hn-classics/_stories/1976/13177611.md

---
created_at: '2016-12-14T17:14:58.000Z'
title: How to run a meeting (1976)
url: https://hbr.org/1976/03/how-to-run-a-meeting
author: trendoid
points: 147
story_text: 
comment_text: 
num_comments: 44
story_id: 
story_title: 
story_url: 
parent_id: 
created_at_i: 1481735698
_tags:
- story
- author_trendoid
- story_13177611
objectID: '13177611'

---
![MAR15\_12\_523197071](/resources/images/article_assets/1976/03/MAR15_12_523197071.png)

Why have a meeting anyway? Why indeed? A great many important matters
are quite satisfactorily conducted by a single individual who consults
nobody. A great many more are resolved by a letter, a memo, a phone
call, or a simple conversation between two people. Sometimes five
minutes spent with six people separately is more effective and
productive than a half-hour meeting with them all together.

Certainly a great many meetings waste a great deal of everyone’s time
and seem to be held for historical rather than practical reasons; many
long-established committees are little more than memorials to dead
problems. It would probably save no end of managerial time if every
committee had to discuss its own dissolution once a year, and put up a
case if it felt it should continue for another twelve months. If this
requirement did nothing else, it would at least re-focus the minds of
the committee members on their purposes and objectives.

But having said that, and granting that “referring the matter to a
committee” can be a device for diluting authority, diffusing
responsibility, and delaying decisions, I cannot deny that meetings
fulfill a deep human need. Man is a social species. In every
organization and every human culture of which we have record, people
come together in small groups at regular and frequent intervals, and in
larger “tribal” gatherings from time to time. If there are no meetings
in the places where they work, people’s attachment to the organizations
they work for will be small, and they will meet in regular formal or
informal gatherings in associations, societies, teams, clubs, or pubs
when work is over.

This need for meetings is clearly something more positive than just a
legacy from our primitive hunting past. From time to time, some
technomaniac or other comes up with a vision of the executive who never
leaves his home, who controls his whole operation from an
all-electronic, multichannel, microwave, fiber-optic video display dream
console in his living room. But any manager who has ever had to make an
organization work greets this vision with a smile that soon stretches
into a yawn.

There is a world of science fiction, and a world of human reality; and
those who live in the world of human reality know that it is held
together by face-to-face meetings. A meeting still performs functions
that will never be taken over by telephones, teleprinters, Xerox
copiers, tape recorders, television monitors, or any other technological
instruments of the information revolution.

## Functions of a Meeting

At this point, it may help us understand the meaning of meetings if we
look at the six main functions that meetings will always perform better
than any of the more recent communication devices.

1\. In the simplest and most basic way, a meeting defines the team, the
group, or the unit. Those present belong to it; those absent do not.
Everyone is able to look around and perceive the whole group and sense
the collective identity of which he or she forms a part. We all know who
we are—whether we are on the board of Universal International, in the
overseas sales department of Flexitube, Inc., a member of the school
management committee, on the East Hampton football team, or in Section
No. 2 of Platoon 4, Company B.

2\. A meeting is the place where the group revises, updates, and adds to
what it knows as a group. Every group creates its own pool of shared
knowledge, experience, judgment, and folklore. But the pool consists
only of what the individuals have experienced or discussed as a
group—i.e., those things which every individual knows that all the
others know, too. This pool not only helps all members to do their jobs
more intelligently, but it also greatly increases the speed and
efficiency of all communications among them. The group knows that all
special nuances and wider implications in a brief statement will be
immediately clear to its members. An enormous amount of material can be
left unsaid that would have to be made explicit to an outsider.

But this pool needs constant refreshing and replenishing, and
occasionally the removal of impunities. So the simple business of
exchanging information and ideas that members have acquired separately
or in smaller groups since the last meeting is an important contribution
to the strength of the group. By questioning and commenting on new
contributions, the group performs an important “digestive” process that
extracts what’s valuable and discards the rest.

Some ethologists call this capacity to share knowledge and experience
among a group “the social mind,” conceiving it as a single mind
dispersed among a number of skulls. They recognize that this “social
mind” has a special creative power, too. A group of people meeting
together can often produce better ideas, plans, and decisions than can a
single individual, or a number of individuals, each working alone. The
meeting can of course also produce worse outputs or none at all, if it
is a bad meeting.

However, when the combined experience, knowledge, judgment, authority,
and imagination of a half dozen people are brought to bear on issues, a
great many plans and decisions are improved and sometimes transformed.
The original idea that one person might have come up with singly is
tested, amplified, refined, and shaped by argument and discussion (which
often acts on people as some sort of chemical stimulant to better
performance), until it satisfies far more requirements and overcomes
many more objections than it could in its original form.

3\. A meeting helps every individual understand both the collective aim
of the group and the way in which his own and everyone else’s work can
contribute to the group’s success.

4\. A meeting creates in all present a commitment to the decisions it
makes and the objectives it pursues. Once something has been decided,
even if you originally argued against it, your membership in the group
entails an obligation to accept the decision. The alternative is to
leave the group, but in practice this is very rarely a dilemma of
significance. Real opposition to decisions within organizations usually
consists of one part disagreement with the decision to nine parts
resentment at not being consulted before the decision. For most people
on most issues, it is enough to know that their views were heard and
considered. They may regret that they were not followed, but they accept
the outcome.

And just as the decision of any team is binding on all the members, so
the decisions of a meeting of people higher up in an organization carry
a greater authority than any decision by a single executive. It is much
harder to challenge a decision of the board than of the chief executive
acting on his own. The decision-making authority of a meeting is of
special importance for long-term policies and procedures.

5\. In the world of management, a meeting is very often the only
occasion where the team or group actually exists and works as a group,
and the only time when the supervisor, manager, or executive is actually
perceived as the leader of the team, rather than as the official to whom
individuals report. In some jobs the leader does guide his team through
his personal presence—not just the leader of a pit gang or construction
team, but also the chef in the hotel kitchen and the maitre d’hôtel in
the restaurant, or the supervisor in a department store. But in large
administrative headquarters, the daily or weekly meeting is often the
only time when the leader is ever perceived to be guiding a team rather
than doing a job.

6\. A meeting is a status arena. It is no good to pretend that people
are not or should not be concerned with their status relative to the
other members in a group. It is just another part of human nature that
we have to live with. It is a not insignificant fact that the word order
means (a) hierarchy or pecking order; (b) an instruction or command; and
(c) stability and the way things ought to be, as in “put your affairs in
order,” or “law and order.” All three definitions are aspects of the
same idea, which is indivisible.

Since a meeting is so often the only time when members get the chance to
find out their relative standing, the “arena” function is inevitable.
When a group is new, has a new leader, or is composed of people like
department heads who are in competition for promotion and who do not
work in a single team outside the meeting, “arena behavior” is likely to
figure more largely, even to the point of dominating the proceedings.
However, it will hardly signify with a long-established group that meets
regularly.

Despite the fact that a meeting can perform all of the foregoing main
functions, there is no guarantee that it will do so in any given
situation. It is all too possible that any single meeting may be a waste
of time, an irritant, or a barrier to the achievement of the
organization’s objectives.

## What Sort of Meeting?

While my purpose in this article is to show the critical points at which
most meetings go wrong, and to indicate ways of putting them right, I
must first draw some important distinctions in the size and type of
meetings that we are dealing with.

Meetings can be graded by size into three broad categories: (1) the
assembly—100 or more people who are expected to do little more than
listen to the main speaker or speakers; (2) the council—40 or 50 people
who are basically there to listen to the main speaker or speakers but
who can come in with questions or comments and who may be asked to
contribute something on their own account; and (3) the committee—up to
10 (or at the most 12) people, all of whom more or less speak on an
equal footing under the guidance and control of a chairman.

We are concerned in this article only with the “committee” meeting
though it may be described as a committee, a subcommittee, a study
group, a project team, a working party, a board, or by any of dozens of
other titles. It is by far the most common meeting all over the world,
and can perhaps be traced back to the primitive hunting band through
which our species evolved. Beyond doubt it constitutes the bulk of the
11 million meetings that—so it has been calculated—take place every day
in the United States.

Apart from the distinction of size, there are certain considerations
regarding the type of meeting that profoundly affect its nature. For
instance:

Frequency—A daily meeting is different from a weekly one, and a weekly
meeting from a monthly one. Irregular, ad hoc, quarterly, and annual
meetings are different again. On the whole, the frequency of meetings
defines—or perhaps even determines—the degree of unity of the group.

Composition—Do the members work together on the same project, such as
the nursing and ancillary staff on the same ward of a hospital? Do they
work on different but parallel tasks, like a meeting of the company’s
plant managers or regional sales managers? Or are they a diverse
group—strangers to each other, perhaps—united only by the meeting
itself and by a common interest in realizing its objectives?

Motivation—Do the members have a common objective in their work, like a
football team? Or do they to some extent have a competitive working
relationship, like managers of subsidiary companies at a meeting with
the chief executive, or the heads of research, production, and marketing
discussing finance allocation for the coming year? Or does the desire
for success through the meeting itself unify them, like a neighborhood
action group or a new product design committee?

Decision process—How does the meeting group ultimately reach its
decisions? By a general consensus, “the feeling of the meeting”? By a
majority vote? Or are the decisions left entirely to the chairman
himself, after he has listened to the facts, opinions, and discussions?

### Kinds of meetings

The experienced meeting-goer will recognize that, although there seem to
be five quite different methods of analyzing a meeting, in practice
there is a tendency for certain kinds of meetings to sort themselves out
into one of three categories. Consider:

The daily meeting, where people work together on the same project with a
common objective and reach decisions informally by general agreement.

The weekly or monthly meeting, where members work on different but
parallel projects and where there is a certain competitive element and a
greater likelihood that the chairman will make the final decision
himself.

The irregular, occasional, or “special project” meeting, composed of
people whose normal work does not bring them into contact and whose work
has little or no relationship to the others’. They are united only by
the project the meeting exists to promote and motivated by the desire
that the project should succeed. Though actual voting is uncommon, every
member effectively has a veto.

Of these three kinds of meetings, it is the first—the workface type—that
is probably the most common. It is also, oddly enough, the one most
likely to be successful. Operational imperatives usually ensure that it
is brief, and the participants’ experience of working side by side
ensures that communication is good.

The other two types are a different matter. In these meetings all sorts
of human crosscurrents can sweep the discussion off course, and errors
of psychology and technique on the chairman’s part can defeat its
purposes. Moreover, these meetings are likely to bring together the more
senior people and to produce decisions that profoundly affect the
efficiency, prosperity, and even survival of the whole organization. It
is, therefore, toward these higher-level meetings that the lessons of
this article are primarily directed.

## Before the Meeting

The most important question you should ask is: “What is this meeting
intended to achieve?” You can ask it in different ways—“What would be
the likely consequences of not holding it?” “When it is over, how shall
I judge whether it was a success or a failure?”—but unless you have a
very clear requirement from the meeting, there is a grave danger that it
will be a waste of everyone’s time.

### Defining the objective

You have already looked at the six main functions that all meetings
perform, but if you are trying to use a meeting to achieve definite
objectives, there are in practice only certain types of objectives it
can really achieve. Every item on the agenda can be placed in one of the
following four categories, or divided up into sections that fall into
one or more of them.

#### 1\. Informative-digestive

Obviously, it is a waste of time for the meeting to give out purely
factual information that would be better circulated in a document. But
if the information should be heard from a particular person, or if it
needs some clarification and comment to make sense of it, or if it has
deep implications for the members of the meeting, then it is perfectly
proper to introduce an item onto the agenda that requires no conclusion,
decision, or action from the meeting, it is enough, simply, that the
meeting should receive and discuss a report.

The “informative-digestive” function includes progress reports—to keep
the group up to date on the current status of projects it is responsible
for or that affect its deliberations—and review of completed projects in
order to come to a collective judgment and to see what can be learned
from them for the next time.

#### 2\. Constructive-originative

This “What shall we do?” function embraces all items that require
something new to be devised, such as a new policy, a new strategy, a new
sales target, a new product, a new marketing plan, a new procedure, and
so forth. This sort of discussion asks people to contribute their
knowledge, experience, judgment, and ideas. Obviously, the plan will
probably be inadequate unless all relevant parties are present and
pitching in.

#### 3\. Executive responsibilities

This is the “How shall we do it?” function, which comes after it has
been decided what the members are going to do; at this point, executive
responsibilities for the different components of the task have to be
distributed around the table. Whereas in the second function the
contributors’ importance is their knowledge and ideas, here their
contribution is the responsibility for implementing the plan. The fact
that they and their subordinates are affected by it makes their
contribution especially significant.

It is of course possible to allocate these executive responsibilities
without a meeting, by separate individual briefings, but several
considerations often make a meeting desirable.

First, it enables the members as a group to find the best way of
achieving the objectives.

Second, it enables each member to understand and influence the way in
which his own job fits in with the jobs of the others and with the
collective task.

Third, if the meeting is discussing the implementation of a decision
taken at a higher level, securing the group’s consent may be of prime
importance. If so, the fact that the group has the opportunity to
formulate the detailed action plan itself may be the decisive factor in
securing its agreement, because in that case the final decision belongs,
as it were, to the group. Everyone is committed to what the group
decides and is collectively responsible for the final shape of the
project, as well as individually answerable for his own part in it.
Ideally, this sort of agenda item starts with a policy, and ends with an
action plan.

#### 4\. Legislative framework:

Above and around all considerations of “What to do” and “How to do it,”
there is a framework—a departmental or divisional organization—and a
system of rules, routines, and procedures within and through which all
the activity takes place. Changing this framework and introducing a new
organization or new procedures can be deeply disturbing to committee
members and a threat to their status and long-term security. Yet leaving
it unchanged can stop the organization from adapting to a changing
world. At whatever level this change happens, it must have the support
of all the perceived leaders whose groups are affected by it.

The key leaders for this legislative function must collectively make or
confirm the decision; if there is any important dissent, it is very
dangerous to close the discussion and make the decision by decree. The
group leaders cannot expect quick decisions if they are seeking to
change the organization framework and routines that people have grown up
with. Thus they must be prepared to leave these items unresolved for
further discussion and consultation. As Francis Bacon put it—and it has
never been put better—“Counsels to which time hath not been called, time
will not ratify.”

### Making preparations

The four different functions just discussed may of course be performed
by a single meeting, as the group proceeds through the agenda.
Consequently, it may be a useful exercise for the chairman to go through
the agenda, writing beside each item which function it is intended to
fulfill. This exercise helps clarify what is expected from the
discussion and helps focus on which people to bring in and what
questions to ask them.

### People

The value and success of a committe meeting are seriously threatened if
too many people are present. Between 4 and 7 is generally ideal, 10 is
tolerable, and 12 is the outside limit. So the chairman should do
everything he can to keep numbers down, consistent with the need to
invite everyone with an important contribution to make.

The leader may have to leave out people who expect to come or who have
always come. For this job he may need tact; but since people generally
preserve a fiction that they are overworked already and dislike serving
on committees, it is not usually hard to secure their consent to stay
away.

If the leader sees no way of getting the meeting down to a manageable
size, he can try the following devices: (a) analyze the agenda to see
whether everyone has to be present for every item (he may be able to
structure the agenda so that some people can leave at half time and
others can arrive); (b) ask himself whether he doesn’t really need two
separate, smaller meetings rather than one big one; and (c) determine
whether one or two groups can be asked to thrash some of the topics out
in advance so that only one of them needs to come in with its proposals.

Remember, too, that a few words with a member on the day before a
meeting can increase the value of the meeting itself, either by ensuring
that an important point is raised that comes better from the floor than
from the chair or by preventing a time-wasting discussion of a subject
that need not be touched on at all.

### Papers

The agenda is by far the most important piece of paper. Properly drawn
up, it has a power of speeding and clarifying a meeting that very few
people understand or harness. The main fault is to make it unnecessarily
brief and vague. For example, the phrase “development budget” tells
nobody very much, whereas the longer explanation “To discuss the
proposal for reduction of the 1976–1977 development budget now that the
introduction of our new product has been postponed” helps all committee
members to form some views or even just to look up facts and figures in
advance.

Thus the leader should not be afraid of a long agenda, provided that the
length is the result of his analyzing and defining each item more
closely, rather than of his adding more items than the meeting can
reasonably consider in the time allowed. He should try to include, very
briefly, some indication of the reason for each topic to be discussed.
If one item is of special interest to the group, it is often a good idea
to single it out for special mention in a covering note.

The leader should also bear in mind the useful device of heading each
item “For information,” “For discussion,” or “For decision” so that
those at the meeting know where they are trying to get to.

And finally, the chairman should not circulate the agenda too far in
advance, since the less organized members will forget it or lose it. Two
or three days is about right—unless the supporting papers are
voluminous.

#### Other ‘paper’ considerations:

The order of items on the agenda is important. Some aspects are
obvious—the items that need urgent decision have to come before those
that can wait till next time. Equally, the leader does not discuss the
budget for the re-equipment program before discussing whether to put the
re-equipment off until next year. But some aspects are not so obvious.
Consider:

  - The early part of a meeting tends to be more lively and creative
    than the end of it, so if an item needs mental energy, bright ideas,
    and clear heads, it may be better to put it high up on the list.
    Equally, if there is one item of great interest and concern to
    everyone, it may be a good idea to hold it back for a while and get
    some other useful work done first. Then the star item can be
    introduced to carry the meeting over the attention lag that sets in
    after the first 15 to 20 minutes of the meeting.

<!-- end list -->

  - Some items unite the meeting in a common front while others divide
    the member one from another. The leader may want to start with unity
    before entering into division, or he may prefer the other way
    around. The point is to be aware of the choice and to make it
    consciously, because it is apt to make a difference to the whole
    atmosphere of the meeting. It is almost always a good idea to find a
    unifying item with which to end the meeting.

<!-- end list -->

  - A common fault is to dwell too long on trivial but urgent items, to
    the exclusion of subjects of fundamental importance whose
    significance is long-term rather than immediate. This can be
    remedied by putting on the agenda the time at which discussion of
    the important long-term issue will begin—and by sticking to it.

<!-- end list -->

  - Very few business meetings achieve anything of value after two
    hours, and an hour and a half is enough time to allocate for most
    purposes.

<!-- end list -->

  - It is often a good idea to put the finishing time of a meeting on
    the agenda as well as the starting time.

<!-- end list -->

  - If meetings have a tendency to go on too long, the chairman should
    arrange to start them one hour before lunch or one hour before the
    end of work. Generally, items that ought to be kept brief can be
    introduced ten minutes from a fixed end point.

<!-- end list -->

  - The practice of circulating background or proposal papers along with
    the minutes is, in principle, a good one. It not only saves time,
    but it also helps in formulating useful questions and considerations
    in advance. But the whole idea is sabotaged once the papers get too
    long; they should be brief or provide a short summary. If they are
    circulated, obviously the chairman has to read them, or at least
    must not be caught not having read them. (One chairman, more noted
    for his cunning than his conscientiousness, is said to have spent 30
    seconds before each meeting going through all the papers he had not
    read with a thick red pen, marking lines and question marks in the
    margins at random, and making sure these were accidentally made
    visible to the meeting while the subject was being discussed.)

<!-- end list -->

  - If papers are produced at the meeting for discussion, they should
    obviously be brief and simple, since everyone has to read them. It
    is a supreme folly to bring a group of people together to read six
    pages of closely printed sheets to themselves. The exception is
    certain kinds of financial and statistical papers whose function is
    to support and illustrate verbal points as reference documents
    rather than to be swallowed whole: these are often better tabled at
    the meeting.

<!-- end list -->

  - All items should be thought of and thought about in advance if they
    are to be usefully discussed. Listing “Any other business” on the
    agenda is an invitation to waste time. This does not absolutely
    preclude the chairman’s announcing an extra agenda item at a meeting
    if something really urgent and unforeseen crops up or is suggested
    to him by a member, provided it is fairly simple and
    straightforward. Nor does it preclude his leaving time for general
    unstructured discussion after the close of the meeting.

<!-- end list -->

  - The chairman, in going through the agenda items in advance, can
    usefully insert his own brief notes of points he wants to be sure
    are not omitted from the discussion. A brief marginal scribble of
    “How much notice?” or “Standby arrangements?” or whatever is all
    that is necessary.

## The Chairman’s Job

Let’s say that you have just been appointed chairman of the committee.
You tell everyone that it is a bore or a chore. You also tell them that
you have been appointed “for my sins.” But the point is that you tell
them. There is no getting away from it: some sort of honor or glory
attaches to the chairman’s role. Almost everyone is in some way pleased
and proud to be made chairman of something. And that is three quarters
of the trouble.

### Master or servant?

Their appointment as committee chairman takes people in different ways.
Some seize the opportunity to impose their will on a group that they see
themselves licensed to dominate. Their chairmanship is a harangue,
interspersed with demands for group agreement.

Others are more like scoutmasters, for whom the collective activity of
the group is satisfaction enough, with no need for achievement. Their
chairmanship is more like the endless stoking and fueling or a campfire
that is not cooking anything.

And there are the insecure or lazy chairmen who look to the meeting for
reassurance and support in their ineffectiveness and inactivity, so that
they can spread the responsibility for their indecisiveness among the
whole group. They seize on every expression of disagreement or doubt as
a justification for avoiding decision or action.

But even the large majority who do not go to those extremes still feel a
certain pleasurable tumescence of the ego when they take their place at
the head of the table for the first time. The feeling is no sin: the sin
is to indulge it or to assume that the pleasure is shared by the other
members of the meeting.

It is the chairman’s self-indulgence that is the greatest single barrier
to the success of a meeting. His first duty, then, is to be aware of the
temptation and of the dangers of yielding to it. The clearest of the
danger signals is hearing himself talking a lot during a discussion.

One of the best chairmen I have ever served under makes it a rule to
restrict her interventions to a single sentence, or at most two. She
forbids herself ever to contribute a paragraph to a meeting she is
chairing. It is a harsh rule, but you would be hard put to find a
regular attender of her meetings (or anyone else’s) who thought it was a
bad one.

There is, in fact, only one legitimate source of pleasure in
chairmanship, and that is pleasure in the achievements of the
meeting—and to be legitimate it must be shared by all those present.
Meetings are necessary for all sorts of basic and primitive human
reasons, but they are useful only if they are seen by all present to be
getting somewhere—and somewhere they know they could not have gotten to
individually.

If the chairman is to make sure that the meeting achieves valuable
objectives, he will be more effective seeing himself as the servant of
the group rather than as its master. His role then becomes that of
assisting the group toward the best conclusion or decision in the most
efficient manner possible: to interpret and clarify; to move the
discussion forward; and to bring it to a resolution that everyone
understands and accepts as being the will of the meeting, even if the
individuals do not necessarily agree with it.

His true source of authority with the members is the strength of his
perceived commitment to their combined objective and his skill and
efficiency in helping and guiding them to its achievement. Control and
discipline then become not the act of imposing his will on the group but
of imposing the group’s will on any individual who is in danger of
diverting or delaying the progress of the discussion and so from
realizing the objective.

Once the members realize that the leader is impelled by his commitment
to their common objective, it does not take great force of personality
for him to control the meeting. Indeed, a sense of urgency and a clear
desire to reach the best conclusion as quickly as possible are a much
more effective disciplinary instrument than a big gavel. The effective
chairman can then hold the discussion to the point by indicating that
there is no time to pursue a particular idea now, that there is no time
for long speeches, that the group has to get through this item and on to
the next one, rather than by resorting to pulling rank.

There are many polite ways the chairman can indicate a slight impatience
even when someone else is speaking—by leaning forward, fixing his eyes
on the speaker tensing his muscles, raising his eyebrows, or nodding
briefly to show the point is taken. And when replying or commenting, the
chairman can indicate by the speed, brevity, and finality of his
intonation that “we have to move on.” Conversely, he can reward the sort
of contribution he is seeking by the opposite expressions and
intonations, showing that there is plenty of time for that sort of idea,
and encouraging the speaker to develop the point.

After a few meetings, all present readily understand this nonverbal
language of chairmanship. It is the chairman’s chief instrument of
educating the group into the general type of “meeting behavior” that he
is looking for. He is still the servant of the group, but like a hired
mountain guide, he is the one who knows the destination, the route, the
weather signs, and the time the journey will take. So if he suggests
that the members walk a bit faster, they take his advice.

This role of servant rather than master is often obscured in large
organizations by the fact that the chairman is frequently the line
manager of the members: this does not, however, change the reality of
the role of chairman. The point is easier to see in, say, a neighborhood
action group. The question in that case is, simply, “Through which
person’s chairmanship do we collectively have the best chance of getting
the children’s playground built?”

However, one special problem is posed by this definition of the
chairman’s role, and it has an extremely interesting answer. The
question is: How can the chairman combine his role with the role of a
member advocating one side of an argument?

The answer comes from some interesting studies by researchers who sat in
on hundreds of meetings to find out how they work. Their consensus
finding is that most of the effective discussions have, in fact, two
leaders: one they call a “team,” or “social,” leader; the other a
“task,” or “project,” leader.

Regardless of whether leadership is in fact a single or a dual function,
for our purposes it is enough to say that the chairman’s best role is
that of social leader. If he wants a particular point to be strongly
advocated, he ensures that it is someone else who leads off the task
discussion, and he holds back until much later in the argument. He might
indeed change or modify his view through hearing the discussion, but
even if he does not it is much easier for him to show support for
someone else’s point later in the discussion, after listening to the
arguments. Then, he can summarize in favor of the one he prefers.

The task advocate might regularly be the chairman’s second-in-command,
or a different person might advocate for different items on the agenda.
On some subjects, the chairman might well be the task advocate himself,
especially if they do not involve conflict within the group. The
important point is that the chairman has to keep his “social leadership”
even if it means sacrificing his “task leadership.” However, if the
designated task advocate persists in championing a cause through two or
three meetings, he risks building up quite a head of antagonism to him
among the other members. Even so, this antagonism harms the group less
by being directed at the “task leader” than at the “social leader.”

### Structure of discussion

It may seem that there is no right way or wrong way to structure a
committee meeting discussion.

A subject is raised, people say what they think, and finally a decision
is reached, or the discussion is terminated. There is some truth in
this. Moreover, it would be a mistake to try and tie every discussion of
every item down to a single immutable format.

Nevertheless, there is a logical order to a group discussion, and while
there can be reasons for not following it, there is no justification for
not being aware of it. In practice, very few discussions are inhibited,
and many are expedited, by a conscious adherence to the following
stages, which follow exactly the same pattern as a visit to the doctor.

#### “What seems to be the trouble?”

The reason for an item being on a meeting agenda is usually like the
symptom we go to the doctor with: “I keep getting this pain in my back”
is analogous to “Sales have risen in Germany but fallen in France.” In
both cases it is clear that something is wrong and that something ought
to be done to put it right. But until the visit to the doctor, or the
meeting of the European marketing committee, that is about all we really
know.

#### “How long has this been going on?”

The doctor will start with a case history of all the relevant background
facts, and so will the committee discussion. A solid basis of shared and
agreed-on facts is the best foundation to build any decision on, and a
set of pertinent questions will help establish it. For example, when did
French sales start to fall off? Have German sales risen exceptionally?
Has France had delivery problems, or less sales effort, or weaker
advertising? Have we lost market share, or are our competitors’ sales
falling too? If the answers to all these questions, and more, are not
established at the start, a lot of discussion may be wasted later.

#### “Would you just lie down on the couch?”

The doctor will then conduct a physical examination to find out how the
patient is now. The committee, too, will want to know how things stand
at this moment. Is action being taken? Do long-term orders show the same
trend? What are the latest figures? What is the current stock position?
How much money is left in the advertising budget?

#### “You seem to have slipped a disc.”

When the facts are established, you can move toward a diagnosis. A
doctor may seem to do this quickly, but that is the result of experience
and practice. He is, in fact, rapidly eliminating all the impossible or
far-fetched explanations until he leaves himself with a short list. The
committee, too, will hazard and eliminate a variety of diagnoses until
it homes in on the most probable—for example the company’s recent
energetic and highly successful advertising campaign in Germany plus new
packaging by the market leader in France.

#### “Take this round to the druggist.”

Again, the doctor is likely to take a shortcut that a committee meeting
may be wise to avoid. The doctor comes out with a single prescription,
and the committee, too, may agree quickly on a single course of action.

But if the course is not so clear, it is better to take this step in two
stages: (a) construct a series of options—do not, at first, reject any
suggestions outright but try to select and combine the promising
elements from all of them until a number of thought-out, coherent, and
sensible suggestions are on the table; and (b) only when you have
generated these options do you start to choose among them. Then you can
discuss and decide whether to pick the course based on repackaging and
point-of-sale promotion, or the one based on advertising and a price
cut, or the one that bides its time and saves the money for heavier
new-product promotion next year.

If the item is at all complex or especially significant, it is important
for the chairman not only to have the proposed course of the discussion
in his own head, but also to announce it so that everyone knows. A good
idea is to write the headings on an easel pad with a felt pen. This
saves much of the time wasting and confusion that result when people
raise items in the wrong place because they were not privy to the
chairman’s secret that the right place was coming up later on in the
discussion.

## Conducting the Meeting

Just as the driver of a car has two tasks, to follow his route and to
manage his vehicle, so the chairman’s job can be divided into two
corresponding tasks, dealing with the subject and dealing with the
people.

### Dealing with the subject

The essence of this task is to follow the structure of discussion as
just described in the previous section. This, in turn, entails listening
carefully and keeping the meeting pointed toward the objective.

At the start of the discussion of any item, the chairman should make it
clear where the meeting should try to get to by the end. Are the members
hoping to make a clear decision or firm recommendation? Is it a
preliminary deliberation to give the members something to go away with
and think about? Are they looking for a variety of different lines to be
pursued outside the meeting? Do they have to approve the proposal, or
merely note it?

The chairman may give them a choice: “If we can agree on a course of
action, that’s fine. If not, we’ll have to set up a working party to
report and recommend before next month’s meeting.”

The chairman should make sure that all the members understand the issue
and why they are discussing it. Often it will be obvious, or else they
may have been through it before. If not, then he or someone he has
briefed before the meeting should give a short introduction, with some
indication of the reason the item is on the agenda; the story so far;
the present position; what needs to be established, resolved, or
proposed; and some indication of lines of inquiry or courses of action
that have been suggested or explored, as well as arguments on both sides
of the issue.

If the discussion is at all likely to be long or complex, the chairman
should propose to the meeting a structure for it with headings (written
up if necessary), as I stated at the end of the section on “Structure of
discussion.” He should listen carefully in case people jump too far
ahead (e.g., start proposing a course of action before the meeting has
agreed on the cause of the trouble), or go back over old ground, or
start repeating points that have been made earlier. He has to head
discussion off sterile or irrelevant areas very quickly (e.g., the
rights and wrongs of past decisions that it is too late to change, or
distant prospects that are too remote to affect present actions).

It is the chairman’s responsibility to prevent misunderstanding and
confusion. If he does not follow an argument or understand a reference,
he should seek clarification from the speaker. If he thinks two people
are using the same word with different meanings, he should intervene
(e.g., one member using promotion to mean point-of-sale advertising
only, and another also including media publicity).

He may also have to clarify by asking people for facts or experience
that perhaps influence their view but are not known to others in the
meeting. And he should be on the lookout for points where an interim
summary would be helpful. This device frequently takes only a few
seconds, and acts like a life belt to some of the members who are
getting out of their depth.

Sometimes a meeting will have to discuss a draft document. If there are
faults in it, the members should agree on what the faults are and the
chairman should delegate someone to produce a new draft later. The group
should never try to redraft around the table.

Perhaps one of the most common faults of chairmanship is the failure to
terminate the discussion early enough. Sometimes chairmen do not realize
that the meeting has effectively reached an agreement, and consequently
they let the discussion go on for another few minutes, getting nowhere
at all. Even more often, they are not quick enough to close a discussion
before agreement has been reached.

A discussion should be closed once it has become clear that (a) more
facts are required before further progress can be made, (b) discussion
has revealed that the meeting needs the views of people not present, (c)
members need more time to think about the subject and perhaps discuss it
with colleagues, (d) events are changing and likely to alter or clarify
the basis of the decision quite soon, (e) there is not going to be
enough time at this meeting to go over the subject properly, or (f) it
is becoming clear that two or three of the members can settle this
outside the meeting without taking up the time of the rest. The fact
that the decision is difficult, likely to be disputed, or going to be
unwelcome to somebody, however, is not a reason for postponement.

At the end of the discussion of each agenda item, the chairman should
give a brief and clear summary of what has been agreed on. This can act
as the dictation of the actual minutes. It serves not merely to put the
item on record, but also to help people realize that something
worthwhile has been achieved. It also answers the question “Where did
all that get us?” If the summary involves action by a member of the
meeting, he should be asked to confirm his acceptance of the
undertaking.

### Dealing with the people

There is only one way to ensure that a meeting starts on time, and that
is to start it on time. Latecomers who find that the meeting has begun
without them soon learn the lesson. The alternative is that the prompt
and punctual members will soon realize that a meeting never starts until
ten minutes after the advertised time, and they will also learn the
lesson.

Punctuality at future meetings can be wonderfully reinforced by the
practice of listing late arrivals (and early departures) in the minutes.
Its ostensible and perfectly proper purpose is to call the latecomer’s
attention to the fact that he was absent when a decision was reached.
Its side effect, however, is to tell everyone on the circulation list
that he was late, and people do not want that sort of information about
themselves published too frequently.

There is a growing volume of work on the significance of seating
positions and their effect on group behavior and relationships. Not all
the findings are generally agreed on. What does seem true is that:

  - Having members sit face to face across a table facilitates
    opposition, conflict, and disagreement, though of course it does not
    turn allies into enemies. But it does suggest that the chairman
    should think about whom he seats opposite himself.

<!-- end list -->

  - Sitting side by side makes disagreements and confrontation harder.
    This in turn suggests that the chairman can exploit the
    friendship-value of the seats next to him.

<!-- end list -->

  - There is a “dead man’s corner” on the chairman’s right, especially
    if a number of people are seated in line along from him (it does not
    apply if he is alone at the head of the table).

<!-- end list -->

  - As a general rule, proximity to the chairman is a sign of honor and
    favor. This is most marked when he is at the head of a long, narrow
    table. The greater the distance, the lower the rank—just as the
    lower-status positions were “below the salt” at medieval
    refectories.

### Control the garrulous

In most meetings someone takes a long time to say very little. As
chairman, your sense of urgency should help indicate to him the need for
brevity. You can also suggest that if he is going to take a long time it
might be better for him to write a paper. If it is urgent to stop him in
full flight, there is a useful device of picking on a phrase (it really
doesn’t matter what phrase) as he utters it as an excuse for cutting in
and offering it to someone else: “Inevitable decline—that’s very
interesting. George, do you agree that the decline is inevitable?”

### Draw out the silent

In any properly run meeting, as simple arithmetic will show, most of the
people will be silent most of the time. Silence can indicate general
agreement, or no important contribution to make, or the need to wait and
hear more before saying anything or too good a lunch, and none of these
need worry you. But there are two kinds of silence you must break:.

1\. The silence of diffidence. Someone may have a valuable contribution
to make but be sufficiently nervous about its possible reception to keep
it to himself. It is important that when you draw out such a
contribution, you should express interest and pleasure (though not
necessarily agreement) to encourage further contributions of that sort.

2\. The silence of hostility. This is not hostility to ideas, but to you
as the chairman, to the meeting, and to the process by which decisions
are being reached.

This sort of total detachment from the whole proceedings is usually the
symptom of some feeling of affront. If you probe it, you will usually
find that there is something bursting to come out, and that it is better
out than in.

### Protect the weak

Junior members of the meeting may provoke the disagreement of their
seniors, which is perfectly reasonable. But if the disagreement
escalates to the point of suggesting that they have no right to
contribute, the meeting is weakened. So you may have to take pains to
commend their contribution for its usefulness, as a pre-emptive measure.
You can reinforce this action by taking a written note of a point they
make (always a plus for a member of a meeting) and by referring to it
again later in the discussion (a double-plus).

### Encourage the clash of ideas

But, at the same time, discourage the clash of personalities. A good
meeting is not a series of dialogues between individual members and the
chairman. Instead, it is a crossflow of discussion and debate, with the
chairman occasionally guiding, meditating, probing, stimulating, and
summarizing, but mostly letting the others thrash ideas out. However,
the meeting must be a contention of ideas, not people.

If two people are starting to get heated, widen the discussion by asking
a question of a neutral member of the meeting, preferably a question
that requires a purely factual answer.

### Watch out for the suggestion-squashing reflex

Students of meetings have reduced everything that can be said into
questions, answers, positive reactions, and negative reactions.
Questions can only seek, and answers only supply, three types of
responses: information, opinion, and suggestion.

In almost every modern organization, it is the suggestions that contain
the seeds of future success. Although very few suggestions will ever
lead to anything, almost all of them need to be given every chance. The
trouble is that suggestions are much easier to ridicule than facts or
opinions. If people feel that making a suggestion will provoke the
negative reaction of being laughed at or squashed, they will soon stop.
And if there is any status-jostling going on at the meeting, it is all
too easy to use the occasion of someone’s making a suggestion as the
opportunity to take him down a peg. It is all too easy and a formula to
ensure sterile meetings.

The answer is for you to take special notice and show special warmth
when anyone makes a suggestion, and to discourage as sharply as you can
the squashing-reflex. This can often be achieved by requiring the
squasher to produce a better suggestion on the spot. Few suggestions can
stand up to squashing in their pristine state: your reflex must be to
pick out the best part of one and get the other committee members to
help build it into something that might work.

### Come to the most senior people last

Obviously, this cannot be a rule, but once someone of high authority has
pronounced on a topic, the less senior members are likely to be
inhibited. If you work up the pecking order instead of down it, you are
apt to get a wider spread of views and ideas. But the juniors who start
it off should only be asked for contributions within their personal
experience and competence (“Peter, you were at the Frankfurt
Exhibition—what reactions did you pick up there?”).

### Close on a note of achievement

Even if the final item is left unresolved, you can refer to an earlier
item that was well resolved as you close the meeting and thank the
group.

If the meeting is not a regular one, fix the time and place of the next
one before dispersing. A little time spent with appointment diaries at
the end, especially if it is a gathering of five or more members, can
save hours of secretarial telephoning later.

### Following the meeting

Your secretary may take the minutes (or better still, one of the
members), but the minutes are your responsibility. They can be very
brief, but they should include these facts:

  - The time and date of the meeting, where it was held, and who chaired
    it.

<!-- end list -->

  - Names of all present and apologies for absence.

<!-- end list -->

  - All agenda items (and other items) discussed and all decisions
    reached. If action was agreed on, record (and underline) the name of
    the person responsible for the assignment.

<!-- end list -->

  - The time at which the meeting ended (important, because it may be
    significant later to know whether the discussion lasted 15 minutes
    or 6 hours).

<!-- end list -->

  - The date, time, and place of the next committee meeting.

A version of this article appeared in the [March
1976](/archive-toc/3762) issue of Harvard Business Review.