Accountability maps look clean on paper. They fail when someone tests them. Five questions to pressure-test yours.
Thomas Collins, GAICD.
The Financial Accountability Regime now applies across banking, insurance and superannuation. It does something deceptively simple. It requires that for every significant responsibility in the organisation, one accountable person can be named, and that person’s obligations are written down and enforceable.
On paper, most boards are ready. The accountability statements are drafted, the accountability map is built, the roles are allocated. The trouble is that the regime is not tested on paper. It is tested when something goes wrong and a regulator asks who was accountable, and whether the map matched reality.
The map and the territory
An accountability map is a model of how the organisation is meant to work. Reality is how it actually works. The two drift. A responsibility moves in practice but not on the map. An accountable person leaves and the obligation sits unassigned for a quarter. Two executives each believe the other owns a control. None of this shows up until it is examined, and by then the gap is the story.
The board’s exposure is that it attested to a structure it never pressure-tested. “We had a map” is not the same as “the map was true.”
Five questions to pressure-test yours
A board does not need to re-run the legal mapping exercise to govern this well. It needs to test whether the accountability the document describes actually exists.
Pick any significant responsibility at random. Can we name the accountable person in under ten seconds, and would they agree it is theirs?
Where two functions meet, handovers, shared controls, outsourced services, is accountability genuinely allocated, or does it fall into the gap between two statements?
When an accountable person last changed roles or left, did the map update at the same time, or weeks later?
Have we tested a real scenario against the map, or only confirmed the document is complete?
If the regulator interviewed our accountable persons separately, would their accounts of who owns what actually agree?
Where boards get caught
The common failure is treating accountability as a one-time drafting project that finished at go-live. The regime is continuous. Every reorganisation, every new product, every outsourced arrangement changes the territory and should change the map. Boards that built a clean structure in year one and never revisited it are the ones most exposed in year three, because the document has quietly stopped describing the organisation.
The other trap is confusing volume with clarity. A thicker accountability pack is not a safer one. Regulators are not impressed by length. They are interested in whether, when it mattered, one person was clearly accountable and knew it.
What good governance of accountability looks like
Good boards treat the accountability map as a living control. They review it on a cycle, not just after an incident. They ask management to walk a real scenario through it at least once a year. And they keep the language plain, because an accountability framework that only a lawyer can read is one no executive will use under pressure. The point is not a tidy document. It is that when the hard moment comes, no one in the room has to ask who was responsible.
