Open any programme's risk register and you will find it populated with respectable, tractable risks: resource availability, data quality, integration complexity, testing capacity. All real, all worth tracking. What you will not find is the risk that actually threatens the programme — because the risks that sink transformations are almost never the ones it feels safe to write down. They live in a second, unwritten register, carried in the heads of people who all know the same uncomfortable things and each assume it isn't their place to say them.
This gap between the register on the wall and the register in the room is not a governance oversight. It is a predictable product of incentives. The formal register captures the risks that can be raised without cost. The dangerous ones carry a social price to name, and so they don't get named — until they arrive as issues, when it is too late to have called them risks.
The risks that don't get written down
They share a signature: everyone senses them, no one benefits from raising them.
The register no one maintains
- The disengaged sponsor — the programme has no real air cover, and everyone can feel it
- The workstream that's quietly behind — visible to peers, invisible to the report
- The decision being avoided — deferred so long that deferral has become the decision
- The estimate no one believed — agreed to keep the plan intact, not because it was real
- The dependency on one person — the individual whose departure would stop everything
Why silence is rational
It is tempting to blame culture, but the silence is usually a rational response to how the programme is run. Raising the sponsor's disengagement means criticising the most powerful person in the structure. Flagging another workstream's slippage means breaking ranks with a peer who will be in the next meeting. Naming the avoided decision means owning the discomfort of forcing it. In each case the person who speaks bears the cost and the programme gets the benefit — a textbook reason for a rational person to stay quiet and hope someone else goes first.
The dangerous risks aren't hidden. They're known by everyone and written by no one — because naming them costs the namer and helps only the programme.
The status report that reports comfort
The unwritten register has a close relative: the status report that stays green not because the programme is healthy but because green is the colour that avoids a difficult conversation. When "amber" invites scrutiny the team would rather not attract, the rational move is to hold at green and privately hope to recover before anyone notices. The report stops describing reality and starts managing perception — and the board, reading comfort, sees no reason to intervene until the green finally, suddenly, goes red.
Making it safe — or making it independent
There are only two real remedies. The first is cultural: a programme where raising a hard risk is rewarded rather than punished, where sponsors invite the bad news that protects them. That is worth building and genuinely rare. The second is structural: someone whose job is to surface the unwritten register precisely because they carry none of the social cost of naming it — no peer to break ranks with, no sponsor to offend, no green to protect. When candour is expensive for insiders, independence is what makes it affordable.
Where we sit. Roltrader's delivery and assurance layers read the signals the written register omits — the decision that keeps slipping, the reconciliation that keeps not tying out — and report them without a stake in the answer. But the first move is free: ask, in every steering meeting, what isn't on the register that everyone in the room already knows. The most important risks are usually the ones no one has been rewarded for writing down.