The right question is never "how big is this?" It is "can I walk it back, and what does walking it back cost?" Most decisions treated as one-way doors are revolving.
Most organisations sort decisions by size. Big ones go up. Small ones stay down.
It is the wrong axis, and it is expensive in both directions.
What size actually predicts
Size predicts attention. It does not predict risk.
A large, easily reversed decision gets six weeks of scrutiny because the number is large. A small, permanent one gets waved through because the number is small. The organisation spends its caution where caution is cheapest to waste.
Ask a different question. Can we walk this back, and what does walking it back cost?
The test, in one line
Write down the cost of undoing it.
Not whether you could undo it in principle. The actual price: the rework, the notice period, the credibility, the second conversation with the same supplier.
If that number is small, you are not making a big decision. You are making a slow one.
If that number is large, you have found one of the genuinely irreversible calls, and it deserves every hour of the scrutiny that the other ninety percent have been borrowing.
Regulation does not change the test
This is the objection I hear most, usually from people in health, utilities or financial operations, and it deserves a straight answer.
Regulation tells you which doors are one-way. It does not tell you to treat every door that way.
A hospital cannot improvise its way through a clinical governance requirement. It also cannot afford to run every rostering change, supplier substitution and process adjustment through the same gate, and the ones that do get run through it are not safer for it. They are just later.
The reversibility test exists to protect the scrutiny, not to bypass it. Spend it where undoing is expensive.
Two lists, one meeting
Try this with a leadership team.
Take the eight decisions currently open. Put each one on a list: cheap to undo, or expensive to undo.
The cheap list gets decided in the meeting. Out loud, with an owner and a date.
The expensive list gets a proper process, and now it gets the whole of it, because it is no longer queued behind six decisions that never needed one.
Most teams find their expensive list has two items on it. They had been running all eight at that speed.
