Project charter

Project. Cross border settlement rebuild

Sponsor. Name, role, the person accountable for the benefit

Project manager. Name, role

Date. Version and date

---

Why this project exists

Two or three sentences on the situation that makes this worth funding. What is happening now, what it costs, and what changes if nothing is done. Written so somebody outside the department understands it.

Settlement instructions are reconciled by hand across three regions, which takes eleven staff days a week and produces an average of four breaks a month. Each break takes two days to resolve and two of the last twelve were reported to the regulator. Volume is forecast to rise by a third next year on the same headcount.

What it will produce

The deliverables, at the level a sponsor cares about rather than the level a plan does. Three to six lines.

• A single reconciliation engine covering all three regions
• Automated break detection with an exception queue
• Migration of twenty four months of historical instructions
• Decommissioning of the two regional tools it replaces

Objectives and how success is measured

Each objective needs a measure, a baseline and a target. An objective without a baseline cannot be shown to have been met.

What is not in scope

The three or four things a reader would otherwise assume were included. This section prevents more argument than any other.

• Client facing reporting, which stays on the existing platform
• The Asia Pacific region, which is on a separate regulatory timetable
• Any change to the underlying ledger

Milestones

Summary level only. A charter is not a plan and a charter with thirty milestones is a plan wearing the wrong title.

Budget and funding

The authorised amount, what it covers, and where the money comes from. State the contingency separately and say who releases it, because a budget with contingency buried inside it will be spent as though it were scope.

Authorised 1.4 million, funded from the operations change portfolio. A management reserve of 140000 is held by the sponsor and released against a written request.

Known risks

The three or four that would change the decision to proceed. Detail belongs in the RAID log.

• The acquirer's certification lead time is outside our control and sits on the critical path
• One controls engineer holds the legacy rig knowledge and no handover is scheduled
• The volume forecast is the finance planning figure and has not been validated against the settlement data

Assumptions and constraints

What is being taken as true, and what cannot move.

• The regulatory reporting deadline in November is fixed and cannot be renegotiated
• The finance system will accept new cost codes without a change request, which is being tested
• No additional headcount is available before the next financial year

The project manager's authority

What the project manager may decide without asking, and what has to be escalated. A charter that appoints somebody without saying this leaves them escalating everything, which is the only safe behaviour when authority is undefined.

The project manager may commit spend to 25000 per decision within the authorised budget, agree changes that do not affect the milestone dates above, and select suppliers from the approved framework. Anything affecting a milestone, the budget or the regulatory position goes to the sponsor.

Authorisation

The signature line is the point of the document. A charter nobody signed authorises nothing, and the project it describes has no more standing than an email.

Sponsor. Signature, name, date

Project manager. Signature, name, date

---

How to use this

Keep it to a page or two. The charter exists so that in month seven, when somebody asks why the project is doing something, there is a document that answers rather than a recollection.

Write it with the sponsor rather than for them. A charter drafted alone and sent for signature gets signed without being read, and the value was in the conversation that settled the objectives and the authority limits.

Revisit it when the business case changes. A charter is not a plan, so it should not change when the plan does, and it should change when the reason for the project does.