In development

PMO Autopilot

A multi-agent PMO that runs your program controls across every project, in whatever method each one uses.

Not available to buy yet, and not yet priced. Register and we will tell you when it is ready.

12 agents running in parallel
85 projects managed at once
1 human program manager accountable

Why we are building it

The problem this closes

PMO cost scales with portfolio size, and most of that cost is coordination rather than judgement: chasing status, reconciling a schedule against what teams actually did, spotting the dependency nobody logged, and rebuilding the same board pack every month. That work is structured, repetitive and evidence-based — which makes it a good fit for agents and a poor use of a senior program manager. The judgement stays human. The assembly does not have to.

Contents

What it does

  • A team of AI agents, each delivering a recognised PMO persona — reporting analyst, risk and issue coordinator, dependency manager, schedule analyst and benefits tracker — rather than one general assistant.
  • Runs across a whole portfolio at once, and manages each initiative in the method it actually uses: PMP, Agile, Scrum or SAFe.
  • Connects to Atlassian, Monday.com and Microsoft (Azure) Planner, and reads the delivery record as it actually is, instead of as it was last summarised into a status report.
  • Drafts the status pack, the risk register update and the steering-committee papers from live data, with every figure traceable to the item it came from.
  • Flags the things a PMO is meant to catch: schedule slippage against baseline, dependencies with no owner, risks past their review date, and status that is greener than the underlying evidence supports.
  • Keeps a decision and change log, so what was reported and who approved it remains reconstructable months later.
  • Every output carries an explicit human approval step. The agents assemble and flag; a named person decides and signs.

What it looks like at scale

In real use, up to twelve agents run in parallel across as many as eighty-five projects, coordinated by a single human program manager. The agents carry the assembly and the monitoring across every project at once; the program manager carries the judgement and the accountability. The point is not to remove the program manager — it is to let one person hold a portfolio that would otherwise need a whole PMO to service.

What it deliberately does not do

It does not make decisions, approve changes or set status on its own authority. It does not replace a program manager — it removes the assembly work that stops one from doing the job. It will not report a position the underlying delivery record does not support, and where the data is missing it says so rather than estimating. Governance is being built against ISO 42001 and the NIST AI Risk Management Framework, with the same human-accountability model we apply to client AI work.

Who it is for

Program managers, PMO leads and CIOs carrying a portfolio large enough that coordination has become its own cost centre, and who need reporting that holds up in front of a board or an auditor.

What it saves

The proportion of PMO effort spent assembling rather than deciding — the status cycle, the reconciliation and the pack build, which recur every month regardless of what the program is doing.

Format

Web application with connectors to Atlassian, Monday.com and Microsoft Planner. Subscription.

Status

In development. Commercial terms will be published with the first release.

Next step

Tell us you want this one

Registering your interest puts you at the front of the release and shapes what we finish first. If enough people ask for a different product before this one, we change the order.

A short first conversation, then a clear recommendation.

Tell us what the program is, where it is, and what has to hold up. If we are not the right firm, we will say so and point you to who is.