Unclear decision rights are the silent tax of a growing company.
At 20 people, the founder decides almost everything and everyone knows it. At 80 or 200, the same habit creates queues, second-guessing and decisions that get re-opened in every meeting. Managers already spend a large share of their time here: a 2019 McKinsey survey of 1,228 managers found they spend on average 37% of their working time making decisions (McKinsey, 2019). Confidence in the result is low. Gallup reported in 2020 that only 24% of U.S. managers strongly agree their peers make well-thought-out decisions, and just 14% are satisfied with the speed of decision-making (Gallup, 2020). Gallup does not state the sample on that page, and the Booz data is older and self-reported, so treat the numbers as direction, not precision. The direction is clear, though.
Start with a decision map, not a framework name.
RAPID, DACI and RACI all help, but the label matters less than writing down your real decisions. List the 15 to 25 decisions that come up every quarter: pricing changes and discounts above a threshold, senior hires, budget reallocations, new markets, product roadmap trade-offs, vendor contracts, policy exceptions. For each one, record five things in one table:
- Owner: the one person who makes the call. Never a committee, never "the leadership team".
- Input: who must be heard before the decision, and by when.
- Veto or agree: who can block it, and on what grounds only (legal, budget, safety).
- Informed: who needs to know the outcome to do their job.
- Escalation: the specific condition that moves the decision up a level.
Match the owner to the stakes and the reversibility.
A useful rule is to push reversible, low-cost decisions as far down as competence allows, and to keep irreversible, high-cost ones close to the CEO or board. Set thresholds in numbers, not adjectives: "discounts up to 15% sit with the sales lead; above that, the CFO decides" is a decision right; "significant discounts need approval" is not. Then check three warning signs: the same decision has two owners, a decision has an owner who lacks the information to make it, or a veto holder uses the veto to re-open the substance rather than to protect the area they are responsible for.
More on Decision Science — every week.
Join 2,400+ leaders getting the decision intelligence briefing.
Make escalation rare and predictable.
Escalation should be triggered by conditions, not by discomfort. Good triggers are cost above a fixed amount, impact on more than one function, a legal or regulatory question, or a material change to a strategic commitment. When a decision escalates, the owner still prepares the recommendation, so the next level decides between clear options instead of starting over. Once a decision is made, record it, including the owner and the reasoning; a shared decision log stops the same question from coming back every month.
Hypothetical example.
Imagine a 140-person B2B software company where the CEO, the CRO and the CFO all believe they own pricing exceptions. Large deals stall for a week while three people trade emails, and two customers receive different terms for the same package. The leadership team maps the decision in one session: the CRO owns discounts up to 20%, the CFO must be consulted above 10% and owns anything above 20%, the CEO is informed weekly, and any non-standard contract term escalates to the CFO and legal. The company now has a written rule that anyone can read and test. If it does not cut the delay, they revise the thresholds at the next quarterly review instead of arguing deal by deal.
Review the map when the company changes shape.
Decision rights drift after funding rounds, new executive hires, acquisitions and reorganizations, which are exactly the moments when clarity matters most. Re-read the map whenever a new executive joins and at least once a year, and ask each owner one question: which of your decisions did someone else actually make last quarter? Where a single high-stakes decision has unclear ownership and strong disagreement, it can help to run it through a structured process first: our AI decision council (try it at no cost) makes the interpretations, objections and constraints visible before a human owner signs off, and a Decision Sprint turns one such decision into a decision-ready brief.
A one-page checklist to start this week.
Write down your 15 to 25 recurring decisions. Give each one a single owner. Add numeric thresholds and escalation triggers. Limit vetoes to defined grounds. Publish the map where the whole leadership team can see it, log decisions as they are made, and review it at the next leadership offsite. If your decisions involve employment law, works-council (Betriebsrat) participation or corporate governance rules, check them with qualified advisors; this article is general guidance, not legal advice.
Research context
In a McKinsey survey, 72% of senior executives said bad strategic decisions were as frequent as good ones or were the prevailing norm.
McKinsey & Company: Untangling your organization's decision making (2017)This independently published statistic provides context for the topic; it does not substantiate every claim in this essay.
One decision framework, every week.
The week's best articles on making and executing high-stakes decisions, backed by sourced research.