How many hours you spend in meetings says less about how you feel at work than whether those meetings are any good. In two 2006 surveys led by Steven Rogelberg, the larger of 676 full-time employees in the US and UK, how effective people judged their meetings was linked to their job satisfaction whatever their meeting load, while the link for sheer meeting time depended on the job and the person.¹

Why meetings fail, read across the meeting-science studies in this guide, comes down to three things: meetings designed badly, talk in the room that circles problems instead of solving them, and meetings that interrupt more work than they move forward. In a widely cited study of 92 videotaped team meetings, complaining and criticizing were more strongly linked to poor outcomes than problem-solving talk was to good ones.²

Why meetings fail: what meeting science studies

Meeting science studies the meeting itself rather than the team that holds it, and a 2026 Annual Review by Steven Rogelberg, Liana Kreamer and Jon Gray sorts about 30 years of it into five strands: meetings as stressors, as a tool for performance, as a platform for voice, as a stage for leadership and as an expression of culture.³

The review uses a standard definition. A work meeting is a purposeful, work-related interaction between at least two people, with more structure than a chat and less than a lecture; meetings are typically scheduled in advance and last 30 to 60 minutes on average. The authors are candid about the field’s limits: many of its measures were built for a single study and still need rigorous validation, and no theory unique to meetings has yet been proposed and tested.³ Most findings below come from surveys in which people rate their own meetings.

Bad meetings are common but not universal. In a 2015 study led by Jennifer Geimer, 1,081 employees from 41 countries, most in the US, UK and Australia, described their typical meetings in their own words. Of 967 comments coded for tone, 49 percent were positive, 30 percent negative and 21 percent ambivalent, so fewer than half called meetings an effective or good use of time.⁴

Company research points the same way. In Microsoft’s 2023 Work Trend Index, a survey of 31,000 full-time employed or self-employed workers in 31 markets run for Microsoft by Edelman Data x Intelligence, people ranked inefficient meetings as their number one productivity disruptor and too many meetings as number three.⁵ Microsoft sells meeting software, and the report is not peer-reviewed.

In the larger 2006 survey, employees reported spending an average of 5.6 hours a week in 4.2 scheduled meetings, and supervisors reported more.¹

Design failures: no agenda, late starts, overruns

Design, the choices an organizer makes before anyone arrives, is one of the best-studied reasons meetings fail. In two surveys published in 2009 by Desmond Leach and colleagues, using an agenda and having good facilities such as the room and equipment stood out among the design features that went with meetings rated more effective.⁶

The larger survey asked 958 employees in the US, UK and Australia about a typical week’s meetings; there, starting on time and ending when expected mattered too. Together, the design features accounted for an extra 20 percent of the variation in how effective people rated their meetings, after background factors were taken into account. Part of the link ran through involvement: well-designed meetings had wider participation, which went with higher ratings, and larger meetings went with lower involvement. The authors note that surveys like these cannot establish cause.⁶

Lateness has also been tested by experiment. Joseph Allen, Nale Lehmann-Willenbrock and Steven Rogelberg first found in an employee survey that meeting lateness was linked to lower meeting satisfaction and effectiveness. Then, in a lab experiment with 270 US undergraduate students, they randomly assigned 48 groups to start on time, 5 minutes late or 10 minutes late. Groups that started 10 minutes late rated their meeting as less satisfying and effective and produced fewer, lower-quality and less feasible ideas than on-time groups; a 5-minute delay made no significant difference. Late groups also had less time, as a late meeting with a fixed end would.⁷

A 2011 study summarized in the 2026 review adds that sending the agenda in advance and ending on time were related to how attendees rated a meeting’s quality.³ The open comments in Geimer’s study point to one more design failure: employees are often invited to meetings of little personal relevance, and many organizers skip basic practices such as circulating an agenda.⁴

Which kinds of meeting talk go with failure?

Complaining and criticizing during meetings went with worse team and company outcomes, and problem-solving and action planning with better ones, in a 2012 study by Simone Kauffeld and Nale Lehmann-Willenbrock that videotaped regular team meetings and coded what was said.²

The study

Moderate evidence

What was said in 92 team meetings, and how the teams fared (2012)

The researchers coded the interaction in 92 regular team meetings with a scheme called act4teams, and gathered team and company outcomes through questionnaires and telephone interviews. Teams that did more problem-solving and action planning were more satisfied with their meetings, and better meetings went with higher team productivity. Constructive meeting talk was also related to the organization’s success 2.5 years later. Criticizing others and complaining showed negative links with these outcomes, stronger than the positive links of constructive talk.²

The study observed real meetings instead of asking people to rate them, which is rare in this field. But it is a single sample, and Geimer and colleagues note its findings come from a German-speaking setting and cover only verbal behavior.⁴ Correlations also cannot show that complaining causes poor results; struggling teams may simply complain more.

The pattern reaches beyond the meeting. In a 2016 online survey of 440 working adults, Lehmann-Willenbrock, Allen and Belyeu linked counterproductive meeting behaviors to lower work engagement and higher emotional exhaustion, and good meeting behaviors to the reverse, through meeting satisfaction and perceived effectiveness.⁸ The 2026 review adds that interrupting and monopolizing the conversation have been shown to hinder speaking up and trust among coworkers.³ Meetings are one place where a team’s speaking-up climate, central to research on what makes a team effective, shows day to day.

Meeting quality and job satisfaction go together

How people rate their meetings is linked to how they feel about their jobs, so a failed meeting may matter beyond its hour. In the larger of Rogelberg’s 2006 surveys, perceived meeting effectiveness accounted for about 27 percent of the variation in job-related enthusiasm and 19 percent in job satisfaction.¹

A 2010 follow-up surveyed 201 and 785 working adults. In the larger survey, satisfaction with meetings still predicted overall job satisfaction after accounting for satisfaction with pay, promotion, coworkers, supervision and the work itself, though it added only about 1 percent to the variation explained, and the link was stronger for people who attended more meetings.⁹ Both studies are cross-sectional, so they show association rather than cause.

Myth

The fix for bad meetings is simply fewer meetings.

Fact

Fewer meetings address only part of the problem: in 2006 surveys, meeting effectiveness was linked to job satisfaction whatever people's meeting load, and in a one-day survey of 304 employees, for people who rated that day's meetings effective, more meetings went with higher job satisfaction.

When meetings interrupt more work than they do

Meetings also fail through volume and timing, and Rogelberg’s team treats many meetings as interruptions to people’s main work. In the 2006 surveys, a heavier meeting load went with lower job satisfaction for people with a strong drive to get their own tasks done, but with higher satisfaction for people less driven that way. In the larger survey, more meeting time went with higher satisfaction in jobs that depend on colleagues and lower satisfaction in jobs done mostly alone.¹

Daily patterns matter too. The 2026 review reports that a higher meeting load, in number and length, has been linked to more fatigue and perceived workload by day’s end, and that people who saw meetings scattered across their day ahead expected a worse mood and a less productive day.³ Scattered meetings leave short gaps, and after each one part of your attention can stay on the meeting you just left, the effect researchers call attention residue.

Perceived overuse matters as well. In two surveys of 174 and 213 working adults, published in 2019 by Joseph Mroz and colleagues, believing that one’s organization holds too many or overly long meetings was linked to rating team meetings as less effective and less satisfying, and to stronger intentions to quit. The data were cross-sectional and self-reported.¹⁰

Which meeting statistics can you trust?

Many of the most repeated meeting statistics rest on thin evidence. Writing in MIT Sloan Management Review in 2007, Steven Rogelberg, Cliff Scott and John Kello attributed the figure of about 11 million meetings a day in the United States to a 1998 white paper by MCI Conferencing, a company document rather than a peer-reviewed study.¹¹

The claim that senior leaders spend nearly 23 hours a week in meetings has a similar pedigree. The 2007 article attributes a figure of nearly 23 hours for senior managers to “some estimates,” and a 2017 Harvard Business Review article by Leslie Perlow, Constance Noonan Hadley and Eunice Eun, which puts executives’ average at nearly 23 hours, up from less than 10 in the 1960s, does not name the underlying research in its opening.¹¹¹² The same authors reported that in their own survey of nearly 200 senior executives only 17 percent said their meetings were generally productive uses of time; that is an opinion survey whose full method is not published.¹²

Even the 2026 review labels its figure of nearly 1 billion meetings a day worldwide as estimates and extrapolations.³ Our own search of research databases and the web could not trace the widely shared claim that bad meetings cost US businesses $37 billion a year to any study, so we do not treat it as a finding. Usage data measures time instead of estimating it, for a different group: in Microsoft’s Teams usage data for commercial customers, the heaviest 25 percent of meeting users spent 7.5 hours a week in meetings in early 2023.⁵

What the research suggests fixing first

Meeting research supports five changes, starting with punctuality, the only one tested in a randomized experiment, then agendas, invitation lists, steering talk toward solutions and asking attendees how meetings went. None has been tested in a large randomized field trial, so treat them as well-supported habits, not guarantees.

Before

• No agenda sent ahead
• People invited for whom it is not relevant

During

• Late start or overrun
• Complaining and criticizing instead of solutions

After

• No owners or next steps
• Nobody asks attendees how it went

1. Start on time and end when people expect

Punctuality is the one feature here with experimental support: in the lab, a 10-minute late start lowered students’ ratings of the meeting and the ideas their groups produced.⁷ In Leach’s larger survey, starting on time and ending when expected both went with higher effectiveness ratings.⁶ Start without latecomers, and if an item runs long, schedule it separately rather than overrunning.

2. Send an agenda ahead and work through it

In Leach’s second survey, of 292 people describing the last meeting they attended on a given day, a written agenda sent before the meeting and the extent to which the meeting worked through its agenda were both linked to higher effectiveness ratings.⁶ Send it early enough for people to prepare.

3. Invite people for whom the meeting is relevant

Larger meetings went with lower involvement, and involvement went with effectiveness.⁶ The 2026 review observes that leaders tend to over-invite and suggests asking relevant non-attendees for input beforehand and sharing notes afterward.³ In a two-wave survey of 319 working adults, employees whose managers kept meetings relevant, allowed voice and managed time reported higher work engagement.¹³

4. Steer the talk toward solutions and next steps

Name the problem, ask for options, and end each item with an action and an owner, the kind of problem-solving and action-planning talk that went with better outcomes in Kauffeld’s study.² When complaining starts to loop, ask what the group can do about it. In Microsoft’s survey, 55 percent of respondents said next steps at the end of a meeting are unclear.⁵

5. Ask attendees, not only the organizer

People who chaired a meeting rated it as more effective than other attendees did in Leach’s second survey.⁶ The 2026 review recommends that organizations track meeting volume and satisfaction, for example in engagement surveys.³ Asking may itself help: in a 2026 preregistered field experiment by Microsoft researchers with 361 employees of a global technology company, prompts to reflect on meeting goals beforehand made no statistically significant difference to rated effectiveness, but the authors report that the post-meeting surveys both groups answered seemed to work as an intervention of their own.¹⁴

WiserHours

Six habits for the next meeting you run

• Start on time, even if people are missing
• Send the agenda in advance
• Invite only people with a reason to be there
• End each item with an action and an owner
• Finish when you said you would
• Ask two attendees how it went

Find this checklist online:
https://wiserhours.com/meetings/why-meetings-fail/#checklist-why-meetings-fail-fixes

The table below sums up how strong the evidence is for each cause of failure covered here.

The bottom line

Fix how meetings run before you count how many there are: in Rogelberg’s 2006 surveys, how effective meetings were went with job satisfaction more consistently than how many hours they took. The failures to fix first are the ones this guide’s studies keep finding: no agenda, late starts and overruns, the wrong people in the room, and talk that turns to complaint instead of decisions and owners.