Making Dumb Groups Smarter – From HBR December 2014

While groups are an important part of working life, they don’t always work effectively for a number of reasons. The success or otherwise of a group depends on the ability to effectively share all information held by the members of the group so that the group can work together in a cohesive manner. This however, is not always the reality and when groups fail, they usually do so as a result of one of the following:

  1. Amplification of errors
  2. Cascade effects
  3. Polarising groups
  4. Focusing on what everybody knows


  1. Amplifying Errors:

This happens in a number of ways, including:

The planning fallacy – underestimating the costs and time required to complete a project.

Over confidence leads us to believe that our predictions are more accurate than they actually are and these errors have both been found to be aggravated in groups – they tend to be even more optimistic than individuals.

The representative heuristic, which leads us to think that things which are similar in one way are similar in other ways has been found to be increased in groups compared to individuals.

The sunk-cost fallacy occurs when people stick with a project which is failing because of the time and resources already invested in it. It has been shown that groups have also been found to be even more likely than individuals to be increasingly committed to a plan that is failing, especially if they identify strongly with the group.

Framing effects, which refers to how we view things, depending on the way in which they are presented to us.

  1. Cascade Effects

This refers to a trickle in one direction that turns into a cascade and it can happen when people get the impression from the outset, possibly from the group leader, that the general consensus is going in a certain direction and others, individually concur, for a number of reasons, which include not wanting to appear ignorant or adversarial.  As each person concurs, it creates a cascade effect whereby others joining the group or decision making process go along with what they think is the general consensus. Individuals then self-censor themselves and don’t voice opposing arguments. This can result in the group making the wrong decision as opposing views and information are not given airtime.

  1. Polarising Groups

This happens when a person who is already inclined to take risks, consults with another and becomes more inclined to take the risk. Group deliberation in this instance appears to cause a “risky shift”. The opposite has also been found, whereby when a person who is more cautious in terms of risk taking consults with another, they become even less risk inclined.


  1. Focusing on “What Everybody Knows”

This happens when most of the group have most of the information but some members have additional information that is either not shared, or not focused on. The “common knowledge effect” occurs when information held by all carries more weight than information held by only a few members of the group. Common information has been found to have a disproportionality large impact on discussions and conclusions, while information that is held by only a few, is given disproportionality little weight, and this can lead to poor decisions.



Making Groups Wiser:


Group failures can have disastrous effects, however there are some practical safeguards to the above pitfalls. These include:

Silencing the leader – by expressing their own views, a leader can unwittingly promote self-censorship. Others may not want to be seen to disagree with the leader and as a result will not want to give a conflicting view or opinion. To combat this, the leader should indicate their desire to hear independent information from the outset and refuse to take a firm view until all the information has emerged.

Priming critical thinking – Research has found that when people are given a “getting along” task to do before engaging in a group deliberation, they are much more likely to stay silent in the face of opposing views than those who were given a critical thinking task to perform. If the leader of the group encourages information disclosure from the beginning people are more likely to speak up.

Rewarding group success – cascades are less likely to happen if people know that they have nothing to gain by a correct individual decision and everything to gain from a correct group decision. Identifying with the groups success will encourage people to speak up and against the general consensus with information that could be critical to the outcome.

Assigning roles – the pitfall of focusing on “what everyone knows” is less likely to happen if every member of the group is assigned a specific role – before deliberations begin. If the group members recognise that each person has something potentially different to bring to the group from the outset, they will be more open to hearing it and giving it the appropriate consideration.

Establishing contrarian teams – doing this involves appointing a “red team” whose mission it is to defeat the primary team by exposing mistakes and potential vulnerabilities and are given clear incentives to do so.

The Delphi method – this mixes the pros of individual decision making with social learning. Individuals offer first round votes anonymously. Then they vote again, with a requirement that the second round estimates have to fall within the middle quartiles of the first round. The process is repeated until they converge on an estimate. The process is often interspersed with group discussion. The anonymity of the process offers protection to the individual group members and therefore reduces the problem of self-silencing.


Therefore, in conclusion, it appears that while there are numerous pitfalls that groups can unwittingly fall prey to, by being aware of these, we can take preventative steps to avoid them and reap more of the rewards that are traditionally associated with group work.