The Science of Meetings – Leading Better with Rogelberg
As a project manager, you spend a large part of your working week in meetings — and run many of them yourself. Yet most of us never receive any training in how to hold meetings. Organizational psychologist Steven Rogelberg has studied meetings for decades, and in his book The Surprising Science of Meetings he sums up what actually separates meetings that create value from meetings that merely steal time.
His conclusion is a hopeful one: the problem is not meetings themselves, but how they are led. And you can do much better with small means — no new tools required, just more deliberate meeting leadership.
What does the research say?
One of Rogelberg's most thought-provoking findings concerns self-perception: meeting leaders systematically rate their own meetings as better than the participants do. The person who calls the meeting talks the most, controls the agenda, and therefore experiences the meeting as worthwhile — while the participants sit quietly counting the minutes. That means you, as the meeting leader, cannot trust your own gut feeling; you have to actively ask and observe.
The research also points to what characterizes good meetings. They have a clear purpose — a meeting with no question to answer or decision to make should often be an email instead. They have a tight agenda, and Rogelberg suggests phrasing it as questions to answer rather than headings, which makes it obvious when the meeting is done. They are no longer than they need to be: meetings tend to fill the time that was booked, so shorter default durations sharpen focus. And they have the right participants — better a few people who are directly affected than many who "maybe should be there".
Finally, Rogelberg emphasizes active facilitation: the meeting leader's job is to draw in the quiet ones, rein in the dominant ones, and protect the purpose of the meeting. Techniques like having everyone write down their thoughts before the discussion starts give quieter participants a way in and reduce the risk that the first voice steers the entire conversation.
What does this mean for you as a project manager?
Phrase the agenda as questions. Don't write "Delivery status" — write "Can we hold the delivery date of the 15th — and if so, what will it take?". Then everyone knows why the item exists, who needs to be there, and when the question has been dealt with — and the meeting gets a natural ending.
Shrink the meeting — in both duration and attendee list. Try 25 minutes instead of 30, 45 instead of 60. Invite only the people needed for the questions, and send notes to everyone else. That is respect for other people's time, not stinginess — and it also makes the conversation in the room both faster and more honest.
Ask how the meeting was. Since your own experience as the meeting leader is unreliable: end with a one-minute evaluation now and then, or ask a participant afterwards what could have been better. Small adjustments based on honest answers do more than a perfect template.
How to practice this
Facilitating for real — interrupting a monologue kindly but firmly, drawing out the quiet expert's objection, steering a high-pressure steering group meeting all the way to a decision — these are skills you build through practice, not through reading. In Project-simulator you practice meeting conversations against AI counterparts in a safe simulator environment and get evidence-based feedback grounded in what you actually said. You can try it free for a week.
In Project-simulator you practice these conversations against an AI counterpart and get feedback grounded in research like this.
Try free for 7 days