For founders: your calendar is your strategy document.

Michael Porter and Nitin Nohria tracked 60,000 hours of 27 CEOs’ calendars for their 2018 HBR study (“How CEOs Manage Time”). The headline number: the average CEO works 62.5 hours a week and spends 3% of it with customers. Less than with consultants. For founders who built their company by being in front of customers, that number is the starkest warning shot in the research.

I’ve sat across from a lot of founders with calendars that look like that. The conversation always starts the same way — “I don’t know where the week went” — and it always ends the same way: a pared-back calendar that stops pretending every meeting is load-bearing.

01. The 300,000-hour meeting

Bain consultants Mankins, Brahm and Caimi reported a finding in HBR (“How a Weekly Meeting Took Up 300,000 Hours a Year”) that I quote to every founder I coach: at one of their clients, a single weekly executive committee meeting consumed approximately 300,000 hours a year once all the downstream prep meetings — the pre-reads, the pre-meeting-meetings, the department-level rollups — were counted.

That is not a big-company problem. It’s a pattern. The moment you have three direct reports who each have two direct reports, every weekly meeting you call creates a cascade of meetings underneath it. Your one hour is twenty hours somewhere in the org. You are almost never aware of this cost.

02. One-way doors vs two-way doors

Jeff Bezos’s 1997 shareholder letter (SEC filing) framed decisions as two types. One-way doors are irreversible — hiring a VP, raising a round at a given valuation, taking on specific debt. Two-way doors are reversible — launching a feature, changing a website copy, running an experiment. Most decisions are two-way doors, which means most decisions don’t need the meeting apparatus founders build around them.

The single highest-leverage change I’ve seen founders make is to run two-way-door decisions in a doc, over 24 hours, with a designated decider. No meeting. The rate at which decisions happen approximately doubles. Nobody wants to admit how much of the calendar was defending reversible choices.

03. Andy Grove’s leverage formula

In High Output Management (1983), Andy Grove defined managerial leverage as output divided by time invested.Every meeting is on the denominator. High-leverage meetings are the ones where your presence directly enables someone else’s output — approving a hire, removing a blocker, making a go/no-go call. Low-leverage meetings are the ones where you’re briefed on things you didn’t need to know.

Run your recurring calendar through that filter. For each meeting, ask: what output did I produce? If the honest answer is “I was updated,” the meeting is probably a doc.

04. The founder multiplier

On the calculator, you can pick an opportunity-cost multiplier. For most roles, we recommend 1.5×–2×. For founders, the real number is higher, and I’ll tell you why: your time gates everyone else’s. A decision you delay by a week is a week of delay for the people waiting on it. For a founder with eight direct reports, every hour you sit in a low-value meeting is eight hours of downstream drag. 2.5–3× is closer to honest for founders of any company past a dozen people.

05. 1:1s that do what they should

Ben Horowitz’s rule in The Hard Thing About Hard Things is that in a 1:1, the manager should talk 10% of the time and listen 90%. The meeting belongs to the employee. They set the agenda, they surface the issues, they use the time.

Almost no founders I’ve worked with run 1:1s this way by default. They arrive with a list of things they want to tell the person. The signal that you’re doing it wrong is that you’re doing most of the talking. The signal that you’re doing it right is that the report leaves with an answer to something, and you leave with an update about how the person is doing.

06. What to cut this week

If you have exactly one hour to spend on this, here’s the audit: export your recurring calendar, run every meeting through the calculatorat a 2.5× founder multiplier, and for the top five by cost, ask what one-way-door decision they produce. The ones that don’t produce one are your candidates to cut. Do it as a reversible experiment — put them back if you miss them — and see which ones actually come back.

Most don’t.