Only 9% of managers say they can rely on colleagues in other functions and units all the time, against 84% for their own reporting line.
Why it matters when the plan changes
A plan is approved in a single meeting. The organisation that has to carry it was built for the previous plan, and its priorities, decision rights, meetings, incentives and habits still point there. Execution risk is what sits in that gap. It is not a judgement about whether the plan is good or whether the people are capable; it is a reading of whether this arrangement can carry this demand, within a stated horizon.
The tension is that execution risk is usually named late and named wrongly. A slow decision reads as a leader without pace, a missed milestone as weak discipline. Those are hypotheses, not findings. Naming the risk early, with the evidence that supports and contradicts it, is what turns it into something a leader can act on before the cost arrives.
In practice
A group announces a new operating model in January. By March two milestones have slipped and the explanation in the room is that the regional leads are slow. The plan actually depends on four handoffs between product and regional teams that no single person owns end to end. The risk was never in the pace of any individual; it was in an interface nobody had been given. Reading it that way changes the move from replacing a leader to assigning an owner.
Evidence
Managers report they can rely on colleagues in other functions and units far less often than they can rely on their own reporting line.
Donald Sull, Rebecca Homkes and Charles Sull, Why Strategy Execution Unravels and What to Do About It, Harvard Business Review (2015)The number of communication paths inside a group grows faster than the group itself, so coordination cost rises with every person added.
Brooks's law, after Frederick P. Brooks, The Mythical Man-Month (1975)
What it cannot tell you
Execution risk describes the fit between a plan and the organisation asked to carry it. It says nothing about whether the plan is commercially sound, whether the market timing is right or whether the numbers behind it hold. A red rating does not mean the strategy is wrong, and a green one does not guarantee that outside conditions will stay favourable.
How Atlas reads it
Execution risk is the question the whole method exists to answer. The organisation is read in five layers, the reading is turned into a time-bounded forecast with evidence on both sides, and the forecast carries a rating, a confidence, an owner and a review date. The rating describes the arrangement around a person, never the person.
Questions
Business risk sits in the market, the numbers and the outside world. Execution risk sits inside the company, in roles, ownership, dependencies and operating patterns. Harvard Business Review's 2015 research found only 9% of managers rely on colleagues outside their own reporting line, evidence of exactly where execution risk lives instead of commercial conditions.
Before commitment, and again whenever the plan materially changes. Six events warrant a fresh reading: new strategy, new ownership, new operating model, new pressure, new dependency and new evidence. Waiting for friction to become failure removes the options that early information would have kept open.
Neither alone. It is a property of the fit between the two: what the plan now demands, and what the roles, ownership and dependencies around a person actually support. The rating describes the arrangement, never a person's worth, and the first move is usually structural.
In four states. Green means the arrangement is in sync with the demand, yellow means it needs adjustment, red means it slows execution, and blind means the signal is insufficient and that is said plainly rather than guessed. Each rating carries a confidence and an owner.
The plan itself, the roles and decision rights around it, the dependencies it relies on, observed operating patterns, and behavioural evidence read against the demand. This includes structural reasoning such as Brooks's law from The Mythical Man-Month, 1975, on how coordination cost rises as more people are added. Every material claim carries its source and confidence.