once, choose deliberately, and the question is settled. Client material deserves the same care in an AI tool as in any other system, and when colleagues start following you onto the tool, business terms are the natural next step. The later stages pick that up.

Takeaway

Give it your world once: standing instructions, project spaces, and the narrowest connections that do the job. What it knows and what it may touch stay decisions you make, not defaults you inherit.

Questions save minutes; delegated tasks save hours. Dabbling uses the model for fragments of work you are already doing, a sentence here and an idea there, so the task's shape and its hours stay yours. Delegating hands the model a whole task with a brief and makes you the reviewer: it produces the full first version, and your time moves from producing to directing and certifying. The real hours live on the delegating side of that line.

Delegate the task, keep the judgment.

Delegation changes where your time goes, not how much care the work gets. A real brief plus a serious review still puts your judgment on the page. What disappears is the hours of production in between.

What you're building
The delegation brief
What handing over a whole task looks like in practice. This one hands over a proposal first draft.
Sample output — yours will look like this
Brief: proposal first draft
Written once for this task · reused and improved each time it runs
The task

Produce the full first draft of the proposal for the new enquiry, ready for my review. Not an outline and not suggestions: the complete draft.

The material

The enquiry notes from Outlook, our last two proposals for similar work from SharePoint, and the outline I sketched. Work from these, not from general knowledge.

Audience and tone

The client's operations director, who reads quickly and dislikes filler. Our standard tone: direct, concrete, no superlatives.

What good looks like

Three pages at most. The scope section mirrors their language from the enquiry. Anything the notes don't cover is flagged as an open question, not invented.

The review that follows

I check every fact, name and number against the material, judge the structure, and send back one round of pointed corrections before anything leaves my desk.

The systems and the task are one firm's. The five sections are the part to copy; they fit any task with a known good shape.

This brief is the personal ancestor of the saved instruction the Stage 2 build runs automatically. Write it well once and it outlives the chat it was written in.
The moves
01

Pick work with a known good shape

Proposals, summaries, reports, first-pass reviews: you can describe what good looks like, so you can check what comes back.

First versions, not final calls

Drafts, options and analyses to react to. Decisions stay with you.

Stay inside its strengths

Language-shaped work first. Hold back the tasks where you cannot verify the output until your verification habits are solid.

02

Write the brief so it holds the standard

Material, audience, tone, and what good looks like, stated up front. The brief is where your judgment enters the work.

Say what must not happen

"Anything the notes don't cover is flagged, not invented" belongs in the brief. The model follows the standard you state, not the one you assume.

03

Review like the work carries your name

Facts, names and numbers checked; structure judged; nothing forwarded unread. Because it does carry your name.

04

Plan the second round

One revision with pointed feedback is part of the job, not a failure of the tool.

Brief, review, iterate: the same loop as the last article, on a bigger unit of work. The habits carry over unchanged; what grows is the size of what you hand over.

Tasks you delegate well once are worth making repeatable. That is the last article of this stage.

Takeaway

Delegate whole tasks with a real brief, and review like the work still carries your name, because it does. The hours come back on the delegating side of the line.

The first delegation is an experiment; the fifth should be a routine. A good result that disappears into the chat history gets rebuilt from memory next month, a little differently, with a little less of what made it work. A captured asset is that same working brief, saved as a reusable instruction that runs the same way every time and can be handed to a colleague. Capture is the difference between being personally faster and building something the firm keeps.

When it works twice, capture it so it works every time.

This is also the point where paying for the tool earns its keep. A free plan carries you through the working loop. The machinery of capture and context, project spaces, connections, saved skills and scheduled tasks, mostly lives on the paid plans. You are not paying for a smarter chatbot; you are paying for the features that turn a chatbot into a delegation system.

What you're building
The starter library
What capture looks like a few months in: five briefs, one skill, one scheduled task, kept in one agreed place.
Sample output — yours will look like this
Brief
Proposal first draft
the full brief from the last article: task, material, audience, standard, review
Brief
Weekly client update
from the week's notes and sent messages, in my voice, to the person who reads it
Brief
Meeting notes to actions
the transcript in, owners and deadlines out, open points flagged
Brief
Contract dates extraction
agreements in, every deadline and notice period tabled for checking
Brief
Enquiry to intake entry
a messy enquiry email becomes a clean, structured entry, ready to file
Skill
How we write summaries
the firm's structure and tone as a multi-step instruction, shared with two colleagues
Scheduled
Monday pipeline summary
arrives at the start of the week, built from the connected calendar and the CRM; the instruction gets improved, not each output

The entries are one person's. The shape is the target: your most-repeated tasks captured and named, the most-delegated one written as a skill, one piece of recurring work arriving on its own.

Every asset here makes the next delegation cheaper, and this library is the seed of something larger: Stage 4 turns it into a firm-wide asset. It starts here, personal and small.
The moves
01

Capture the briefs that worked

Your best instructions, kept where you can rerun them, named so you can find them.

Capture at the moment it works

The brief that produced the best draft so far is in this week's chats right now. Saving it takes a moment; rebuilding it from memory next month loses a little of what made it work.

02

Turn the most-repeated task into a skill

A skill is a multi-step instruction the model applies on demand: how your firm writes summaries, structures reviews, formats deliverables.

Reusable, versionable, shareable

A skill can be handed to a colleague and improved in one place. It is the first asset you build that belongs to the firm rather than to a chat.

03

Schedule the truly recurring work

Recurring work set to run on a rhythm: the Monday pipeline summary now arrives instead of being produced.

Review what arrives

The certify rule still holds; what changes is when your attention enters. Improve the instruction, not each output, and the quality compounds.

04

Keep the library like code

One agreed place, improved in the master copy, reviewed now and then for what has gone stale.

Scattered private variants are how capture quietly dies: three colleagues each running a slightly different version of the same brief, none of them the best one. The master copy is the asset; everything else is a copy of it.

The starter library, concretely: five briefs that work, one skill, one scheduled task. That completes what this stage set out to build.

Takeaway

When it works twice, capture it: templates, skills, scheduled tasks. Personal speed fades with the person; a library compounds.

Closing this stage
Operating well at this altitude

Most readers should run at this altitude for a while, and staying here is fine. Operating well at Stage 1 looks like: the setup is trusted, the delegation habit is real, and the starter library grows on its own. Stage 2 is there for the day a shared workflow needs more than personal leverage, not because a next stage exists.

Stage 2 · First business win

Bring AI into one real workflow

This is where AI stops being a personal trick and becomes something the firm does: one shared workflow, with a quality bar that survives the enthusiast leaving.

The articles

Stage 2 starts with a decision: which workflow gets your firm's first serious AI effort. Candidates are not the problem; any working firm has plenty of tasks that could be automated. The hard part is picking one you can actually win. The workflow that would look best in a demo is usually heavy on judgment, full of exceptions, and wired into half your systems, and that is exactly why it fails as a first build. The right first pick is less glamorous: it runs often, it has a clear start and finish, and a mistake is easy to catch. This article takes you from a long list to one commitment.

Pick a workflow you can win, not one that would impress in a demo.

One thing to know before you start looking: the workflows worth winning are not always the ones people complain about. A step everyone grumbles about may cost the firm very little, while a quiet step nobody mentions holds up every job that passes through it. What matters is where work sits and waits, and where your most qualified people lose their week.

What you're building
The shortlist memo
The document this article produces: five candidates, one recommendation, ready for your management team.
Sample output — yours will look like this
Memo
ToThe management team
FromThe operations lead
ReFirst workflow for automation — shortlist and recommendation
The shortlist, ranked
  1. Proposal drafting. The first version of every proposal, written from the intake notes. Runs weekly, has a clear start (the enquiry) and finish (the sent proposal), and the inputs already sit in our systems: intake notes in Outlook, past proposals in SharePoint, rates in Excel. A partner can judge quality by reading one draft.
  2. Supplier invoice checking. Amounts are reconciled by hand between the accounting system and Excel. Passes the tests, but fewer hours are at stake and the delay costs us little.
  3. Support inbox replies. High volume and real pain, but the right reply varies too much case by case for a first build. Revisit after the first win.
  4. The monthly client status report. Well bounded and easy to check, but it runs twelve times a year; a pilot would take a quarter to judge.
  5. A client assistant covering enquiry to renewal. Raised more than once, and set aside: no clear boundaries, no way to measure it, and it touches half our systems.
Recommendation

Proposal drafting. It passes all seven tests, the hours at stake are real and weekly, and we can check quality by reading the drafts. Next step: map the current process and count one week as the baseline.

A sample, not a prescription: the candidates and systems are this firm's. A software business might recommend support replies; an accounting practice the seasonal crunch. The shape of the document is the part to copy.

The end product of this article: a short document your management team can read and challenge. The next article maps the recommended workflow in detail; the rest of the list keeps its place for later.
The moves
01

Take stock of where work piles up

The inventory is a structured look at the firm's day-to-day work, and it deserves real attention: this list decides where the first months of effort go. Three questions organize it.

Where does work sit and wait?

Go through the queues and handoffs: quotes waiting to be written, tickets waiting to be answered, invoices waiting to be checked. Shared inboxes and handoff points hold more of this than any job description shows.

Where do people retype and reconcile?

Every place someone copies data between systems, cross-checks two lists, or re-enters something that already exists somewhere is a candidate.

Where does the same work repeat?

The reports, replies, summaries and first drafts that get produced to the same pattern every week.

02

Look at what people already hand to AI

If the firm has been through Stage 1, part of the list writes itself. The tasks people already delegate to a model on their own initiative are strong signals for what the firm should automate properly.

Ask what people do, not what they think

"What do you already hand to AI?" gets concrete answers. "What should we automate?" gets opinions. In many firms the proposal queue shows up here on its own, because associates are already drafting from intake notes with a model and tidying the result by hand.

03

Filter with the seven tests

Every candidate faces the same seven questions. One clear no is usually reason enough to set a candidate aside for now.

1Frequent. It happens daily or weekly, so there is enough volume to judge the result and the win pays back quickly.
2Bounded. It has a clear start and finish, with defined inputs and outputs.
3Painful. It costs real hours or real delay today, so winning it is worth something.
4Measurable. You can put a number on how it runs today; the next article captures that number as your baseline.
5Low error-stakes. A mistake is cheap to catch and cheap to fix, because early versions will make some.
6Mature. The process is stable and understood, not something the firm is still redesigning.
7Fed by reachable data. The inputs it needs exist in systems you can connect, current and complete, not on paper or in one person's head.

Proposal drafting passes all seven at most firms that live on proposals: weekly volume, a clear start and finish, real hours at stake, an easy quality check, and inputs that already sit in the firm's systems. Whatever passes at your firm moves on to the ranking.

04

Rank what survives

Score each surviving candidate on two dimensions, and take your first workflow from the high-impact, low-effort corner. Impact means the hours it drains, how often it runs, what delay costs, what an error costs. Effort means the systems it touches, the exceptions it produces, the judgment it needs.

High impact · low effort
Proposal drafting
The first win. Real hours every week, three systems, few exceptions.
High impact · high effort
Support inbox replies
Worth winning, too heavy for a first build. This is what Stage 3 is for.
Low impact · low effort
Invoice checking · the monthly report
Easy wins that move little. Good later, once the muscle exists.
Low impact · high effort
The client assistant
Neither first nor soon. Off the list until it can be bounded and measured.

The candidates placed are the sample shortlist's. Place your own survivors the same way, and the choice usually makes itself.

05

Commit to one bounded task

The commitment is a single bounded task, not a department. If proposals are the bottleneck, the first win is drafting the first version of one, not reinventing how the firm sells.

Put the commitment in writing

One line: the task, where it starts and ends, and the person who owns the win. That line becomes the top of the map you build next.

Where firms typically start: proposal drafting at an agency, first-line support replies at a software business, the seasonal crunch at an accounting practice, after-hours enquiries at a local service firm. Treat these as places to look, and test what you find against the seven tests.

Takeaway

Candidates are everywhere; choosing is the skill. List broadly, filter with the seven tests, rank by impact against effort, and commit to one bounded task.

Every workflow exists twice. The official version lives in handbooks and job descriptions, written from memory, often by someone who no longer does the work. The real version runs this week, full of workarounds, exceptions and quiet fixes people have built over time. Builds fail because they are made from the official version and run against the real one. This article's job: a map of the real process, made in about a week, with the number it actually costs attached.

Map the real one. The gap between the two is where the surprises live.

One part of the real process is nearly invisible: people repair bad data without noticing they do it. They spot the stale price, remember the address that one system gets wrong, fill the field the form left empty. That silent repair disappears the moment a machine runs the step. And a model handed wrong data does not stall the way a script would; it produces something plausible from it. So the map has to capture the data behind the work, not just the steps.

What you're building
The process map
The one-page deliverable this article shows you how to make.
Sample output — yours will look like this
Process map — proposal draftingone page · a new colleague could run it
The steps — who, and which systemsWhat recording found — and memory missed
01
Enquiry arrives and is logged
front office · reads: the shared Outlook inbox · writes: a new record in HubSpot
02
Partner scopes the ask
partner · reads: the enquiry + that client's past work in SharePoint · judgment call
Decision pointstandard pitch or custom — the fork the route choice runs on
03
Associate assembles the draft
associate · reads: past proposals in SharePoint + the Excel rate sheet · writes: the draft in Word
Quiet fixthe day-rate HubSpot still shows is stale — corrected by hand from the Excel sheet
04
Partner reviews and prices
partner · marks up the Word draft · prices against the Excel rate sheet
Exceptionrush requests jump the queue — labeled as such, not smoothed over
05
Proposal goes out
associate · sends the PDF from Outlook · logs it on the HubSpot record
runs as the handbook says — worth knowing too
The before-numbercounted across one real week
How often · how longevery pass, arrival to sent, incl. waiting
Where it waitsthe handoffs and queues
Ownerone named person

The systems are this firm's, not a requirement — yours might read Gmail for Outlook, Salesforce for HubSpot, Google Drive for SharePoint. What matters is that every step names its real ones.

The whole deliverable on one page, at the depth the next article runs on. The route choice needs to know each step's real systems and forks before it can say what should be a rule, what needs AI, and what needs an agent; the right-hand column is what recording finds and memory misses.
The moves
01

Record one full pass

The person who does the work records one real pass, screen and voice, the way they would show a new joiner. Recording produces the real process; writing from memory produces the official one.

Use the recorder already on the machine

A solo video call with recording on, a screen-recording tool, or the system's built-in recorder, whichever is already at hand. The transcript matters more than the video.

Record a real, untidied item, not a staged one

Narrate the off-screen parts as you go ("here I'd normally ring the client back") and don't clean anything up: the workarounds are the most valuable footage.

If several people touch the workflow, each records their part

For the proposal workflow: one recording from the associate, one from the partner. The handoff between them is the most important footage.

02

Set up the mapping space, then let a model draft

Hand the transcript to a model with clear instructions and it returns a structured first draft of the map. The setup decides whether this works once or becomes a habit.

Mapping one workflow? A chat is enough

Upload the transcript with an instruction describing the output: steps in order, who does each, the systems read and written, exceptions marked as such, the decision points. One page, plain language.

Mapping more than one? Make it a standing setup

A dedicated project for process mapping, with that instruction saved once. Every transcript comes back in the same shape, and the space keeps past maps, so a new map's handoffs connect to workflows the model has already seen.

That growing collection is the beginning of the firm's process library, the one Stage 4 makes official. The fifth map costs a fraction of the first.

03

Certify the draft yourself

The person who does the work corrects the draft against the recording. A person certifies the map, not the model.

Fix what the recording missed

Wrong order, skipped steps, and the exceptions that didn't occur in the recorded pass. Mark every decision point; the next article chooses the route on exactly those forks.

Hold it to the bar

One page, plain language, and a new colleague could run the workflow from it. If they couldn't, it isn't done.

04

Check the data the work runs on

An automation is only as good as the data it reads, and the map is where you find out what that data really is.

Name the systems
What each step reads from and writes to. One line each on the map.
Spot-check real records
Four failures matter: wrong, out of date, incomplete, inconsistent between systems.
Mark the quiet fixes
Each one gets cleaned at the source or written into the build. Never left silent.
05

Count one real week

The baseline is one real week of the workflow, counted while it happens, not remembered afterwards and not asked about, because people's sense of time saved is not a reliable measure.

Put a tally where the work happens

A shared sheet is enough. Count every pass that week and time each one door to door, arrival to finished, including the waiting. Delay usually hides in the handoffs, not the doing.

This number is what the pilot will be judged against; without it nobody can say whether the new way is better, including you. No baseline, no build.

06

Name one owner

A single person owns the map, the baseline, and the decisions the next moves ask for. Named now, not after something goes wrong. When the pilot needs its go or no-go, this is whose name is on it.

Takeaway

The map is an asset, not admin. Record the real process, check the data it runs on, count a real week, name one owner; everything you build next stands on it.

The map from the last article lists the workflow step by step, and every step is one of two kinds. Mechanical steps run the same way every time: the same situation always gets the same action. Judgment steps depend on what is in front of you: something has to be read, weighed or worded before the step can happen. Mark each step as one or the other and most of the routing work is done, because the rule of this article is simple: give every step the simplest tool that can handle it.

Give every step the simplest tool that can handle it.

The tools come in three sizes. A fixed rule does the same thing every time: if this happens, do that. An AI step takes on a single judgment task with your instructions and guidelines. It drafts the reply in your tone or pulls the key figures from the contract, then it stops. An agent gets a goal, boundaries and sign-off points, and carries out several steps in a row, choosing the order to fit the case in front of it. Every size up buys flexibility and gives up predictability. That is why the simplest tool that fits is the right choice, not the most capable one.

The mechanism
The build, wired
The proposal workflow as it is actually set up on an automation platform: one trigger, two rules, one AI step, two human stops.
Trigger
A new enquiry lands
The automation watches the shared Outlook inbox and the website form. Either one starts a pass; nobody has to notice the enquiry for the workflow to begin.
Rule
Create the client record
writes: a new record in HubSpot with the enquiry attached. Same action every pass, so it is configured once and never thought about again.
Waits for a person
Partner scopes the ask
The workflow assigns the partner a task and stops until it is answered: standard pitch or custom. This is the map's decision point, built in as a configured stop rather than left as a habit.
AI step
Draft the proposal
reads: the intake notes, that client's past proposals in SharePoint, the current rate sheet in Excel · writes: a draft in Word, with anything the notes don't cover flagged. It runs on the saved instruction: written once, improved deliberately.
Waits for a person
Partner reviews and prices
The draft waits for sign-off, and nothing leaves without it. Which results need a person's approval is the firm's own written choice; this firm drew the line at everything a client sees.
Rule
Send and file
writes: the PDF goes out from Outlook, the file lands in SharePoint, the HubSpot record gets the log entry. Mechanical again, so a rule closes the pass.
Where an agent would enterNothing on this build needs one. An agent earns its place where rules keep producing exceptions until nobody can hold them all. A support inbox is the classic case: the right reply depends on who is writing and what has already happened. Stage 3 starts there.
The systems are this firm's; swap in your own. What transfers is the shape of the build: a trigger starts the pass, rules carry the mechanical steps, one AI step carries the judgment inside its instructions, and the two stops the firm chose are configured into the flow, not remembered.
The moves
01

Mark every step: mechanical or judgment

Go through the map with its owner and mark each step as one or the other. The decision points the map already carries show you where the judgment sits.

The test is a sentence

If you can honestly finish the sentence "when this happens, we always do that", the step is mechanical. If the sentence needs "it depends", it is a judgment step.

Disagreement is information

When two people who know the workflow disagree about a step, it is usually two steps written as one. Split it on the map before you route it.

02

Hand the mechanical steps to rules

Rules are predictable and fast, and they should carry as much of the workflow as possible.

Start with what your systems already offer

Most systems on the map, from the inbox to the client record to the file store, have automation built in for exactly these moves: when an enquiry arrives, create the record; when a proposal is sent, file it and log it. Where two systems don't reach each other, the automation platforms bridge them.

Keep each rule small enough to say out loud

One rule per line of the map. If nobody can say from memory what a rule does, nobody will trust it when the workflow misbehaves.

03

Give each judgment step one AI task

An AI step is a single task, handed over with instructions and guidelines. It does that task and stops. What happens before and after stays with you.

For occasional use, a saved instruction is enough

The associate opens the standing instruction: draft the first version from these intake notes, in the firm's tone, at this length, and flag anything the notes don't cover. Attach the notes, get a draft back.

For every pass, wire it into the flow

The same instruction, saved once inside the automation that carries the workflow, so every new enquiry produces its draft without anyone opening a chat. That is the AI-step node in the build above. The instruction gets written once and improved deliberately, instead of being retyped a little differently each time.

Either way, the instructions are the real asset. The tone, the structure and the guidelines survive even if the tool underneath changes.

04

Build the stops in, don't rely on habit

The steps that stay with people become configured stops in the flow: the workflow assigns the task, waits for the answer, and cannot continue without it.

Decision points become assigned tasks

The partner's scoping call arrives as a task with the enquiry attached, and the pass waits until it is answered. Built this way, the decision point cannot be skipped on a busy day.

Approvals become required sign-offs

Which results need a person's approval before they count is the firm's own written choice. Many firms draw the line at whatever a client sees. Wherever your firm draws it, the stop belongs in the build, not in anyone's memory.

05

Know when a stretch needs an agent

If the steps in your map hold for nearly every case, automate them as written. If the rules keep producing exceptions until nobody can hold them all, that stretch is agent territory.

What handing over a stretch means

The agent gets a goal, the boundaries it works within, and the points where it pauses for your sign-off. It chooses the order of the steps to fit the case. You still decide which systems it may use and which rules it follows.

The big automation platforms now offer agents as a built-in option, so getting one is not the hard part. Deciding how much of the route to hand over is. On the proposal build, nothing needs one; the work that varies case by case arrives with Stage 3's support inbox.

06

Choose ready-made or custom-made

Someone has to build the route, and the choice follows the map. It is a question of fit, not of ambition.

Ready-made: the workflow as a finished product

A vendor has already built it, the way bookkeeping tools read receipts on their own. A product has one shape, so the adapting is done by your process, not by the software. That trade is worth making when your workflow looks like everyone else's.

Custom-made: built around the map you drew

The map stays the map instead of bending to a product's shape. This used to mean a software project. With standard connectors, models that can operate tools, and mature automation platforms, a custom workflow is now assembled from proven parts and hardened in weeks rather than built from scratch in months.

Where it runs is a separate decision: a vendor's cloud, or systems you run yourself. Data sensitivity, what you already run on, cost at your volume, and who can maintain it all weigh in. And before anything gets built, three lines go on the map in writing: what it may see, what it may do alone, and who owns it — an alert when it fails, and one named person who responds.

Takeaway

Rules for the steps that never change, AI for the judgment steps, an agent for the stretches that vary. The simplest combination that covers your map is the right one.

The route is built. Now it has to prove itself. A hard switch, where the old way is dropped on day one, puts everything on a setup nobody has watched under real load and leaves the way back improvised. A parallel run tries the new way while the old way keeps producing the real output, so nothing breaks while you learn. Pilot in parallel, and decide before the first run what result means go.

Stopping is a result, not a failure.

The referee is the baseline from the map, the counted week. The pilot's numbers are compared against that count, not against how the old way felt. And the pass mark gets written before the first run for a simple reason: a pass mark chosen after the results are in will bend toward whatever result you got.

The mechanism
The parallel run
Both ways run for a month. One of them is real; the counts decide which one survives.
The old wayAssociate drafts by hand
Keeps producing everything clients see, exactly as before. Every pass timed.
The new wayThe build drafts every pass
Drafts every proposal too. Timed and reviewed; nothing it produces is shipped.
Week 1Week 2Week 3Week 4
The end dateThe counts against the pass mark. Go, no-go, or one written extension.
Nothing breaks while you learn, because the old way still does the real work. The comparison this produces is the whole point: the same passes, done both ways, counted.
What you're building
The pilot plan
One page, written before the first run. The sign-off boxes stay empty until the end date.
Sample output — yours will look like this
Pilot plan: proposal drafting
Written before the first run · owner: the named owner from the map · runs one month, end date fixed
Pass mark, set before the first run

A pass has to take less time than the baseline week's count, door to door, and the share of drafts the partner accepts without rework has to beat the share we counted in the baseline week. Both reference numbers come from the map. Nothing here is estimated.

How it runs

The build drafts every proposal from the intake notes in Outlook. The associate still writes theirs as usual. Both are timed, and the partner's review decides which drafts would have gone out.

Who judges

The associates who draft daily, and the partner who signs off. The person who built the setup supports the pilot and does not judge it.

What we log

Every pass, timed door to door, both ways. Every case the new way got wrong, and why. The difficult cases from the map (the urgent request that arrives by text, the client record with the stale rate) get pushed through deliberately.

On the end date
Go No-go Extend, in writing, with a new end date
Signed by the owner
On the end date

The systems named are this firm's; swap in your own. What transfers is that everything except the counts is written before the first run.

During the pilot the only things added are the counts and the log of what broke. On the end date, one box gets marked and signed.
The moves
01

Keep the old way running

The old route keeps producing the real output for the whole pilot. The comparison is the point of the exercise.

Same passes, both ways

The build drafts every proposal, and for this month the associate still writes theirs as usual. Both are timed, and the partner's review says which drafts would have gone out.

Clients see the old way's output until the pilot passes

The new way's drafts are compared, not shipped. That is what makes an honest pilot affordable: a bad week costs learning, not clients.

02

Write the pass mark before the first run

Two or three numbers on one page, each set against the baseline week's count.

Use measures the baseline already has

The time a pass takes door to door, and the share of drafts accepted without rework. If the pass mark needs a number the baseline week didn't capture, go back and extend the baseline first. A comparison without a referee turns into an argument.

This is the discipline the whole article hangs on: go and no-go are defined before anyone has seen a result.

03

Fix the end date and name the judges

A pilot gets a fixed end date, and the people who do the work daily judge it, not the person who built the setup.

A month is typical for a weekly workflow

Long enough for the workflow to run often enough to judge. Short enough that a decision has to come. A fixed date beats an open-ended trial that nobody ever closes.

The builder supports, the team judges

A demo run by the builder always looks good. The real test is an ordinary Tuesday under real load, in the hands of the people who will live with the workflow afterwards.

04

Test it on the difficult cases

If nothing broke, the pilot was probably too gentle. Put the difficult cases through on purpose.

The map's exceptions are the test set

The workarounds and exceptions the map recorded, like the urgent request that arrives by text or the client record with the stale rate, are exactly the cases to push through the new route while the old way is still there to catch them.

Every miss goes in the log with a line about why. That log becomes the work list for the rollout, not a mark against the pilot.

05

Decide on the end date

On the end date the owner reads the counts against the pass mark and makes the call: go, no-go, or one extension in writing with its own end date.

Did it beat the baseline?
The counted result against the counted week, not an impression.
Did the team accept it?
Drafts used as they came out count. Drafts quietly rewritten do not.
What broke, and why?
The log of misses, read as the rollout's first work list.

A pilot that ends in a clear no has done its job. It cost you weeks. A workflow kept alive on hope costs quarters.

Takeaway

Run the new way beside the old, judged by the people who do the work against the counted baseline, with the pass mark set before you start. Stopping is a legitimate result.

The pilot passed. Now the new way has to become the way the work is done, and this is where many rollouts quietly fail: the new tool gets added on top of the old routine, every old step survives, new checks grow around it, and the workflow ends up with more steps than before. The alternative is to rebuild: remove the steps that only existed because a person did the work by hand. A workflow counts as rolled out when the new way is the default, not an option.

Rolled out means the default, not an option.

Expect the first weeks after rollout to run slower than the pilot suggested. The whole team is now learning what the pilot group already knew. Budget for that dip instead of reading it as failure. It is the cost of the last mile, not a sign the workflow is wrong.

The mechanism
The routine, rebuilt
What happened to each step of the proposal routine when the new way became the default.
Kept
Enquiry arrives and is logged
a pass starts the same way it always did, now by rule instead of by hand
Removed
The Monday meeting where drafts were assigned
it only existed because drafting was manual; the drafts now exist before the meeting would have started
Changed
Associate assembles the draft → associate reviews the draft
the hour that went into assembling now goes into reading, checking and sharpening
Not added
A second review of the draft itself
an add-on rollout would have created one; the partner's review already exists and stays the single check
Kept
Partner reviews and prices
the sign-off this firm chose to keep
Added
A scheduled look at the workflow's output
automated work drifts as inputs, tools and models change; unwatched, it degrades quietly
A useful check: the rebuilt routine has fewer steps than the old one. More steps than before means the tool was added on top.
Rebuilt means steps removed, one step changed in kind, and one watchful step added. It does not mean the old routine with a tool on top.
The moves
01

Set the default date

From a named date, the new way is how the workflow runs. The old way remains only as the documented fallback.

Close the choice

As long as both routes stay open, the busy week picks the familiar one. The announced date closes the choice: from then on, a new case goes down the new route without anyone deciding that it should.

02

Remove the steps that manual work required

Walk the map once more and ask of each step: does this still need to exist now that the drafting runs on its own?

Coordination built around waiting goes first

Assignment meetings, chasing lists, status check-ins. These existed to coordinate manual work, and each one removed gives time back to the team.

Name the steps that changed in kind

The associate's hour moves from assembling to reviewing. Say that clearly during the rollout. People should hear that their step changed shape, not that it disappeared.

03

Give it a champion beside the owner

One person on the team who uses the workflow daily, answers colleagues' questions, and collects what annoys them.

A lighter job than it sounds

A few questions a week, not a new role. It is how new routines survive a busy month: small questions get answered right away instead of turning into reasons to go back to the old way.

Champion and owner are different jobs

The champion lives inside the workflow and speaks for the team. The owner holds the map, the numbers and the decisions. In a small firm one person may do both, but the two jobs still exist.

04

Point the measures at the new way

If people are still judged by numbers that reward the old way of working, the old way comes back quietly.

Judge the workflow by its measured outcome, the number the next article tracks, and check that nobody is scored on something the rollout deliberately removed. A team praised for the volume of hand-written drafts will keep writing drafts by hand.

05

Watch it, and keep the way back

Someone checks the workflow's output on a schedule, and the owner can switch back to the documented old way at any time, without calling a meeting.

The default test
A new case today goes down the new route without anyone deciding that it should.
The absence test
The workflow still runs when the champion is on holiday, because written instructions, not one person's memory, carry it.
The rollback test
The owner knows exactly how to switch back, and the old way is documented well enough that switching back actually works.

Pass all three and the rollout is real. Fail one and the workflow is still an experiment with a rollout's name.

Takeaway

Rolled out means the new way is the default, someone owns its health, and the way back exists. Rebuild the workflow around what runs on its own; don't bolt the new tool onto the old routine.

The workflow runs. The last move decides what it earned. The tempting measure is adoption, how much the new way gets used, because it is easy to count and comfortable to report. But what counts is the verified outcome: a result that was accepted and used, like the draft that went to the client at the quality the firm actually ships. Count verified outcomes against the baseline, and let that number make the keep, revise or retire call.

A busy tool is not the same thing as a better workflow.

Ask the team how it feels and you will get an answer, just not a measurement. People's sense of saved time is unreliable in both directions. The decision comes from the counted numbers. The team's experience still matters, as context for the revise decision, not as the measure.

What you're building
The review page
The filled-in form that makes the keep, revise or retire call. This sample shows a workflow that earned its keep.
Sample output — yours will look like this
Workflow review: proposal drafting
One page · reviewed on the scheduled date · next review already in the calendar
Before and afterThe baseline week from the map, next to the same measures from this period. Both sides come from counting: the baseline week then, the workflow's own records now.
What counted as a successA draft that passed the partner who signs off, at the quality we actually ship. Drafts sent with light edits count. Drafts rewritten from scratch are listed separately, with the reason.
The full costThe tools' price for the period, plus the human time still inside the workflow: preparing what the AI step reads, reviewing what it produces, correcting what it got wrong.
Cost per successCost divided by successes, set against the same figure from the baseline week. A good result now costs less than it did in the baseline week, in hours more than in money.
Decision
Keep Revise Retire
What we would do differentlyTwo or three lines, written while the experience is fresh. This is what makes the second workflow cheaper than the first.
Owner
Date
Next review

The workflow is this firm's; the form transfers as it is. Whatever your firm automated, these six entries are the whole file.

One page per workflow, kept current. When Stage 3 grows the portfolio, a page like this exists for every workflow that runs, and the decision line keeps the portfolio honest.
The moves
01

Count the successes

Verified outcomes over a set period, taken from the workflow's own records, not from memory.

Define accepted before you count

A draft counts when it passed the person who signs off, at the quality the firm actually ships. Drafts that go out with light edits are wins. Drafts rewritten from scratch are counted separately, and their reasons feed the revise pile.

02

Count the full cost

The cost is the tools' price for the period plus the human time still inside the workflow.

The time hides in three places

Preparing what the AI step reads, reviewing what it produces, and correcting what it got wrong. The review hour is part of what a result costs. Leaving it out flatters every workflow equally.

03

Divide, then compare

Cost divided by successes, set against the same number from the baseline week. The workflow wins if a good result now costs less, in money or in scarce hours.

Both sides of the comparison come from counting: the baseline week priced the old way of working, and the measured period prices the new. If the new number is not clearly better, that is what the revise decision is for. The log from the pilot usually says where the cost is hiding.

04

Decide on a date

Keep, revise or retire, decided on a date that went into the calendar when the rollout ended, not whenever someone remembers to ask.

Keep
It beats its baseline. It stays, and its page stays current.
Revise
It almost does. The pilot's log usually says how, and the revision gets its own review date.
Retire
It does not. A retired workflow is not wasted; it taught you what the next one needs.
05

File the dossier

One page: the before and after, the decision with its date, and the lines about what you would do differently, written while the experience is fresh.

With the page filed, the firm has run the whole loop once: picked, mapped, routed, piloted, rolled out, measured. Everything the loop produced, from the map to this page, is reusable the next time it runs.

Operating well at this level is a destination in itself. Stage 3 is there when the next constraint genuinely needs machines acting unattended, not because a next stage exists.

Takeaway

Count verified outcomes, not usage, and price them against the baseline. The number, not the mood, makes the keep, revise, or retire call.

Closing this stage
Operating well at this altitude

Many firms should operate here, and some should stay. Operating well at Stage 2 looks like: the workflow runs as the default, its dossier is current, its number gets checked monthly, and no heroics are involved. Stage 3 is there when the next constraint genuinely needs machines acting unattended, not because a next stage exists.

Stage 3 · Expansion

Let automation and agents carry more of each workflow

You have won once. Now the loop runs again and again, and the route ladder extends up into automation and agents, with the rule that saves the most pain: rules before agents.

The articles

Stage 3 lets machines act without someone watching each run, and what they act on is your systems and your data. At Stage 1 a connection acted as you, inside your own accounts, and the blast radius was your own inbox. What runs unattended needs more: its own login and its own narrow permissions, so what it can touch is decided, recorded and revocable. The difference is not the tool; it is the accountability around it. This article gets both ready: grant the least access that does the job, and fix unfit data before you build on it.

Machines act at machine speed; readiness is what makes that a feature.

The Stage 2 map already did half the work here. It marked every point where a person quietly repairs bad records on the way through. At this stage that marking becomes law, because a machine repeats the error faithfully, at speed, in every run.

The mechanism
The data sort
Every dataset a build touches lands in one of three piles, and the pile decides what happens next.
Clean
Build on it now
Accurate, current, consistently structured.
The template folder in SharePoint: current versions, one naming pattern.
→ The build reads it as it is.
Improvable
Standardize in the build
The right data in a messy shape.
Client records where half the phone numbers carry country codes and half do not.
→ The build includes the clean-up step.
Unfit
Fix first
Wrong, stale, or contradictory.
Two systems that disagree about which contact is current.
→ Blocks the build until it is fixed.
Automation does not clean data; it industrializes it. A machine repeats the error faithfully, at speed, in every run. That is why unfit data blocks the build instead of being fixed later.
The sort is quick to do and expensive to skip. The quiet fixes the Stage 2 map recorded tell you exactly where to look: every one of them is a pile-two or pile-three finding already made.
The moves
01

Give every build its own identity

Each automation gets its own login rather than borrowing whoever built it, so the access survives that person's departure and can be switched off without touching anyone's account.

The proposal build shows why

It reads the CRM and the template folder under its own login, and nothing else. When its builder changes jobs a year later, nothing about the automation changes, and nobody has to remember what was connected through her account.

02

Grant the least access that does the job

The invoice workflow reads invoices, not the whole accounting system. Grant access like it will be audited, because one day it may be.

Never let one system hold all three

Access to private data, exposure to content outsiders can write, and a way to send data out. Drop any one of those three legs and the worst case collapses. It is the single most useful scoping rule at this stage, and it costs nothing to apply.

03

Sort the data into the three piles

Clean, improvable, unfit, as in the figure. Nothing runs on the unfit pile.

Start from the map's quiet fixes

Every point where the Stage 2 map recorded a person correcting data on the way through is a finding: the stale rate, the wrong address, the empty field. Each one names a dataset and tells you which pile it belongs in.

04

Build the context the firm owns

Stage 1 gave your personal AI its world; this stage gives the firm one. The firm-scale version is deliberately low-tech: a small set of plain documents the firm owns, in a folder the firm controls.

What goes in

What the firm does and for whom, how it writes, its service conventions, the decisions people keep re-explaining. One routing file on top tells any tool where to look first. A "how we write" file, a services file, one page of conventions per major client.

Why plain documents

Because the files are ordinary documents rather than any vendor's memory, the asset survives every change of tool and model underneath it. Every workflow in this stage starts informed instead of briefed from scratch.

Context ages the way data does, so it gets the same discipline as everything that runs unattended: a scheduled job folds each week's new decisions in and retires what has expired, a periodic check catches duplicates and contradictions, and one named person approves what changes. Without that maintenance, a shared context layer becomes a contradictory, expensive mess within a quarter.

05

Ask four questions before any build in this stage

Four questions, asked before anything gets connected. An afternoon spent here saves the weeks that follow a build on sand.

1What can it see? The scopes are listed and no wider than the workflow needs.
2Who is it? Its own credentials, revocable without touching anyone's personal account.
3What is it standing on? The data it touches is sorted, and nothing runs on the unfit pile.
4What does it know? The context files exist, are current, and someone owns them.
Takeaway

Least access that does the job, an identity of its own, no builds on unfit data, and a context layer the firm owns. Machines act at machine speed; readiness is what makes that a feature.

The fastest wins in this stage are not the cleverest ones. The steps that never change go first, for three reasons: the rule already exists, on a checklist or in someone's head, so automating it is transcription rather than invention; they fail loudly and fix cheaply, because a file in the wrong folder is visible and reversible in a way a wrong judgment is not; and every step a plain rule handles costs almost nothing per run, which starts to matter once the platforms begin metering your volume. Automate the steps that never change before the ones that think.

Automate the steps that never change before the ones that think.
Example · a real one

A large delivery company automated one boring internal task: account-unlock requests. Around eight hundred a month, each dropping from thirty-five minutes of a person's time to twenty, roughly two hundred hours a month recovered, and the build took five hours on a standard automation platform. No cleverness anywhere: plain rules, with manager approval kept exactly where it was.

The mechanism
Sized by what happens if it breaks
Every build in this stage lands in one of three bands, and the band sets the rules it is built under.
Read-only
Dashboards, digests, morning summaries
If it breaks, someone sees a stale number.
Build these freely; they are how the firm learns the platform.
Internal records
Trackers, bookings, handover notes
A mistake costs an internal correction.
Build with the harness from article four in place.
Client-facing
Anything that leaves the building
A mistake reaches a client under your name.
Runs in parallel against the current way first, exactly per the Stage 2 pilot, before a client ever sees it.
The proposal workflow's own extensions follow the same bands: the pipeline digest built freely, the filing and chasing rules built with the harness, and anything client-facing earning its place through a parallel run.
The moves
01

Start with the steps around the build you have

The proposal workflow from Stage 2 is surrounded by predictable work that never made it into the first build: the filing, the logging, the chasing, the weekly digest.

Transcribe the checklist, don't invent

These rules already exist, on a checklist or in someone's head. Automating them is writing down what the firm already does, which is why they go first and why they rarely surprise anyone.

02

Know your billing unit

The platforms meter volume in different units, and the unit matters more than the sticker price.

1Per task. A charge for each step executed: cheap for short workflows, adds up on long ones. Built-in helper steps are often free.
2Per credit. A standard step costs one credit, but AI steps consume by size and complexity, so the bill tracks what the AI actually does.
3Per execution. A charge per workflow run, however many steps it contains: predictable for long workflows.

The same workflow can be cheap on one meter and expensive on another. Model a month of your real volume on each before committing.

03

Size the blast radius before you build

The bands in the figure set the rules: read-only freely, internal records with the harness, client-facing through a parallel run first.

The band is about failure, not importance

A small client-facing automation outranks a large internal one, because the question is what happens if it breaks, not how much it does.

04

Keep the familiar front end, and know the bulk boundary

Two practitioners' rules of thumb, useful precisely because they are unglamorous.

Let the machinery change, not the screen

If the team lives in a spreadsheet, let the new workflow read and write that same spreadsheet instead of forcing a new screen. The machinery changes underneath while the habit survives.

Know where chat-with-review stops

Working through a list inside a chat tool, with a person reviewing, is comfortable up to somewhere around one or two hundred items a run, at real usage cost. Past that, or the moment nobody reviews each item, the job belongs on an automation platform built for volume. Rules of thumb, not physics.

05

Put generative production behind a brand gate

The safe first taste of generative AI inside automation is production work: one approved brief becomes many drafts, variants or formats, produced by a model inside the workflow.

What makes it safe is the brand gate: a person or a hard checklist that everything passes before it leaves the building. The shape is predictable even though the content varies, which is why it belongs with the predictable work.

06

Notice when you have outgrown the platform

You have outgrown a platform when you spend more time working around it than in it: workflows split in two to fit its limits, exception handling that fights the tool, costs that climb faster than volume.

That is a normal milestone, not a mistake, and the map you keep from Stage 2 moves with you.

The volume is real
It runs often enough that the build pays back.
The meter is understood
You know what unit you pay in and roughly what a month costs at your volume.
Failure is visible
When it breaks, something tells someone. Silent failure is the expensive kind.
Takeaway

Automate the steps that never change first: they are specified, cheap to run, and cheap to fix. Know your billing unit before the bill teaches it to you, and size every build by what happens if it breaks.

An agent is the first thing you will run that chooses its own path through the work. One word, two meanings, so let's separate them: the popular beginner courses use "agent" for a chat persona with standing instructions, and if you have built one of those, that was real work; keep it. This stage means something with more consequence: software that acts in your systems, toward a goal you set, choosing the order of its steps within your rules, without you watching each pass. The first one gets a narrow job, real boundaries, and a suggest mode to prove itself in.

An agent is a job description, not a science project.

The support inbox has been waiting for this since Stage 2: it sat on the shortlist marked "kept for later", because the right reply varies too much for rules. That varying shape is exactly what an agent is for, and inbox triage is the classic first agent job: constant volume, quick to check, cheap to correct.

What you're building
The agent's job description
Written before the agent exists. The definition is the work; the capability comes built into the platform.
Sample output — yours will look like this
Job description: support inbox triage
Written before the first run · owner named · starts in suggest mode
The job

Read each incoming ticket, label it by category, and propose the reply. Route anything it cannot place to a person. One kind of work in one place: triaging this inbox, not "handling support".

The boundaries

Reads: the support inbox, the client list, and the answer library in SharePoint. Nothing else. Scoped in the platform itself, not just described in the prompt.

What always goes to a person

The firm's own written list. This firm routes complaints, anything mentioning a deadline or legal step, and every sender not on the client list. However confident the agent is.

What a right answer looks like

Ten real tickets, labeled and answered the way the team would, attached as examples. The agent is graded against these, not against a feeling.

The mode it starts in

Suggest: it labels each ticket and proposes the reply; the team sends it or fixes it. Acting on its own is earned later, category by category.

The scorecard

A running tally of right, wrong, and escalated, kept per category. This becomes the promotion file the next two articles run on.

The systems and the routing list are this firm's; swap in your own. What transfers is that every section is written down before the agent exists.

The same shape as the delegation brief from Stage 1 and the saved instruction from Stage 2, one size larger: a goal, boundaries, sign-off points, and examples of right answers.
The moves
01

Decide whether before which

The platforms you already use offer agents as a built-in option, so capability is not the question. The first design question is whether this work should be delegated to an agent at all.

Two kinds of work fail the gate

Work you cannot check, because you would never know whether it was done right. And work a fixed rule already handles, because an agent adds variability where none was needed.

02

Pick a job that is narrow, frequent, and checkable

One kind of work in one place. Agents with sprawling toolkits choose badly; small and focused wins.

Narrow
Triaging the support inbox, not "handling support".
Frequent and checkable
It runs daily, and a person can tell at a glance whether a given decision was right.
Low-stakes on error
A mislabeled ticket costs a correction, not a client.
03

Rehearse with the persona you already have

If you built a chat persona in Stage 1, it makes a useful rehearsal space, at zero stakes.

Give it a mandatory stop phrase and make it end every session with a short report of what it did and where it struggled. You have just rehearsed ceilings, off-switches and logging, the whole grammar of the next article, before anything real depends on them.

04

Write the job description before the agent exists

The goal, the boundaries, the sign-off points, and examples of right answers, as in the figure. The definition is the work.

Boundaries live in the platform, not the prompt

The systems it may use and the data it may see are scoped in the platform itself. A prompt describes intentions; a scope enforces them.

Attach real examples

Ten tickets labeled and answered the way the team would. Right answers you can point at beat qualities you can only describe.

05

Launch in suggest mode, with the scorecard running

In suggest mode the agent drafts the decision and a person confirms it: it does the reading and the routing work, while every outcome still passes a human hand.

Act mode is earned, not assumed

A category moves up when its suggestions have stopped needing correction. The path from suggest to act, and the doors that never open, are the next two articles.

Tally from day one

Right, wrong, escalated, per category. The promotion decisions ahead need this record, and it cannot be reconstructed later.

Takeaway

First agent: one narrow, frequent, checkable, low-stakes job, launched in suggest mode with its job description written first. Capability is easy to get; the definition is the work.

Everything in this stage that runs unattended needs a structure that keeps it dependable. Errors in autonomous runs compound: a bad night, unwatched, is an open-ended cost unless something bounds it. The structure is the harness, and it has four parts: ceilings, an off-switch, a logbook, and one page of instructions for the colleague who is not you. Give every unattended run a ceiling, a logbook, and an off-switch.

The harness is the product.

Ceilings, logs and off-switches are what turn something that works in a demo into something the firm can lean on. None of it needs a security team. It needs an afternoon of configuration and the habits this article describes.

The mechanism
The harness around the triage agent
The four guards, wired around the agent from the last article. Each one is configured, not remembered.
Before it runs · ceilings
The limits are set in the platform
A spend it cannot exceed, a time limit per run, a cap on steps and retries. A bad night becomes a bounded cost instead of an open one.
While it runs · the logbook
Every run leaves an auditable trace
What it read, what it decided, what it did, attributed to the agent that did it. The triage agent ends every night with a table: ticket, label, action taken, and one plain line at the bottom: "3 tickets skipped: sender not in the client list." That skipped line is the practice working: obstacles reported, not improvised around.
When something looks wrong
Alert, named person, off-switch
An alert reaches the owner; the off-switch is one known way to stop it now, that works, that more than one person can reach; and the documented old way from the Stage 2 rollout is still there to fall back on.
On a schedule · the health check
A person reads a sample of runs
And asks one question: is this still doing what we meant? Drift arrives quietly, as inputs, tools and models change underneath the workflow. The health check is a habit, not a system.
The same harness fits every unattended build in this stage, agent or not. What changes per build is the size of the ceilings and the schedule of the check, both sized to the blast-radius band from article two.
The moves
01

Set the ceilings

A spend it cannot exceed, a time limit per run, a cap on steps and retries. Ceilings turn a bad night into a bounded cost.

02

Wire the off-switch and write the one page

One known way to stop it now, that works, that more than one person can reach. And one page of instructions, written for the colleague who is not you: what this is, what normal looks like, what to do when it is not, who owns it.

03

Log in a shape you can audit

Have the agent produce structured output: a table or a labeled list an owner can audit line by line, rather than prose that can hide a wrong call inside a nice paragraph.

Obstacles reported, not improvised around

"Could not find the May invoice" is a good result. A guessed invoice number is not. Require the report, and read it.

04

Keep the toolkit minimal, and vet what you install

Every tool the agent holds is something it can use at 2 a.m. with nobody watching. Give it only what the job description needs.

Shared skills and plugins are software from the internet

Read what one does before installing it. Skills can run tools and execute code; treat them with the care you would give any software the firm adopts.

05

Fence the computer-operating experiment

An agent that operates a computer directly, screen, mouse and keyboard, gets its own fence: isolate it like a contractor on day one.

Its own user account, a machine that holds nothing else, access granted surface by surface, and a hard payment ceiling if a card goes anywhere near it.

06

Schedule the check, and decide who operates it

Someone runs this machinery, and deciding who is part of hardening. On a managed platform the vendor keeps it alive while you own the definitions and the checks; running it yourself buys full control of your data and of cost at volume, and it is a real operating job: updates, backups, monitoring.

Data sensitivity, what you already run, volume, and who would honestly maintain it make the call. The full where-should-it-run question gets its own treatment in Stage 4.

The overnight test
If it misbehaves at 2 a.m., the ceilings cap the cost and the log shows what happened.
The absence test
Someone other than the builder can read the one page and act, including switching it off.
The drift test
The next health check is in a calendar, and the last one actually happened.
Takeaway

Ceilings, off-switch, logbook, one page of instructions, and a scheduled look at what it is actually doing. The harness is what makes unattended work dependable.

How much should the system be allowed to do on its own? Most firms answer by feel: a good week builds confidence, a bad morning revokes it, which means the answer is really a mood. This article replaces the mood with a ladder: autonomy promoted category by category, on performance you can show, plus a set of doors you decide, in advance, to keep human.

Every new capability starts at suggest, where the agent proposes and a person decides, however good the demo looked. A category that keeps earning it climbs to act with review and then to act and report; if corrections rise after a promotion, it moves back down. And some doors never enter the ladder at all: not because the numbers are bad, but because you decided in advance they stay human. Which doors is a business judgment, and it is yours. Many firms choose what clients see, movements of money, and the irreversible. Your list may be shorter or longer; what matters is that it is written before the first promotion, so it never has to be argued case by case.

Demotion is maintenance, not drama.
The mechanism
The autonomy ladder
Three rungs a category climbs on numbers, and the doors that never enter the ladder.

A category climbs on a low, stable correction rate — measured against a bar written before it started.

1 · Suggest
it proposes · a person decides
every category starts here
↑ climb
2 · Act with review
it acts · a person checks shortly after
undoing is easy
↑ climb
3 · Act and report
it acts · the log tells you what it did
scheduled spot checks
← corrections rise: the category steps back down a rung, for a set period
Outside the ladder

The doors you chose to keep human never enter it — exempt by your decision, not by performance.

Trust climbs one category at a time, on numbers. The correction rate decides the rung; you decide the doors.
Example

Password-reset tickets ran six weeks in suggest mode with corrections near zero, earned a month of act-with-review, and now act and report. Refund requests live in the same inbox and score just as well, and they stay in suggest mode anyway, because this firm put movements of money behind a human door. That is not a judgment on the technology; it is a choice about the business.

The moves
01

Cut the work into categories you can score

The ladder works per category, not per agent, so the first job is deciding what a category is.

Group by "same kind of request, same kind of answer"

In a support inbox: password resets, renewal questions, refund requests. Each narrow enough that one correction number means something, and each holds its own rung.

Too broad shows up as noise

If a category's corrections swing week to week for no visible reason, it is probably two categories wearing one label; split it.

02

Write the bar — and your doors — before you start

Both are set in advance, in writing, so promotion day is a check against a page, not a debate about a feeling.

What counts as a correction
Edited before sending, rejected, or undone after. Anything a person had to change, close calls included.
The bar and the period
How low, for how long: weeks rather than days, through at least one busy stretch. One written line per category.
Your permanent doors
The actions your firm keeps behind a person regardless of performance. Your call, written next to the bar.
03

Make corrections countable

The ladder runs on the correction rate, so the review step has to leave a countable trace.

Run suggest mode through a visible review step

The person approves, edits, or rejects each suggestion; that action is the log. Most platforms record it, and where one doesn't, a per-category tally does the job.

Keep the count per category

A blended number can hide one category quietly failing inside three that are fine. This is the scorecard from your first agent, continued: the promotion file.

04

Hold a short promotion review

Promotions happen on a schedule, not on impulse: a short, recurring look at the numbers, category by category.

Under the bar?
The correction rate, for the written period. Not one good week.
Harness held?
No ceiling breaches, no silent failures in the period.
Doors untouched?
The promotion leaves your permanent doors exactly where they were.

Then promote one rung, one category at a time, and log the decision with a date. The list Stage 4 keeps of everything that runs is its natural home.

05

Wire the doors you chose

Permanent doors are configured, not remembered: built into the workflow as required human steps that no promotion can remove.

Each door becomes a required approval

Whatever your firm put on the list routes to a person at every rung. Sent under your name means certified by a person.

If the platform can't enforce the stop

Then the category isn't eligible for promotion at all; a door only counts if it cannot be walked around.

06

Demote without ceremony

When corrections rise, the category moves back down a rung for a set period: no meeting, no post-mortem, no blame. Demotion is the ladder working, not failing; a system that can only ever gain autonomy isn't earning trust, it's accumulating risk.

Takeaway

Autonomy is earned per category, against a correction-rate bar set in advance, and some doors stay human by design. Demotion is maintenance, not drama.

Expansion is not one big program; it is the same loop, run again on the next constraint. The map from Stage 2 is the brief for every build in this stage: record the work as it really happens, count a real week, name one owner. Two sentences carry the whole discipline: map the real process, and no baseline, no build.

Every workflow gets the full loop, and everything that runs gets watched.

Growth changes the failure mode. One workflow that breaks is an incident; ten automations without monitoring are ten quiet degradations racing to be discovered by a client. The monitoring habit from the harness article is what makes ten manageable.

What you're building
The portfolio board
The same loop at different points, on one page. This is what growing well looks like a few months in.
Sample output — yours will look like this
Measured · kept
Proposal drafting
the review page said keep; its number gets checked monthly; owner named, one page of instructions current
In suggest mode
Support inbox triage
the agent proposes, the team confirms; the per-category scorecard counts toward the first promotion review
Being mapped
Supplier invoice checking
the recording is done and certified; the baseline week is counting now; no baseline, no build
Picked next
The monthly client status report
still on the shortlist from the first win; waits its turn rather than running as a side project
Running · watched
Friday pipeline digest
read-only band; a scheduled look sized to its stakes, because a broken digest looks exactly like a quiet week

Five entries at one firm, a few months into the stage. The board is one page. When keeping it current becomes its own job, Stage 4 has arrived.

Every entry carries an owner, a place in the loop, and a way of being watched. The current, owned list of what runs is the door into Stage 4.
The moves
01

Choose the next constraint like the first

The next workflow is chosen the way the first one was: where work waits now. The shortlist from the first win is still the list.

Wins compound; scattershot doesn't

One constraint at a time keeps each build standing on the last one's assets: the map habit, the context files, the platform experience. Scattershot builds compound maintenance instead.

02

Run the loop uncut

Route, pilot against the baseline, harden, measure. The steps that feel skippable on workflow five are the ones that fail on workflow six.

03

Decide build, buy, or hire, per workflow

Some workflows a standard platform covers, some arrive as finished products, and for some the right answer is a specialist who builds and hardens it to fit. The route logic from Stage 2 decides, workflow by workflow, never as a firm-wide policy.

04

Watch everything, with no exception for the small ones

Unmonitored automations degrade quietly, and the cheap ones degrade quietest, because nobody is looking.

The Friday pipeline digest broke in March and nobody noticed until May, because a missing nice-to-have looks exactly like a quiet week. The fix took an hour. The two months of decisions made without it did not.

An alert or a scheduled look, sized to its stakes

The client-facing build gets an alert the moment it fails. The digest gets a monthly glance. What nothing gets is no watcher at all.

05

Keep the list of what runs current

The portfolio board from the figure, kept honest: no orphans, nothing unwatched, nothing that neither pays for itself nor gets retired.

Owned
Every automation has an owner and one page of instructions. No orphans.
Watched
An alert or a scheduled look, sized to its stakes.
Earning
Everything paid for itself or was retired. The measuring keeps running per workflow.

When enough runs that keeping the list current is itself a job, you have arrived at the operating engine. That is Stage 4.

Takeaway

Grow one constraint at a time, run the whole loop each time, and watch everything you leave running. The current, owned list of what runs is the door into Stage 4.

Closing this stage
Operating well at this altitude

Operating well here is a complete destination. It looks like: several workflows running under real harnesses, an autonomy log that says what earned trust, and monitoring that catches drift before clients do. Stage 4 is there when keeping the whole picture current becomes its own job, not because a next stage exists.

Stage 4 · Operating engine

Make AI a dependable part of how the firm runs

What changes when AI stops being a project and becomes something you rely on. What this stage builds is light: documents and routines, not platforms, and far lighter than the governance genre implies. Most firms have none of this, which is exactly the opportunity.

If you self-located straight to this stage, here is what the engine gathers up: the personal collections of prompts and instructions become one shared library, the notes on each automation become one list, the guardrails around each build become one page of rules, and the measuring of each workflow becomes one review of the whole. Each earlier stage teaches its piece in full; this stage only assumes they exist somewhere, however informally.

The articles

The operating engine starts with two short documents. The policy says what your firm does and does not do with AI, in six short sections. The register is the single list of everything that runs, one entry per automation. The policy is the rules; the register is the reality. Together they answer the two questions every serious client, insurer or regulator asks: what do you allow, and what actually runs?

If it runs, it is on the list.

Most firms have none of this, and not only small ones: ask a leadership team anywhere what AI actually runs in their business today, and few can answer with confidence. The governance literature makes the fix sound like a program with a steering committee. It is an afternoon: six sections on one page, one list, one review date. Light is the point, because heavy governance dies of neglect.

What you're building
The register
One list, everything that runs, a name on every entry. This sample carries the firm the playbook has followed since Stage 2.
Sample output — yours will look like this
RunningProposal drafting
What it does
Writes the first draft of each proposal from the firm’s intake notes
Runs by itself
Writes the first draft
Goes to a person
Every draft is checked before it leaves the firm
Responsible
The operations lead
What it touches
The CRM, the proposal templates, the rate sheet
Where it runs
As a service, at the provider
Instructions
On the shared drive
Next review
Quarterly
RunningSupport inbox triage
What it does
Labels every incoming message and proposes replies
Runs by itself
Labels every message and sends the routine replies
Goes to a person
Complaints, legal questions and anything new go to the support lead
Responsible
The support lead
What it touches
The inbox, the client list, the answer library
Where it runs
As a service, on the automation platform
Instructions
On the shared drive
Next review
Monthly, while it is still earning new categories
RunningSupplier invoice checking
What it does
Checks each supplier invoice line by line and explains any variance
Runs by itself
Matches every invoice line
Goes to a person
Every variance goes to the finance lead
Responsible
The finance lead
What it touches
The accounting system, with access to invoices only
Where it runs
As a service, at the provider
Instructions
On the shared drive
Next review
Quarterly
RunningFriday pipeline digest
What it does
Writes a short summary of the week’s pipeline for the Monday meeting
Runs by itself
It only reads and reports, so nothing needs a sign-off
Responsible
The operations lead
What it touches
The CRM and the calendar
Where it runs
As a service, at the provider
Instructions
On the shared drive
Next review
A scheduled look, monthly
RetiredMeeting-notes summaries
Retired at its second review because nobody used the output. Its brief and process notes went to the shared library first, and the entry stays here, dated, so the list stays honest.
Sources of truth — who may change them
Owned“How we price”
Feeds three automations. Changes go through the operations lead, so all three change behavior on the same deliberate day instead of drifting one by one.

The entries are this firm's; the labeled fields are the part to copy, exactly as they appear here. Keep the list wherever the firm keeps its working documents; what matters is that there is exactly one.

Anyone can open it and know what runs, who looks after each one, and which ones were retired. When an insurer or a client asks what touches their data, this list already says.
The moves
01

Write the one-page policy

What your firm does and does not do with AI, in six short sections.

1Scope. What this page covers, and for whom.
2Approved tools. Listed by name; anything else is a question for the person in section five.
3Data that never gets pasted. This firm's line: client financials never go into consumer tools.
4Where a person signs off. The approval line the firm chose, stated once, firm-wide.
5Who fields questions. A name, so questions have somewhere to go.
6When this page is reviewed next. A date, honored.
02

Build the register

One entry per automation, seven fields each, as in the figure. No automation without an entry, and new builds start by adding one.

One list, not several

Keep it wherever the firm keeps its working documents. The place matters less than there being exactly one, because two lists disagree the week after they are created.

03

Decide what each one does on its own

Every automation needs one more decision: which parts of the work it does by itself, and which parts it hands to a person. The answer goes on its register entry.

There are two easy settings, and both waste the automation. If it does everything by itself, the firm carries risk nobody agreed to. If a person has to approve every single thing it produces, that person becomes the bottleneck, and most of the time saved is lost again. The setting that works for most work sits in between: the automation handles the routine cases by itself and hands the unusual ones to the person named on its entry.

The line moves as trust grows

Start careful. When the reviews show corrections are rare, let it handle more by itself. Move the line at the review, based on the numbers, and update the entry so the register always says where the line sits today.

Example

The support inbox triage from Stage 3 began by proposing replies for a person to send. Because corrections stayed rare, it now sends the routine replies itself, and still hands complaints, legal questions and anything it has not seen before to the support lead.

04

Put one name on every entry

An automation nobody is responsible for runs until it breaks, and then fixing it is nobody's job, which in practice means it stays broken or quietly keeps producing the wrong thing.

One person, who knows the workflow and answers for it

When that person leaves, the entry gets a new name the same week or the automation gets retired. No third option.

05

Name who may change the sources of truth

The documents your automations rely on, like a price list or a description of your services, get the same discipline: name who may change them, because a quiet edit there changes what everything built on them does.

The "how we price" document feeds three automations, and the register names the operations lead as its owner. When a partner wants a rate changed, the change goes through her, and all three automations change behavior on the same deliberate day instead of drifting one by one.

06

Keep it light, and check it is working

The biggest software vendors have started building this same kind of inventory into their own products, because even they found that nobody could reliably say what was running. Yours can be far simpler and do the same job.

Someone owns the documents
The policy's review date has been honored at least once.
The list matches reality
Pick an entry, find the automation; pick an automation, find its entry.
They get used
Questions actually reach the named person, and new builds start by adding an entry.
Takeaway

One page of rules, one list of what runs, both owned and current. This is an afternoon's work, and it puts you ahead of most firms of any size.

Every AI tool your firm uses does its work somewhere: on a provider's computers, or on your own. Your email almost certainly runs at a provider; your files might, or they might sit on a machine in the office. Nobody agonized over those choices; each followed a real need. AI is the same decision in new clothing, and by this stage the choices add up to how much control, cost and dependency the firm carries. The rule: for everything that runs, you can say where it runs and why.

Make it a decision, not an accident.

Most tools are services you rent: the provider runs everything, you start through an account, and your protection is the contract. Some can run on computers you own: real effort to set up, someone has to keep it healthy, and in return nothing leaves machines you control. Between the two sit real middle options, and most firms end up using several at once. The point is not to pick a camp; it is to match each automation to the protection it actually needs. Paying for maximum control everywhere means paying for a constraint you do not have.

What you're building
The placement page
One page that says, for everything that runs: where, why, whose job when it breaks, and how the firm would leave.
Sample output — yours will look like this
Where everything runs — and how we would leave
One page · read against the register · reviewed when a provider changes something
The AI assistant and the automation platform · as a service Everyday work with no hard constraint. The plan we are on decides what the provider may do with our data; when it breaks at night, that is the provider's problem. Leaving: we have exported our workflows once to know it works.
The support triage agent · as a service, on the platform Runs where the platform runs, inside the access we granted. Its job description and its instructions live on our shared drive, so rebuilding it elsewhere starts from our documents, not from memory.
The confidential-documents model · on our own server The one hard constraint we actually have: documents that may never leave the building. Someone at the firm is named for keeping it healthy, and was named before it first broke, not during.
The firm's knowledge files · ordinary files, in the folder we control Readable by any tool, movable anywhere. When we change assistants, the new tool reads the same folder and knows the firm on day one.
If we had to work by hand for a week Each automation's instructions end with one short paragraph: how to export it, and how the work gets done by hand from the same checklist it was built from.

The placements are this firm's; the columns are the part to copy: where, why, whose job, and the way out. Most firms' honest page looks like this one: nearly everything as a service, one or two hard constraints handled deliberately.

Everyday work runs comfortably as a service. Work with a hard constraint, like client confidentiality or a rule about where data may be stored, earns a more controlled option: a private space at a large cloud provider, a tool configured so data stays in your own systems or your own country, or software on machines you own.
The moves
01

Match each automation to the protection it needs

Everyday work as a service; a hard constraint earns a more controlled option. The right answer depends on the work, not on fashion.

The middle options are real

A private space at a large cloud provider, where the software runs on their machines inside a walled-off area only your firm can reach. A provider's tool configured so your data stays in your own systems, or in your own country. Most firms end up using several options at once, deliberately.

02

Keep the firm's knowledge in files you own

The documents your tools read before doing any work deserve their own decision, because every tool depends on them.

Ordinary files, in a folder the firm controls

Ordinary files can be read by any tool, moved anywhere, and survive every change of provider. Keeping that knowledge inside one tool instead, in its settings or workspace, is more convenient day to day, and it quietly ties you to that tool: switch providers, and the firm's accumulated knowledge has to be rebuilt from scratch.

03

Ask: when this breaks, whose job is it?

For a service, the provider's. For anything you run yourself, someone at your firm, and that someone is named before the breakage, not during it.

04

Ask: how would we leave?

Providers raise prices, retire products, and get bought. Leaving calmly takes three things you can arrange today.

The knowledge stays yours
The firm's files live in a folder you own and would move untouched.
The export is tested
You have tried getting your work out of each tool once, so you know it works.
The hand-run is written
Each automation's instructions say how the work would get done by hand for a week if it had to be.
05

Run the three questions against the register

Pick any automation on the register and the page answers: where does it run, could you leave, could you change calmly?

When a provider updates or retires something, the register tells you exactly what to re-check. That is what makes provider news an item on a list instead of a small crisis.

06

Leave the agent fleets for later

You will hear more and more about systems where many agents pass work between themselves. The idea is real, and most firms should not start there.

A fleet multiplies everything this playbook taught you to keep in check: what the system can reach, what it can spend, and what can go wrong while nobody watches. The honest readiness test: your individual agents run reliably, they rarely need correcting, and your list of what runs is current. Until then, several simple agents doing one job each beat one clever fleet.

Takeaway

Match each automation to the protection it actually needs, keep the firm's knowledge in files you own, and know for everything that runs where it runs, who cares for it, and how you would leave. These are decisions, not projects.

The engine is documents and routines, and documents and routines only stay alive when people look after them. The people plan is deliberately small: give every standing piece one named person, and lead adoption by example rather than by announcement. The reason for the names is simple: a responsibility held by "the team" is held by nobody.

A team does not out-adopt its leader.

Adoption is cultural before it is technical. Announced tools get quietly ignored; demonstrated tools get copied. The habits in this article are deliberately cheap, because the expensive version, a rollout program with mandatory training, mostly produces attendance.

The mechanism
Three pieces, three names, zero new positions
What ongoing attention the engine actually needs, and what holding it looks like at a twelve-person advisory firm.
The list of what runs
Its reviews actually happen
The operations lead makes sure the quarterly review happens. A few hours a quarter, inside the job she already has.
The shared collection
It gets pruned now and then
A senior consultant tidies the shelves occasionally: retires what has gone stale, re-tries the important pieces when tools change.
The questions
They have somewhere to go
Questions reach the partner who wrote the policy. Because there is a name, they get asked instead of guessed at.
Nothing changed on paper; three things stopped falling through the cracks. These are not new positions and this is not an org chart: a few hours a month folded into jobs people already have, and in a small firm one person can carry all of it.
The moves
01

Give every standing piece one named person

The list needs its reviews to happen, the shared collection needs occasional pruning, and people's questions need somewhere to go. Each gets a name, folded into an existing job.

02

Lead in the open

Adoption follows what leaders visibly do, not what they approve. A leader who uses the tools in the open gives everyone else permission by example.

Five minutes on Monday beats the training budget

A partner opens the Monday meeting with the brief she used to draft a client memo, including the two corrections she had to make before it was right. The opposite is just as visible: a leader who announces AI and never touches it teaches the firm that the tools are for other people.

03

Hold a build day

Each person automates the task they like least, self-chosen, so nobody's work is automated at them.

Whatever proves itself goes into the shared collection with the builder's name on it. The credit is not decoration; it is why the next build day has volunteers.

04

Ask it in interviews

"Show me how you work with AI" says more than any line on a resume, and it tells candidates what kind of firm they are joining.

05

Point the tools at the firm's own people work

Starting with onboarding, because it is the people process every firm has and few maintain.

The firm's onboarding manual, converted by a model into short modules with a quick check after each one and a first-day practical task marked against a checklist. A new hire's gaps show up as flagged answers to talk through together, and the manual finally gets read.

06

Check the three signs

The people side is working when three questions get answered without a pause.

Every piece has its person
Ask who looks after the list, the shared collection, and the questions, and get an answer.
The build day left something behind
Something built at it still runs, and its builder gets the credit.
The leaders' use is visible
Anyone in the firm can name something the leadership actually does with the tools.
Takeaway

A named person for every standing piece, leaders who visibly use the tools, and people who learn by building. The people side costs hours a month, and without it the documents go stale.

Earlier in the playbook you measured one workflow; the engine measures all of them. The discipline is the same one that proved your first workflow: compare what the automation delivers against what the work cost before, and decide from the number, not the mood. What changes at this stage is only the scope: everything the firm runs, together, on a schedule it cannot skip.

A firm that never retires anything is admiring its automations, not measuring them.

These numbers are what earn the engine its budget. Measured wins argue for the next build better than any enthusiasm, and the retired automations are what make the wins credible: a list where everything supposedly works convinces nobody.

What you're building
One quarter's review, on a page
Every automation on the register, three questions each, a written decision with a date. If it takes more than a morning, the measuring has grown too clever.
Sample output — yours will look like this
Keep
Proposal drafting
delivered its best quarter; the number is the argument, with evidence attached, for taking on the next workflow
Keep
Support inbox triage
two categories promoted since last review, corrections low and stable; the scorecard came with the responsible person, not reconstructed in the room
Fix
Intake summaries
the count of summaries the team actually used is down against its best stretch; the new document type it mishandles is already named in its own logs, so the fix leaves the room as a work order
Retire
Meeting-notes summaries
nobody uses the output; second review says so twice. Its brief and process notes go to the shared library before it is switched off, so the next attempt starts warm instead of from zero

Decisions dated, written into the register, retirements included. Quarterly works for most firms; look monthly at anything new or busy, and twice a year is enough for the stable and quiet.

The same three questions for every entry: what did it deliver, what did it cost, and is that worth keeping? The responsible person brings the numbers from the automation's own records; the meeting does not reconstruct them.
The moves
01

Bring the numbers, don't reconstruct them

Results and costs since last time, taken from the automation's own records. The responsible person brings the numbers; the meeting reads them.

02

Compare against the best stretch, not the start

Automations fade as inputs, tools and models shift around them. The review is where fading gets caught on a schedule instead of by accident.

Flat is a finding too

An automation that merely matches its launch quarter while volume grew has quietly gotten worse. The comparison against its best stretch is what shows it.

03

Decide, and write the decision down

Keep it, fix it, or retire it, dated, in the register, retirements included. Retiring things is normal.

04

Let each decision feed something

The review is not just a scorecard; each of its three outcomes leaves the firm with something useful.

A keep
A strong number is the argument for the next workflow, with evidence attached instead of enthusiasm.
A fix
Leaves the review as a clear work order, because the automation's own records usually say what went wrong.
A retirement
What the workflow taught goes onto the shelf before it is switched off. Even a retirement leaves the firm faster than it found it.
05

Keep the measuring simple and the schedule fixed

Three plain questions per automation: what did it deliver, what did it cost, is that worth keeping. Set the dates in advance, and let nothing skip its turn.

Sized to the automation, not to ambition

Quarterly for most. Monthly for anything new or busy. Twice a year for the stable and quiet. If the review takes more than a morning, the measuring has grown too clever.

Three signs the review is honest: nothing gets skipped, fading gets noticed, and the decisions are written down, dated, retirements included.

Takeaway

On a fixed schedule, for every automation: what did it deliver, what did it cost, keep it or not. A firm that never retires anything is admiring its automations, not measuring them.

Sooner or later someone will ask what your firm does with AI: a client before they sign, an insurer at renewal, eventually a rule. A clear, unhesitating answer signals exactly what the person asking wants to know: that someone here is in charge of this. And because most firms cannot yet answer the question well, answering it plainly sets you apart more than the tools themselves do. The rule: whoever asks, the honest answer already exists in writing.

Firms rarely lose work for using AI; they lose it for being vague about it.

If your people picked up the habit early of noting what a model touched in each piece of work and who checked it, none of this is new effort. The engagement letter simply says at firm level what the work already says internally, and the honest sentence writes itself from practice.

What you're building
The questionnaire, answered from the shelf
A client's procurement team sends its AI question list. Every answer comes from a document that already exists, and the reply goes out the same week.
Sample output — yours will look like this
Reply to the procurement questionnaire — AI use
Answered from standing documents · sent the same week · no meeting required
What AI tools does your firm use, and for what? Named in our policy's approved-tools section, and named again in each engagement letter with what they are used for. From: the one-page policy · the engagement letter
What of our data do these tools touch? Listed per automation: what it does, what data it touches, where it runs. The summary attached is current as of this month. From: the register
Who checks the output? Work that reaches you is signed off by a person; where that line sits is written in the policy, and each automation's entry names its responsible person. From: the policy · the register
Is our data used to train these tools? No. The engagement letter states it: drafting and document review are assisted by the named tool under our review; your data is not used to train it. From: the engagement letter
What happens when a tool fails? Each automation carries one page of instructions including how the work runs by hand, and a named person responsible for it. From: the register · the placement page

The questions are typical of the lists clients and insurers now send; the answers are one firm's. What transfers: every answer names the standing document it came from, and none required a meeting.

This is what the whole stage buys: a week of scrambling becomes a same-day reply, assembled from the policy, the register and the letter that already exist.
The moves
01

Put the line in the engagement letter

Clients are told, in the letter, which tools touch their work and what they are used for. A vague "we may use AI" line protects nobody and reads like it is hiding something.

Name the tool, name the use, state the training answer

"Drafting and document review are assisted by [named tool] under our review; your data is not used to train it." One sentence, and the most common question is answered before it is asked.

02

Keep the answers ready in advance

Clients and insurers increasingly send question lists: what do you use, what does it touch, who checks the output. The register already holds those answers.

A short summary, kept current

One page distilled from the register, refreshed when the register changes. It turns a week of scrambling into a same-day reply, as in the figure.

03

Know what the law actually asks of a firm like yours

Less than the headlines suggest. For a typical services firm using general-purpose tools, the current rules come down to two things: your people know how to use the tools responsibly, and nobody is misled about dealing with AI.

Chatbots say they are chatbots, and AI-made content is labeled as such. The heavy obligations target uses like automated hiring decisions, and they sit mostly with the companies that build the models, not with firms that use them. A short internal training and the one-page policy from the start of this stage cover most of what is asked.

This is the one part of the playbook that touches law, and law moves. Treat this as orientation as of mid-2026, not as advice; your counsel has the final word for your firm.

04

Check the three signs you are ready

Readiness is checkable, and all three checks read from documents that already exist.

The letter says it
A client can learn from their engagement letter which tools touch their work, without having to ask.
A question list holds no fear
An AI questionnaire arriving today would be answered this week, from documents that already exist.
The answers match
Ask anyone at the firm what it does with AI, and the answer agrees with the policy.
Takeaway

An honest letter, ready answers, and people who can say what they use. Firms rarely lose work for using AI; they lose it for being vague about it.

The first article of this stage built the list of what runs; this one builds its companion: the collection of what works. They are different things. The register exists so you stay in control; the library exists so the firm gets faster. A personal collection makes one person faster: it lives in someone's account, improves when they remember, and leaves when they do. The shared library makes the firm faster, permanently. Treat proven prompts, templates and instructions like any other thing of value the firm owns.

Documented work compounds; undocumented work walks out the door with its owner.

The library is why the fifth workflow costs a fraction of the first. Every documented process, proven brief and set of instructions is a head start on the next build. Firms that keep them compound; firms that keep starting over do not.

What you're building
The shelves, sorted by the work
People reach for the library with a task in front of them, so the shelves are named after tasks, never after tools.
Sample output — yours will look like this
The proposal shelf
Brief
Proposal first draft
promoted from a senior's personal collection in Stage 1, with credit; improved in place ever since
Guide
The tone guide and the checklist
what the firm's best proposals already do, written down; a new consultant produces a house-standard draft in week one
The intake shelf
Brief
Enquiry to intake entry
a messy enquiry becomes a clean, structured entry; the same instruction the intake automation runs on
Notes
Meeting-notes brief and process notes
kept from the retired automation at its last review; the next attempt at the problem starts warm instead of from zero
The reporting shelf
Skill
How we write summaries
the firm's structure and tone as a multi-step instruction; one good copy, a one-line note whenever it changes

The shelves are one firm's; the sorting rule transfers. A library sorted by tool names goes stale with the first change of provider; a library sorted by the work survives every one.

The new consultant with a proposal due opens the proposal shelf and finds the brief, the tone guide and the checklist. She never has to ask which tool the firm prefers; the shelf answers it in passing.
The moves
01

Promote what proves itself

When something personal proves itself, it moves to the shared shelf, named and credited.

Credit is the engine of the habit

The builder's name stays on the piece. People promote their best work when promotion is visible; they hoard it when it disappears into an anonymous folder.

02

Shelve by the work, not the tool

A proposal shelf, an intake shelf, a reporting shelf. People reach for the library with a task in front of them.

03

Keep one good copy

Improved in place rather than copied into private variants, because variants are how quality quietly forks.

A one-line note when something changes

So a bad change can be undone, and so the next person knows why the brief reads the way it does.

04

Tidy occasionally

Whoever looks after the collection prunes what has gone stale and re-tries the important pieces when the tools underneath them change. A small standing habit, not a committee.

05

Close the loop

Once something works, capture it so someone else can run it without its inventor. That principle built your personal collection back in Stage 1; here it becomes how the whole firm keeps what it learns.

That closes the engine: rules, a clear view of what runs and where, people who keep it alive, honest numbers, ready answers, and a library that compounds.

Takeaway

Promote what works from personal to shared, shelve it by the work it does, and keep it tidy. Documented work compounds; undocumented work walks out the door with its owner.

Closing this stage
Operating well at this altitude

The engine is a steady state, and it is deliberately light. Operating well here looks like: the list of what runs is current, the numbers are honest, the answers are ready, the library keeps growing, and none of it needs a committee. This is the destination for firms with enough running; there is nothing above it to climb to.

How it works

One method, run at rising scale

The first three stages run the same six-move loop at rising scale: on your own recurring work, then on one shared workflow, then across many workflows with automation and agents. The fourth stage is different in kind: the operating engine, the light set of standing rules and shared documents that keeps everything you've built running and compounding. Six principles sit underneath all of it.

The loop · six movesRun in order, one workflow at a time
1

Find the constraint

2

Map the work

3

Choose the route

4

Run the pilot

5

Harden and own

6

Measure the return

What actually breaks first, or drains the most hours. One bounded task, not a whole function.

Watch the process as it really happens, capture a baseline, and name one owner.

Pick the simplest option that clears the bar; move up a rung only when the one below runs out.

Run it beside the current way, with the go/no-go rule decided in advance and a time limit.

Add the guardrails, name the owner, and make it the default way the work gets done.

Track verified outcomes, not usage. Retire whatever stops paying its way.

Move 6 hands back to move 1, pointed at the next constraint
The six principlesUnderneath every stage
01

Problem before tool

Start with the bottleneck you actually want to clear, and let that choose the tool, not the other way around.

02

Rules before agents

An AI that runs on its own is only as reliable as the process behind it, so define the workflow as clear steps and rules first, then let it take over.

03

AI executes, humans certify

Let the system do the work, but a person reviews and signs off on anything a client sees or that would be costly to get wrong.

04

The harness is the product

What makes an automation safe to trust is the structure around it: spending limits, clear boundaries, an off-switch, and a record of what it did.

05

Ground every answer

When a model gives you an answer, make it cite and check its sources, because you stay accountable for the result.

06

Codify and delegate

Once something works, capture it as a repeatable process, so the next time is faster and someone else can run it without you.

Underneath every stage
Book a working session

Forty-five minutes on one of your workflows

Bring the workflow that bothers you most. We map where the work leaks, tell you what we would run the loop on first, and you keep the notes either way. No deck, no obligation.

The AI Operating Playbook, by Sprint Assembly
Link copied