I Spent Three Months Fixing Our Meeting Problem. Here Is What Worked.

I counted my meetings last month. 67. That's 50 hours in a 160-hour month spent talking instead of doing. I manage two engineering teams across three time zones, and somewhere between the daily standups, sprint planning, stakeholder syncs, and "quick alignment calls," I stopped having time to do my actual job.

I decided to fix it. Not with a policy announcement or a Slack message about "meeting culture." With experiments, tools, and an honest look at what worked and what I wasted time on. This is that story.

The data says I'm not alone

Before I started changing anything, I wanted to know if my calendar was abnormal or just normal-bad. Turns out it's normal-bad.

Microsoft's 2024 Work Trend Index measured a 153% increase in weekly meetings since February 2020. Calendly's 2024 data showed the median professional attending 15+ scheduled meetings per week. Clockwise found meeting load grew 12% year over year with no sign of flattening.

So my 67 meetings in a month? That's about 16 per week. Right on the trend line. I wasn't an outlier. I was average. That was the depressing part.

At 45 minutes per meeting on average, 16 per week works out to 12 hours. That leaves 28 hours for actual work. And those 28 hours aren't contiguous; they're chopped into 30- and 60-minute fragments between meetings. Research from Microsoft's own labs shows that this kind of fragmentation cuts deep work capacity by another 20-30%. I wasn't getting 28 hours of productive time. I was getting maybe 20.

What I tried first (and failed)

"No meeting Wednesdays." This lasted exactly two weeks. The first week was great. Empty Wednesday, real work got done. The second week, someone scheduled a "critical" client call on Wednesday because "it's the only time they're free." By week three, Wednesdays looked like every other day. The problem with no-meeting days is that they require universal buy-in, and one exception kills the norm. I'd need executive backing and enforcement to make it stick, and I didn't have either.

Meeting audits. I asked my teams to justify every recurring meeting. Could it be an email? A Slack message? A doc? The exercise felt productive for about a day. Then people got defensive. "Are you saying my meetings are a waste of time?" One team lead told me it felt like a performance review of her management style. She wasn't wrong; that's how it landed, even though I meant it as housekeeping. I backed off. The two meetings we did cancel reappeared under different names within a month.

Time limits on all meetings. I mandated 25-minute meetings instead of 30, and 50 instead of 60. The idea was to create natural breaks and force agendas. What actually happened: people just booked more meetings. Instead of one 60-minute planning session, teams booked two 25-minute sessions. Total meeting time went up, not down. My mistake was constraining the shape of meetings without reducing the demand for them.

What actually moved the needle

After three failed experiments, I tried a different approach. Instead of telling people to meet less, I gave them tools that made meetings unnecessary for specific use cases.

Async standups via Loom. I replaced our Monday standup with a 3-minute Loom from each team lead. Saved 5 hours per week across the team. Some people hated it. They missed the "energy" of the live standup. Most adapted in two weeks. The key was that Loom recordings were short enough to watch at 1.5x speed and had threaded comments for follow-up. A 3-minute Loom watched at 1.5x takes 2 minutes. A 30-minute standup takes 30 minutes no matter how fast you listen. The math was obvious once people experienced it. (If you're evaluating async video tools, our take on which ones survive the consolidation is worth reading.)

AI meeting summaries for skip-able meetings. I set up Fathom on every meeting I was "optional" on. Instead of attending 45-minute calls to catch 5 minutes of relevant content, I read the AI summary afterward. That cut 4 meetings per week from my calendar. The summaries weren't perfect, maybe 80% accuracy on action items. But 80% of the context in 3 minutes of reading beats 100% of the context in 45 minutes of sitting there.

Notion docs for decisions. We moved all non-urgent decisions to Notion. Post the proposal, share it in Slack, set a 48-hour comment window, then resolve. The first month was rough. People kept requesting "a quick call to discuss the doc." I pushed back every time: "Leave your questions as comments. If we can't resolve in writing, we'll schedule a call." About 70% of decisions never needed that call. The other 30% still happened as meetings, but shorter ones because everyone had already read the proposal.

Reclaim.ai for focus time. This was the tool that surprised me. Reclaim auto-blocks focus time on your calendar and defends it against meeting invites. It doesn't reduce meetings directly. It makes sure the gaps between meetings are big enough to use. Before Reclaim, my longest uninterrupted block was 45 minutes on most days. After, I consistently got 2-hour blocks three times per week. That changed how much real work I could do between meetings.

The math, honestly

For my two teams (10 people total), here's the rough ROI. I'm including the caveats because the clean version is misleading.

Before: 10 people x 12 hours/week in meetings = 120 person-hours per week. At an average fully loaded cost of $75/hour, that's $9,000/week in meetings, or $36,000/month.

After: 10 people x 7 hours/week = 70 person-hours per week. Cost: $5,250/week, or $21,000/month.

Monthly savings on paper: $15,000 in recovered productive time.

Monthly tool cost:

• Loom Business: 10 users x $12.50 = $125
• Fathom Pro: 10 users x $19 = $190
• Reclaim.ai: 10 users x $8 = $80
• Slack and Notion: already in our stack

Total tool spend: $395/month.

These numbers assume people actually watch the Looms and read the Notion docs. In practice, maybe 70% do. A few people on my teams still ask me to "recap" the async update in person. Some skip the Notion doc and come to the meeting unprepared. The theoretical 5-hour savings per person is closer to 3.5 hours in reality, because human behavior doesn't optimize as cleanly as a spreadsheet.

Still, even at 3.5 hours saved per person per week across 10 people, that's 35 person-hours. At $75/hour, $2,625/week against $395/month in tools. The tools pay for themselves in two days. That part of the math I'm confident about.

What I'd tell you if you're where I was

Don't announce a "meeting reduction initiative." People hear "your meetings are bad" and get defensive. Instead, just start recording async updates and let people experience the time savings. Within a month, they'll stop requesting the live meetings on their own.

Start with one meeting type. I started with standups because they're the most obviously replaceable. Weekly syncs came next. Don't try to overhaul everything at once. Pick the meeting that wastes the most collective time and replace it. Let success build momentum.

Accept that some people will never go async. Two people on my team still prefer live standups. I let them have a 10-minute call on Mondays while everyone else watches the Looms. Forcing 100% adoption creates more friction than it's worth. Aim for 80%.

1:1s are sacred. Don't cut them. I shortened mine from 45 to 25 minutes by having both parties pre-share agenda items in Notion. But I never canceled them. 1:1s are the one meeting type where the relationship value matters more than the information exchange. Don't optimize the humanity out of your team's week.

Budget 4-6 weeks for the team to adjust. The first two weeks are the worst. People feel disconnected. They worry they're missing things. They schedule "catch-up" calls that partially undo the savings. By week four, the new rhythm clicks. By week six, nobody wants to go back.

Measure before you start. I counted my meetings before I changed anything. That baseline made it obvious, to me and to my manager, that the experiment was worth running. If I'd just said "I think we have too many meetings," nobody would have cared. "I spent 50 hours in meetings last month" got attention.

My calendar isn't perfect. I still have 7 hours of meetings per week, and some of those are still "alignment" calls I can't escape. But 7 is not 12. I got 5 hours back. My teams got a similar amount. The tools exist. The math works. The hard part is the first month.