Chief of Staff interview questions

I would design it around the decisions the team must make, not around a standard meeting calendar.

• I would map recurring decisions to weekly, monthly, and quarterly forums, with one accountable chair and a written pre-read for each.
• A single decision log would capture the owner, commitment, due date, and what evidence would reopen the call.
• After two cycles, I would remove forums that only share status and move that information to an asynchronous update.

I would preserve their functional cadences and add a thin shared layer for cross-functional commitments.

• I would identify the few dependencies that need both executives, then give each one a single accountable owner rather than two co-owners.
• The shared review would cover only decisions, missed handoffs, and resource conflicts, with routine status left in the functional forums.
• I would test the design for one quarter and judge it by fewer unresolved dependencies, not meeting attendance.

I make every forum earn its place through a defined decision or intervention.

• Each agenda item states the decision needed, the owner, and the pre-read; an item without an ask stays asynchronous.
• I track decisions that are reopened, actions that age past due, and issues repeatedly escalated without resolution.
• If a meeting produces only updates for two cycles, I replace it with a written digest and return the time to the executives.

I would give the COO room to lead while protecting the institutional memory behind the current system.

• I would brief them on which forums make real decisions, which are weak, and which commitments cannot pause during the transition.
• We would run one diagnostic cycle before changing the full design, unless there is a clear control or accountability failure.
• I would make the COO the visible sponsor of the revised cadence and document what changed so the team does not treat it as a temporary preference.

The CEO should chair only when their authority or judgment materially changes the outcome.

• Company-level resource allocation, strategy shifts, and unresolved executive conflicts usually warrant the CEO in the chair.
• Operating reviews with established decision rights should sit with the relevant executive, with the CEO receiving exceptions rather than every detail.
• I revisit the chair when the forum stabilizes, because a meeting that always needs the CEO often has weak ownership underneath it.

I would separate the pricing decision into distinct rights instead of forcing one owner onto the entire subject.

• Product can own packaging options, the CFO can own economic guardrails, and the CEO can approve changes that alter company positioning.
• I would write who recommends, who decides, who must be consulted, and which threshold triggers escalation.
• We would test the model on the next pricing change and fix any ambiguity exposed by the actual decision, not debate the matrix in theory.

I first determine whether the decision was weakly made or whether the executive is avoiding commitment.

• The decision log should show the original call, assumptions, owner, and explicit conditions that justify reopening it.
• If new evidence meets those conditions, I schedule a deliberate review; if not, I ask the CEO to reinforce the agreed right and move execution forward.
• Repeated reopening is then addressed as an accountability issue, because the hidden cost is work paused across several teams.

I define escalation as a response to a blocked decision, not as a substitute for working through conflict.

• Teams must state the decision needed, options considered, their recommendation, and the cost of waiting before an executive takes it.
• Clear thresholds such as budget exposure, legal risk, or cross-functional deadlock determine which level receives the issue.
• I review recurring escalations quarterly, because repeated topics usually reveal missing decision rights or an executive avoiding ownership.

I would protect the delegated authority while helping the CEO decide whether the underlying boundary was wrong.

• I would confirm that the executive followed the agreed process and acted within the stated guardrails.
• The CEO can change the policy prospectively, but reversing a valid call without new evidence teaches the team that delegation is not real.
• I would facilitate a direct conversation and update the decision-rights document only if the scope or escalation threshold genuinely needs to change.

I look for faster, more durable decisions at the right level, not whether everyone can recite the model.

• I sample important decisions for cycle time, number of escalations, and whether the final owner was clear from the start.
• I also watch for decisions made too low, such as commitments that breach financial or legal guardrails.
• If the same issue reaches the CEO twice, I treat that as evidence to clarify the boundary or strengthen the accountable executive.

I would force the portfolio into a ranked set of company bets with explicit capacity behind each one.

• Every initiative needs an executive sponsor, one accountable operator, a measurable outcome, and a clear reason it belongs at company level.
• I would show the leadership team the total people and budget consumed, including shared teams that are often double-counted.
• New work enters only by displacing an existing bet, which turns priority language into a real allocation decision.

I set stop criteria when the initiative is approved, before sunk cost and executive identity distort the discussion.

• The criteria usually cover the outcome threshold, time to evidence, total investment, and a strategic assumption that must remain true.
• At review, I distinguish a weak execution plan from a disproven thesis, because the first may need a new owner while the second should end.
• A stop recommendation includes where the people and budget go next, so the decision creates capacity rather than just disappointment.

I would review the thesis before celebrating execution against an obsolete plan.

• I would restate the original assumption, show the new evidence, and quantify how it changes the expected value of the program.
• The choices should be explicit: continue, reshape the bet, pause for evidence, or stop and redeploy the team.
• I would not let delivery progress become the reason to continue, because completing the wrong program efficiently still destroys capacity.

I own the portfolio seams and require each initiative leader to own delivery inside their boundary.

• A shared dependency map identifies only cross-initiative handoffs, scarce resources, and decisions that need executive arbitration.
• Initiative leaders bring a recommendation when a dependency breaks; my role is to route the decision and expose the portfolio consequence.
• If I am chasing task updates, the ownership model is wrong, so I reset the accountable operator rather than absorb their job.

I would make the review a set of allocation decisions, not a parade of project updates.

• The pre-read shows each bet's intended outcome, current evidence, resources consumed, key risk, and recommendation.
• Meeting time goes to initiatives that need a change in funding, owner, scope, or stop decision; healthy work stays in the appendix.
• The output is a revised portfolio with named decisions and resource movements, followed by a short company narrative explaining what changed.

I would add a company-level integration step before plans turn into fixed budgets.

• Each function must show which strategic choice it supports, the outcome it owns, and the dependencies it needs from other teams.
• I would convene cross-functional working sessions around the few company bets, not review functions one by one in isolation.
• The final plan should expose unresolved capacity conflicts for executive decisions rather than hide them in optimistic assumptions.

I keep the strategic choices stable and use quarterly planning to update the evidence and sequence of bets.

• A small set of strategic outcomes anchors the year, while quarterly commitments state what the company must learn or deliver next.
• New information can change investment levels or timing without changing the underlying choice unless a core assumption breaks.
• I document those assumption changes explicitly so normal execution variance does not trigger a fresh strategy exercise.

I would build a base plan plus a small number of funded strategic scenarios instead of forcing false precision.

• The base covers committed operations, while each unresolved choice has a clear cost, trigger date, and decision owner.
• I would align with the CFO on the latest date each choice can remain open without harming hiring or cash planning.
• Once a trigger is reached, the leadership team selects the scenario and releases or removes the reserved capacity.

I would make the contingency visible and centralize it rather than let every function hide a private buffer.

• We would separate committed roles from roles tied to a named growth, revenue, or workload trigger.
• The CFO and CEO would hold a company reserve that can be released when those triggers are met.
• This protects flexibility while preventing ten reasonable local buffers from creating an unaffordable company plan.

I would separate target setting from plan building and resolve the few disputed assumptions first.

• The CEO and CFO should set company guardrails, then functions propose how to meet them rather than renegotiate the ambition in every review.
• I would isolate disagreements about market demand, productivity, or investment and assign one owner to produce the deciding evidence.
• After the cycle, I would shorten templates and remove reviews that did not change a decision.