We were asked to review a programme that had been reported green for eleven consecutive months and then, in month twelve, was declared eight months late. Nobody had lied. The reporting format simply had no way to say that something was going wrong.
Transformation governance is not about control. It is about making bad news travel faster than it wants to.
One owner, named, with authority
Programmes with a committee as owner have no owner. Committees can approve and can review; they cannot decide quickly, and transformation is a stream of small decisions that must not queue.
Name one person. Give them authority to spend within a threshold and to change scope without reconvening anyone. If nobody can be given that authority, the organisation is not ready to start.
In many Jordanian companies this owner is best placed in operations rather than IT. The change is to how the business works; IT is how it is delivered.
The decision log
The single most useful artefact, and almost nobody keeps one. A simple table:
- What was decided.
- When.
- Who decided it.
- What alternatives were considered.
- What we assumed to be true.
Six months later, when someone asks why the system works a particular way, the answer exists. Without it every decision is relitigated by whoever arrives next, and programmes lose months to arguments that were already settled.
It also protects the people who decided. A decision that turned out badly but was reasonable with the information available is defensible — if the information available was recorded.
Status reporting that can say "not fine"
Traffic-light reporting fails because amber is socially expensive. Reporting amber invites scrutiny, so people report green until they cannot.
Replace the colour with three questions, answered in writing:
- What did we complete since the last report? Completed means accepted, not "in testing".
- What is now different from what we expected? Both good and bad.
- What decision do we need from this group, by when?
The third question is what makes the meeting worth holding. A status meeting with no decisions to make is a presentation, and presentations do not surface problems.
Make escalation safe
Whether problems arrive early depends on what happened to the last person who raised one.
If raising a risk produced questioning of competence, the next risk arrives late. If it produced help, the next one arrives early. This is set by the sponsor's behaviour in the first two months and is very difficult to change afterwards.
A practical measure: the first time someone escalates something genuinely difficult, thank them explicitly in front of the group. People will notice.
Review the benefits, not just the delivery
Most governance tracks whether the project is on time. Very little tracks whether it still makes sense.
Twice a year, ask a different question: is the reason we started this still true? Markets shift, regulations change, the process being automated may have been reorganised. A programme delivering on schedule against an obsolete rationale is worse than a late one, because nobody is looking for the problem.
Be willing to stop. Stopping a programme that no longer makes sense is a governance success, though it rarely feels like one.
The standing agenda
Thirty minutes, monthly, in this order:
- Decisions needed today. First, because energy is highest at the start.
- What changed since last time.
- Risks that became more likely, not the full register.
- Benefits check, quarterly only.
Deliberately absent: a demonstration. Demos belong in a separate session with the people who will use the system, not in the meeting where decisions are made.
How to tell whether yours is working
Ask two questions of your last three status reports.
Did any of them contain something the sponsor did not want to hear? If not, your governance is producing reassurance rather than information.
Did any of them result in a decision that changed the plan? If not, the meeting is a ritual and everyone attending knows it.
Good transformation governance is quiet, slightly uncomfortable, and short. Programmes with elaborate governance are usually compensating for the absence of a single accountable owner.
