The Quarterly Meeting Audit: Cutting 6 of 19 Recurring Calls

Nineteen recurring meetings appeared on the average calendar, and six of them had not produced a documented outcome in the previous quarter. Removing those six returned 2.6 hours a week to each attendee, which is roughly a working day a month.

The audit covered a 24-person product organisation at a logistics software company. The organisation had added headcount quickly in the previous year and had never reviewed its calendar.

How the list was built

The list came from the calendar rather than from memory. Any event with three or more occurrences in the quarter and three or more attendees was included. Nineteen met the test across the organisation, which excluded one-to-ones deliberately.

Each entry got three fields: attendee count, total hours per quarter, and the date of the last documented outcome.

The outcome test

Documented outcome was defined narrowly: a decision, a plan or a change recorded in a system someone else could read. Verbal agreement did not count, which was controversial for a week and then accepted as the only definition that could be audited.

Six meetings had no documented outcome in the quarter. Four more had one, but it predated the quarter and repeated the previous one.

The six that went

Two were status meetings replaced by a written update. One was a cross-team sync whose two teams had stopped sharing work. One was a weekly review of a project that had shipped. One was a vendor call that the vendor had stopped attending. One was a meeting whose organiser had left the company eight months earlier.

The last one had continued to meet with nine attendees, which nobody had noticed because it was on a recurring invitation and had an agenda template that was never filled in.

The conversations

Removing a meeting requires a conversation, and the audit script cannot have it. Two organisers defended their meetings with a real argument: the meeting was the only place their teams heard from each other, and the outcome was undocumented but present.

Both were kept and given a written outcome requirement. Both produced a documented outcome in the following quarter, which suggests the argument was right and the documentation was the missing piece.

What the audit could not see

The audit measures documented outcomes, so it is blind to value that arrives as relationships or as early warnings. One cancelled meeting had been the place where a delivery risk surfaced twice, in both cases as an aside.

The team accepted that loss knowingly and replaced it with a fortnightly written risk note, which has surfaced one risk in three months rather than two.

The cost of running it

The audit took one person about six hours across two weeks, mostly reading calendars and asking organisers a single question. It requires calendar access across the organisation, which is a permission question before it is an analytical one.

The team now runs it twice a year. Once a year was too rare, since meetings accumulate faster than that.

One quarter later

Recurring meetings per person fell from 19 to 13. Meeting hours fell from 11.4 to 8.8 per week. The share of meetings with a documented outcome rose from 32 percent to 71 percent, partly because the requirement spread to meetings the audit never touched.

The meeting nobody could cancel

One of the cancelled meetings had to be reinstated after a month, because it was the only place where two teams discovered they were building the same thing. The audit had measured outcomes correctly and had missed the value of collision.

The replacement is a fortnightly written note listing what each team is currently building. It surfaced one duplication in three months. The meeting surfaced two a year, so the note is cheaper and slightly worse.

Nineteen recurring calls once sat on the average calendar and six produced nothing written. Thirteen remain, and the calendar now has room for the work those six were displacing.