What it is for
Any IT or technology program. A three month upgrade and a three year transformation differ in how many rows there are, not in what a row is.
Six delivery profiles sit underneath. One is always on, the rest switch on for what your program is, and most programs run several at once.
| Profile | What it covers |
|---|---|
IT delivery baseline | Always on. The stages, gates and evidence every program owes regardless of type. |
Package, COTS or SaaS | Anything bought rather than built: ERP, CRM, HCM and payroll, EHR, core banking, MES, ITSM. |
Custom software | Something you are building: a digital product, a platform, an internal system with no vendor model to start from. |
Migration | Moving what already exists: data centre exit and cloud, mainframe, endpoint and OS, email and workplace, identity. |
Data and analytics platform | Warehouse and lakehouse, pipelines, the report estate, and the governance that decides whether any of it is trusted. |
Transformation overlay | Sits on top of the others when the program changes how the business works, not just what it runs on. |
Carve-outs, M&A and regulatory programs are combinations of these, plus a date somebody outside your company set. That date is what changes the plan.
What you need
A Google account, a program, and about thirty minutes for the first conversation. That is the whole list. There is nothing to install, no seats to buy, and nobody on your team creates an account at any point.
The workbook is created in a HeyPM folder in Google Drive and shared with you, with full edit access, from the first minute. When the engagement ends, ownership transfers to your Google account.
What it asks of you
The agent does the work. These are the seven moments it hands something back, worth knowing before you start so that nothing about it is a surprise. Everything not on this list runs whether you open it or not.
| It asks you to | What that means | How often |
|---|---|---|
| Describe it once | The way you would brief a new joiner. What is changing, who runs what, the date you are working to. | 30 minutes, once |
| Name your people | Name, work email, what they own. Nobody creates an account, installs anything, or needs a licence. | 5 minutes, once |
| Answer the brief | Each person gets only their own items at eight. Three buttons: all good, running late, stuck. | One tap, daily |
| Approve, or not | It never moves your dates by itself. When they need to move it shows the change and the reasoning, and waits. | When it asks |
| Decide the gaps | In scope, out of scope, or deferred. A gap nobody has decided keeps blocking whatever waits on it. | Weekly |
| Evidence what is done | A link, a document, a ticket reference. Without one it stays open, on purpose, which is the point of a gate. | At each gate |
| Overrule it | It is a Google Sheet. Type over any cell and the next morning takes your value as the truth. You never have to undo anything it did. | Any time |
Do nothing at all and the checks still run, the briefs still go out, and the escalation clock keeps counting. That is the part that stops things going quiet.
Your first conversation
You describe the program the way you would describe it to a new joiner. What is being replaced, which parts of the business it touches, who runs each one, and the date you are working to. The agent asks about the gaps as it goes rather than at the end.
Where the program is a known platform, it starts from the model for that platform and asks you to correct the part that is genuinely yours. Where it is something you built, it starts from a baseline instead: who, what, when, and the artifact each stage owes.
You do not need a charter or a plan. If you have one, send it and it gets read first, but the conversation is enough on its own.
What arrives
One Google Sheet, populated in a single pass. Nothing is written until the whole plan validates, so you never get half a program with the other half missing.
The tabs, and what is in them
There are more tabs than you will ever open, which is deliberate: a program in discovery and a program in hypercare need different ones, and neither should have to build the missing half itself. So the tabs are coloured into three groups, and the colour on the tab is the instruction.
Green — yours
The ones you fill in and the ones you read. If you only ever touch these, the agent has everything it needs.
| Tab | What it holds |
|---|---|
Start here | A live checklist of what is filled in and what is not, with a link into each tab. It ticks itself as you go. |
Program Profile | Core facts about the program, so nothing has to be re-explained in every conversation. |
Milestones | The key dates. One of them is ticked as your real go live, and everything is measured against it. |
Departments | The major areas of work. Everything else hangs off these, so they get named first. |
People | Everyone involved, and how the agent reaches them. |
Roles | Who holds what, and which department they hold it in. |
Process Areas | The specific things being tracked inside each department. This is the level fit gaps hang off. |
Outcomes | What success looks like, with an owner and a date. |
Capacity | Forecast effort by month and by team. Rebuilt every morning, so there is nothing to fill in. |
RAID | Risks, actions, issues and decisions, with the decision ones dated and owned. |
My Actions | What the agent has asked you, and what you answered. |
Agent Settings | How the agent behaves: who it writes to, when, and in whose voice. |
Amber — when you need it
Empty and out of the way until the program reaches them. A tab appears when its moment arrives rather than sitting there looking unfinished for four months.
| Tab | What it holds |
|---|---|
Current State | How each process works today. |
Future State | How it needs to work instead. |
Process Flow | The two of those, drawn. Rebuilt every morning. |
Fit Gap | The differences between today and tomorrow, and what you are doing about each. |
Sequence | The order the steps run in, where that matters. |
Dependencies | What waits on what. A real graph, not a column of notes. |
Cutover | The go live weekend, and everything that has to happen for it to work. |
Training Plan, Sessions, Attendance | Who needs training on what, when each session runs, and who actually came. |
SOPs | The documented, correct way to do something. |
Test Scripts, Sessions, Results | What gets tested and how, each round of it, and what passed, failed or is blocking something. |
Hypercare Log | Support after go live. |
Lessons Learned | What you would do differently. Worth writing down while it still stings. |
Grey — the agent’s own
Hidden by default. They are the agent’s working memory, not yours, and the next section says what is worth knowing about them.
Outcomes, owners and dates
An outcome is something that has to be true before go live. Every one carries an owner, a date, a stage and a status. Those four are what almost every finding is about: an outcome with no owner, no date, a date that contradicts another one, or a status nothing evidences.
Status is set in the workbook using its own words: On track,
Slipping, Blocked. The people answering by email see plainer ones.
Both refer to the same thing.
| In the workbook | What the person reads |
|---|---|
On track | All good |
Slipping | Running late |
Blocked | Stuck |
Dependencies and fit gaps
Dependencies are held as a graph rather than a list, so a change to one date travels to everything downstream of it. A fit gap with no date of its own still becomes an edge in that graph, which is why a gap nobody has dispositioned can block an outcome that looks fine.
This is the one that catches people out. A gap has no date, so it never appears late, and three steps later it is the go live date.
People, cover and escalation
Each person has what they own, who covers them, and who it escalates to when they do not answer. Load is read off the plan rather than from a timesheet, so the agent can tell you who is carrying too much and which items would stop if one person were away for a week.
An item with nobody to escalate to is itself a finding. So is a department with no owner.
The agent’s own tabs
These are hidden, and they are hidden because they are working memory rather than content. You can unhide any of them; two are worth knowing exist.
Agent Log. What it did this morning and in what order, with counts. This is the first thing to look at if a morning seems to have been skipped.Agent State. What it is still carrying. Things raised weeks ago that nobody has answered live here, which is why nothing drops off because a week went by.
The rest — the dated plan it is working from, the capability list, the escalation history — exist so that the agent can answer a question the same way twice. Nothing in them needs maintaining, and editing them by hand is the one kind of edit the next run will overwrite.
What happens before eight
Every working morning, in this order:
- Reads every row in the workbook.
- Runs fifty-two finding rules across all of it.
- Works out which findings need a person, and which person.
- Sends one brief to each of those people, and nothing to anyone else.
- Advances the escalation clock on anything still unanswered.
- Writes what it did to
Agent Log.
It runs whether anyone opens the workbook or not, and nothing in this list is you opening something. There is a page you can read when you want to — see below — but reading it is never how the work gets done.
The brief
Each person gets only their own items, never the whole program. Every item says why it matters, usually who is waiting on it. If somebody has nothing that needs an answer, they get no email at all, which is the main reason the ones they do get are read.
Answering
Three buttons in the email: All good, Running late, Stuck. One tap, from the phone, with nothing to log into. The link in the email is the credential.
The same link opens that person’s own live list, if they want to see everything they own rather than just today’s question.
Escalation
Anything unanswered keeps its clock running and gets asked again the next morning. Only a real answer resets it, and only All good resets it fully. Running late and Stuck both keep it visible, because both mean somebody still has to do something.
Past a threshold it escalates to whoever that person escalates to. If the escalation path is empty, that becomes the finding.
Your agent page
The workbook is where the program lives and the brief is how most people touch it. Between the two there is a page at agent.heypm.ai, which is the same agent reading the same workbook, laid out for a person rather than for a spreadsheet. You sign in with a six digit code to your own inbox. There is no account and no password, because there is nothing to have an account on.
Every page on it is generated from your workbook that morning. A page with nothing behind it is not in the menu at all, so what you see is a description of how far your program has got.
| Page | What it answers |
|---|---|
Start here | What this is and what it did overnight. It leaves the menu once you have read it. |
For you | What is waiting on you specifically, with a count. This is the one to open if you only open one. |
Report | Where the program stands: the window, the next gate, what moved, and the quality of the plan underneath it. |
SteerCo pack | Friday’s meeting, written from the same rows as everything else. |
Program map | The program drawn: departments, process areas and what waits on what. |
Your workbook | What is still missing from the workbook, as steps rather than as a score. |
People | Who is carrying what, who has no cover, and who has nothing to escalate to. |
Settings | The program’s own facts, and how the agent behaves: the hour it runs, who it writes to, in whose voice. |
Each page shows one thing and puts the working behind a fold. The numbers are on the page; the rows they came from are one click away, under whichever line raised the question.
There is a chat on every page of it, and it is the same conversation that built your program in the first place. Ask it the question rather than hunting for the row: what changed since I last looked, what is blocked on me, what would you do first. It answers from your workbook, and when it wants to change something it proposes and waits.
Nobody has to open this page for the program to run. It is there for the morning you want the whole picture rather than your own three items, and for the fifteen minutes before a steering meeting.
Editing by hand
It is a Google Sheet. Type over any cell, any time, and the next run takes your value as the truth. You do not have to tell the agent, and you do not have to undo anything it did.
This is the intended way to work, not a fallback. The agent drafts; you have the last word on every row.
What it proposes, what it applies
Scheduling is proposed, never applied. When dates need to move, you get the proposal and the reasoning, and nothing changes until you accept it.
Downstream recalculation is different: when you move a date yourself, the things that depend on it are recalculated automatically, because leaving them stale would be worse than moving them. Anything that needs a judgement call comes to you instead.
People joining and leaving
Add a row, and they are in from the next morning. Remove a row, and everything they owned becomes unowned, which is deliberate, because unowned work is a finding and should not quietly vanish with the person.
Six things you will want to do
- Add a row to
Peoplewith their name and email. - Put their name in the owner column of whatever they own.
- They get their first brief the next morning. Nothing else is needed. No invitation, no account.
- Type the new date straight into the
Outcomesrow. - The next run recalculates everything that depends on it.
- Anything that now conflicts is raised as a finding, rather than quietly overwritten.
- Set the status on the outcome.
- Put the evidence in the evidence column: a link, a document, a ticket reference.
- Without evidence the agent keeps treating it as open, on purpose.
- Open
Fit Gapand set the disposition: in scope, out of scope, or deferred. - A gap with no disposition keeps blocking whatever waits on it.
- Deferred gaps come back when their date arrives.
- Tell the agent in chat, or set the flag on the program row.
- Nothing is sent while it is paused, and nothing is lost.
- The escalation clock keeps counting, so you find out what built up.
- Tell me who it should belong to and I transfer ownership of the file to them.
- Tell me so the agent writes to the right place.
- Nothing has to be exported, and nothing lives anywhere else.
Who can see what
The workbook is yours to use from the first day and yours to own when we stop, and you share it with whoever you like. People answering by email see only their own items through their own signed link. They never see the workbook, and they never see anybody else’s items.
The agent reads and writes your workbook and sends your email. It touches nothing else.
How answer links work
Each link is signed and specific to one person and one item. There is no login because the link itself is the credential, which is what makes a one-tap answer possible from a phone in a warehouse.
Because the link is the credential, a forwarded brief lets somebody else answer as that person. Treat a brief like any other work email.
If something looks wrong
Two things worth checking before anything else. Agent Log tells you whether
this morning’s run happened and what it found. Agent State tells you what
it thinks is still outstanding, which is usually where a surprise comes from.
If it still looks wrong, email hello@heypm.ai with the program name and roughly when you noticed. I read those myself.