skip to navigation skip to content

The lowdown

 

Sequential search

Two heads are not better than one: a new study co-authored by Dr Vincent Mak of Cambridge Judge Business School questions whether committees make better decisions than individuals.

Hundreds of voting tokens.
Dr Vincent Mak
Dr Vincent Mak

There once was a time when committee decisions were invariably taken in a physical group setting, usually a meeting room. Committee members pored over documents and engaged in lengthy discussions before finally reaching a decision – over competing suppliers or alternative job candidates the panel had just interviewed, for example.

But in the Internet age, time-starved committees often need to make snap decisions on each alternative as it comes along, and members might only be able to email a one-word “yes” or “no” to indicate what they think – a practice that can be called “sequential search by committees”. Decisions taken in this manner can have multi-million-dollar ramifications, as they might concern selecting big investment projects or renting office space.

Conventional wisdom would assume that even decisions taken in this modern manner by like-minded committee members would be sounder than a decision taken by just one individual member – on the basis that two heads are better than one.

Yet a study co-authored by Dr Vincent Mak, Reader in Marketing & Decision Sciences of Cambridge Judge Business School, questions this assumption. The study published in the journal Management Science found that committees made smarter decisions than individuals only when committee decisions required unanimity – and even then, there was no significant improvement over decisions by individuals.

“We think the results are driven by the fact that, at an individual level, each committee member tends to rely on each other to make smart decisions – so everybody ends up cutting corners with the mental effort they put into the decision. On the other hand, a single individual, who has only himself or herself to rely on, cannot afford to cut corners,” says Vincent. “Our findings caution against simplistic assumptions about the quality of committee decision making over individual decision making.”

The study is based on an experiment involving 126 people, who were placed in three-person committees and divided into groupings according to each of three voting rules: unanimity (requiring all three committee members), majority (two members) and minority (one member).

Each committee played an online “sequential search game” in which they were presented one fictional alternative after another, each time voting whether to accept the current alternative. If the voting rule was, for example, majority, the committee accepted the alternative if at least two members voted to accept it – and the game stopped there and then. If there were not enough votes to accept the alternative, the committee moved on to the next alternative and would not be able to go back. When the committee was presented an alternative, they could only compare it with all other alternatives they had already seen, to obtain what the researchers call a “relative rank”. However, each committee member’s payment from the game was higher based on how much better the accepted alternative actually was compared with all the alternatives seen or unseen – an “absolute rank” that would only be revealed to them after they had made their decision.

“While sequential search is now common in every aspect of life, including families deciding on a holiday destination, the study’s findings are of special significance to senior executives who need to make important decisions,” says Mak, who has authored a series of studies on online search.

“In the case of talent searches, if senior executives evaluate and interview candidates sequentially this can take quite a bit of time and may lead to losing excellent prospects to companies that can move faster.

“So the study’s finding that committees don’t usually make significantly better decisions than individuals – and may often make worse decisions – can hold important practical implications for companies. Requiring unanimity in such situations might be best in mitigating the shortcomings of committee decisions.”

The study – entitled “Voting Rules in Sequential Search by Committees: Theory and Experiments” – is co-authored by Vincent Mak of Cambridge Judge Business School, Darryl A. Seale of Lee Business School at the University of Nevada, and by Amnon Rapoport and Eyran J. Gisches of Eller College of Management at the University of Arizona.

Dr Vincent Mak discusses his research into sequential search.
Dr Vincent Mak explains what conclusions the study reached.