What Six Months of Meeting Data Reveals About Your Team

Every recurring meeting is judged against a very short memory. You compare this week to last week, notice nothing dramatic, and move on. A five percent drop in how much a session actually accomplishes does not register against that baseline, because the baseline is itself drifting down at roughly the same rate. Each week feels normal because each week is only slightly worse than the one you have already accepted.

There is also no natural artifact that holds the pattern. Notes from a single meeting describe that meeting. The moment the next one happens, the previous notes are filed away and effectively gone. Nobody sits down to read six months of a meeting in a row, because doing it by hand would take most of a day and there is never a day. So the long slow shape of a meeting getting worse never assembles into something a person can see, which means it never gets fixed. It just continues. Microsoft's 2025 Work Trend Index, drawing on data from 31,000 workers across 31 countries, found that inefficient meetings are the number one barrier to productivity — and the recurring meetings where quality has quietly eroded are the hardest ones to catch, because everyone only ever compares this week to last. The same blind spot appears in the meeting metric most leaders never track.

Minuteory scores each meeting across the same six dimensions every time, which is what makes a trend possible in the first place. When the measure is consistent, the movement means something. A meeting that scored well on outcomes in March and mediocre in June has not changed its format, it has changed its results, and the score records that even when no one in the room felt it happen. For the single-session view, see how to measure whether a single meeting worked.

The piece that turns individual scores into a trend is meeting groups. When you connect a recurring or related series of meetings into a single thread, the sessions stop being isolated and become a narrative over time. You can then read the series as a sequence rather than a pile, and you can ask the assistant to generate an analytical report across that whole meeting group instead of one session. The half day of manual reading that nobody ever does becomes something the system does across the connected thread.

Reading a series this way changes what you notice. A single session gives you a verdict on one meeting. A connected thread gives you a direction, and direction is what you actually make decisions from. You stop asking whether last Tuesday's meeting was any good and start asking whether this meeting has been getting better or worse since the spring. That second question is the one worth answering, because it points at a trend with a cause, an agenda that drifted or a decision that migrated to a smaller group, rather than a single session that can always be explained away as an off day.

Because the scoring is organized by meeting type, the trend also tells you where to look. A weekly operations meeting and a monthly review are held to the criteria that fit each, so you can see that one meeting type is holding steady while another is sliding. That is a far more useful signal than a single company wide sense that meetings feel worse lately. It points at the specific recurring meeting that is drifting, while the others are fine.

A single meeting score answers whether that meeting worked. A trend answers a different and more important question: which of your recurring meetings is getting better, which is getting worse, and how fast. That is the difference between a snapshot and a diagnosis. A snapshot can be explained away as an off day. A downward trend across two months cannot, and it usually has a specific cause: an agenda that outlived its purpose, a participant list that was never updated, a decision that quietly moved to a smaller group and hollowed out the larger meeting.

The trend also protects you from the opposite error, cutting a meeting that is actually improving because it had one bad session. Over time the data separates a real decline from noise, so the decisions you make about which meetings to keep, reshape, or end are based on the pattern rather than the last thing that happened.

None of this asks you to become an analyst. The scoring already happens on every meeting, and connecting a recurring series into a thread is a one time setup. After that the trend assembles itself, meeting by meeting, until the shape that used to take half a day of reading to see is simply there when you open the group. The very thing that made the erosion invisible, the fact that nobody ever reads six months of a meeting in a row, is the work the system quietly takes off your hands.