When I started at Northwind Health, the CEO had forty-seven weekly recurring meetings on his calendar and was averaging three hours of unbroken focus time a week. He was tired, slightly behind on every important decision, and convinced the problem was him. It wasn't. The problem was the calendar.
This essay is the playbook I used to fix it — every question I asked, every meeting I killed, and the parts that made executives nervous before they got grateful.
The audit, step by step
I started by exporting six weeks of calendar data into a spreadsheet. Not three months, not a year — six weeks. Long enough to see the pattern, short enough that the data still felt recent. I tagged every recurring meeting by type, owner, frequency, and last-time-it-actually-mattered.
Before I touched anything, I asked the CEO one question: "If I could give you back five hours a week, what would you spend it on?" He had an answer ready. That answer became the brief for everything that followed.
The categories I ended up with were boring, which was the point. Standing 1:1s with direct reports. Cross-functional syncs. External relationship meetings. "Office hours." Investor check-ins. Operations reviews. Once each meeting had a label, the patterns started to surface — and they were uncomfortable.
What the data actually showed
Three things, mostly. First, seventeen of the forty-seven meetings had no clear owner. They had been put on the calendar by someone who had since left, or for a reason no one could remember. Second, eleven of them ran longer than the discussion required, simply because the calendar slot was an hour. Third — and this was the one that hurt — nine of them were duplicates in everything but name.
"The hardest part of a calendar audit isn't the spreadsheet. It's the conversations you have to have after."
The conversations
I interviewed every meeting owner. Not in a formal way — I caught them at coffee, in Slack DMs, in the five-minute window before another meeting. I asked three questions, always in this order:
- What is the outcome you are trying to get from this meeting?
- When was the last time you got it?
- If we cancelled this meeting next week, what would actually break?
The third question was the one that did the work. Most people, when pressed, could not name anything that would break. A few could, and those meetings stayed. The rest fell into one of three piles: kill it, shorten it, or convert it to async.
What we killed, what we kept
The final numbers looked like this:
The two-day batching was the move that surprised even me. Instead of spreading meetings across the week, we clustered them into Tuesdays and Thursdays — leaving Mondays, Wednesdays, and Fridays largely meeting-free. That single change is what turned three hours of focus time into eleven.
The result, six months in
The CEO is, by his own measure, more decisive than he has been in years. The direct reports who initially resisted the killed meetings now say they prefer the new rhythm — the written updates created a paper trail that was missing before, and the meetings that survived are sharper because there are fewer of them.
I will be honest: the first three weeks were uncomfortable. People felt the loss of meetings they did not realise they had been relying on as social glue. I had to defend a few of my recommendations more than once. But by week five, no one was asking to put anything back.
What I would do differently
Three things. First, I would communicate the audit before I ran it, not after. People resent surprises about their own calendars, even good ones. Second, I would build in a "review window" — a clear month after the changes where anyone could request a meeting back, no questions asked. Most won't. The option matters. Third, I would not present the final numbers as a victory. I would present them as the new baseline, with the assumption that we will need to audit again in twelve months.
The calendar is never finished. It is a living thing. The best you can do is keep an eye on it.