Much about IT projects in public administration is decided before the start.
The technology is rarely the hard part. Requirements, responsibilities and procurement are fixed earlier in a public procedure than anywhere else — and that is exactly why they are where the most can be steered.
Key points
Clear responsibilities carry a project further than any software.
A requirement with a named owner is a decision. Without one it stays a wish.
Procurement sets the points before delivery begins.
Leadership needs a role of its own — not another line in somebody's workload.
Requirements need an owner, not a collection.
Requirements in public administration usually start out broad and thorough: every department reports what it needs. The step that decides things comes afterwards — somebody sets down what the system has to be able to do, and what it expressly does not.
In a public procedure that step weighs more than in the private sector, because requirements are part of the procurement: what stands in the specification has been negotiated. What is missing becomes a change request after the award. A structured way of gathering requirements — following CPRE (IREB®), for instance — is therefore not a formality. It is the cheapest protection a project can have, and it works exactly once: beforehand.
Governance shows in decisions, not in meetings.
HERMES 5 sets out roles, phases and decision points. The model is good and it holds — it does its work where it is read as an instrument of leadership rather than a documentation duty.
There is one question that tells you which it is: does everyone involved know which decision they may take themselves — and which they may not? Where that answer comes without hesitation, critical decisions run at the normal pace. Where it hesitates, they pile up until they have to be escalated. The question can be settled in a single meeting. Settling it is the most effective thing anyone can do at this stage.
Procurement is the moment with the greatest leverage.
Public projects are shaped in the award procedure. Deadlines, award criteria and reference checks reach all the way into operation — and after the award they can only be moved so far.
That is precisely the opportunity. Take the groundwork seriously — feasibility, the range of effort, the state of the market, how many suppliers are realistically available — and you write a more precise tender, receive offers that compare better, and negotiate from a stronger position. The work falls due before the costs do. That is the cheapest moment a project will ever have.
Project management is a role, not an extra duty.
In many administrations an experienced specialist runs the project alongside their day job. That often works remarkably well — and reaches its limit where the day job takes no account of it: attention runs short exactly when the project needs it most.
Bringing in external project management is therefore not a verdict on your own organisation. It is an answer to a shortage of capacity — time-limited, with a clear brief, a clear end, and the express task of leaving the knowledge in the house. The subject expertise stays where it belongs. What is added is the time to see it through.
What is settled before the award does not have to be negotiated after it.
Recommendations for practice
Settle before the start who takes which decision — in writing, with names.
Have one responsible person consolidate the requirements — with a named owner.
Budget the groundwork as a stage of its own — it is cheaper than any correction later.
Fill project management as a role of its own, time-limited and handed back to the line.
Dates before scope: move functions, never the financial year-end.
A project in preparation?
The groundwork is the point where the most can be moved and the least is spent. If that is where you are, let us talk about it.
Discuss the groundwork→