The hidden tax of forgetting: an open study of what meetings lose

Why this study

Claims about how much people forget after meetings are everywhere, usually unsourced, and usually made by people selling a fix. We sell one too, which is exactly why we are publishing the method first and committing to publish whatever we find.

The questions

1. How accurately do participants recall decisions, numbers and commitments one hour, one day and one week after a meeting?
2. Do participants in the same meeting disagree about what was decided, and how often?
3. How much time is spent recovering lost information: re-asking, re-meeting, searching?

Method

• Participants: professionals whose work happens in conversation (investors, consultants, founders, operators), recruited openly.
• Design: participants nominate real meetings they are already holding. Shortly after, and again at one day and one week, they answer a short recall form. Where all participants agree, a reference record of the meeting is used to score accuracy.
• Privacy: no meeting content is published. Scoring compares answers to the reference record; only aggregate, anonymised measures are released.
• Consent: every person in a scored meeting must agree. If one person declines, that meeting is not used.

What we expect, written down in advance

We expect recall of exact numbers to fall faster than recall of decisions, and participants in the same meeting to disagree on at least one commitment in a meaningful share of meetings. We are stating these now so that we cannot quietly move the goalposts later. If we are wrong, we will say so.

What we will publish

• The full method, recall form and scoring rubric, before data collection begins.
• Aggregate results with sample sizes and confidence intervals.
• Limitations, including self-selection: people who sign up for a memory study may not be typical.

Take part

We are looking for people who hold at least five work conversations a week and are willing to answer three short forms per meeting. Use the request form below and mention "study" in the second step.

Questions people ask