Tag Archives: executive decisions

Groupthink in Action

An interesting situation which, I would argue, is representative of Groupthink.

As a brief summary of the situation: a subgroup within a larger group is discussing the possibility of changing the larger group’s structure. In that larger group, similar discussions have been quite frequent, in the past. In effect, the smaller group is moving toward enacting a decision based on perceived consensus as to “the way to go.”

No bad intention on anyone’s part and the situation is far from tragic. But my clear impression is that groupthink is involved. I belong to the larger group but I feel little vested interest in what might happen with it.

An important point about this situation is that the smaller group seems to be acting as if the decision had already been made, after careful consideration. Through the history of the larger group, prior discussions on the same topic have been frequent. Through these discussions, clear consensus has never been reached. At the same time, some options have been gaining some momentum in the recent past, mostly based (in my observation) on accumulated frustration with the status quo and some reflection on the effectiveness of activities done by subgroups within the larger group. Members of that larger group (including participants in the smaller group) are quite weary of rehashing the same issues and the “rallying cry” within the subgroup has to do with “moving on.” Within the smaller group, prior discussions are described as if they had been enough to explore all the options. Weariness through the group as a whole seems to create a sense of urgency even though the group as a whole could hardly be described as being involved in time-critical activities.

Nothing personal about anyone involved and it’s possible that I’m off on this one. Where some of those involved would probably disagree is in terms of the current stage in the decision making process (i.e., they may see themselves as having gone through the process of making the primary decision, the rest is a matter of detail). I actually feel strange talking about this situation because it may seem like I’m doing the group a disservice. The reason I think it isn’t the case is that I have already voiced my concerns about groupthink to those who are involved in the smaller group. The reason I feel the urge to blog about this situation is that, as a social scientist, I take it as my duty to look at issues such as group dynamics. Simply put, I started thinking about it as a kind of “case study.”

Yes, I’m a social science geek. And proud of it, too!

Thing is, I have a hard time not noticing a rather clear groupthink pattern. Especially when I think about a few points in Janis‘s description of groupthink.


Antecedent Conditions Symptoms Decisions Affected


Insulation of the group Illusion of invulnerability Incomplete survey of alternatives


High group cohesiveness Unquestioned belief in the inherent morality of the group Incomplete survey of objectives


Directive leadership Collective rationalization of group’s decisions Failure to examine risks of preferred choice


Lack of norms requiring methodical procedures Shared stereotypes of outgroup, particularly opponents Failure to re-appraise initially rejected alternatives


Homogeneity of members’ social background and ideology Self-censorship; members withhold criticisms Poor information search


High stress from external threats with low hope of a better solution than the one offered by the leader(s) Illusion of unanimity (see false consensus effect) Selective bias in processing information at hand (see also confirmation bias)


Direct pressure on dissenters to conform Failure to work out contingency plans


Self-appointed “mindguards” protect the group from negative information


A PDF version, with some key issues highlighted.

Point by point…


Antecedent Conditions of Groupthink

Insulation of the group

A small subgroup was created based on (relatively informal) prior expression of opinion in favour of some broad changes in the structure of the larger group.

Lack of norms requiring methodical procedures

Methodical procedures about assessing the situation are either put aside or explicitly rejected.
Those methodical procedures which are accepted have to do with implementing the group’s primary decision, not with the decision making process.

Symptoms Indicative of Groupthink

Illusion of unanimity (see false consensus effect)

Agreement is stated as a fact, possibly based on private conversations outside of the small group.

Direct pressure on dissenters to conform

A call to look at alternatives is constructed as a dissenting voice.
Pressure to conform is couched in terms of “moving on.”

Symptoms of Decisions Affected by Groupthink

Incomplete survey of alternatives

Apart from the status quo, no alternative has been discussed.
When one alternative model is proposed, it’s reduced to a “side” in opposition to the assessed consensus.

Incomplete survey of objectives

Broad objectives are assumed to be common, left undiscussed.
Discussion of objectives is pushed back as being irrelevant at this stage.

Failure to examine risks of preferred choice

Comments about possible risks (including the danger of affecting the dynamics of the existing broader group) are left undiscussed or dismissed as “par for the course.”

Failure to re-appraise initially rejected alternatives

Any alternative is conceived as having been tried in the past with the strong implication that it isn’t wort revisiting.

Poor information search

Information collected concerns ways to make sure that the primary option considered will work.

Failure to work out contingency plans

Comments about the possible failure of the plan, and effects on the wider group are met with “so be it.”

Less Obvious

Antecedent Conditions of Groupthink

High group cohesiveness

The smaller group is highly cohesive but so is the broader group.

Directive leadership

Several members of the smaller group are taking positions of leadership, but there’s no direct coercion from that leadership.

Positions of authority are assessed, in a subtle way, but this authority is somewhat indirect.

Homogeneity of members’ social background and ideology

As with cohesiveness, homogeneity of social background can be used to describe the broader group as well as the smaller one.

High stress from external threats with low hope of a better solution than the one offered by the leader(s)

External “threats” are mostly subtle but there’s a clear notion that the primary option considered may be met with some opposition by a proportion of the larger group.

Symptoms Indicative of Groupthink

Illusion of invulnerability

While “invulnerability” would be an exaggeration, there’s a clear sense that members of the smaller group have a strong position within the larger group.

Unquestioned belief in the inherent morality of the group

Discussions don’t necessarily have a moral undertone, but the smaller group’s goals seem self-evident in the context or, at least, not really worth careful discussion.

Collective rationalization of group’s decisions

Since attempts to discuss the group’s assumed consensus are labelled as coming from a dissenting voice, the group’s primary decision is reified through countering individual points made about this decision.

Shared stereotypes of outgroup, particularly opponents

The smaller group’s primary “outgroup” is in fact the broader group, described in rather simple terms, not a distinct group of people.
The assumption is that, within the larger group, positions about the core issue are already set.

Self-censorship; members withhold criticisms

Self-censorship is particularly hard to observe or assess but the group’s dynamics tends to construct criticism as “nitpicking,” making it difficult to share comments.

Self-appointed “mindguards” protect the group from negative information

As with leadership, the process of shielding the smaller group from negative information is mostly organic, not located in a single individual.
Because the smaller group is already set apart from the larger group, protection from external information is built into the system, to an extent.

Symptoms of Decisions Affected by Groupthink

Selective bias in processing information at hand (see also confirmation bias)

Information brought into the discussion is treated as either reinforcing the group’s alleged consensus or taken to be easy to counter.
Examples from cases showing clear similarities are dismissed (“we have no interest in knowing what others have done”) and distant cases are used to demonstrate that the approach is sound (“there are groups in other contexts which work, so we can use the same approach”).

Customer Service on the Phone: Netflix

An interesting piece about the move, by Netflix, to phone-only customer service.
Victory for voices over keystrokes | CNET News.com

Much of it sounds very obvious. Customers tend to prefer phone support instead of email. Customer service representatives who take more time on the phone with customers are more likely to make people happy. Many customers dislike offshoring. Customer service can make or break some corporations. Customers often have outlandish requests. Hourly salaries in call centres will vary greatly from one place to the other, even within the same area.

In other words, Netflix has done what many people think a company should do. We’ll see how it all pans out in the end.

The main reason this piece caught my attention is that I have been doing surveys (over the phone) about the quality of the service provided by customer service representatives over the phone. Not only am I working in a call centre myself (and can certainly relate with the job satisfaction which comes from empathy). But several of the surveys I do are precisely about the points made in this News.com piece. The majority of the surveys I do are about the quality of the service provided by customer service representatives (CSRs) at incoming call centres for a big corporation. So I hear a lot about CSRs and what they do well. Or not so well. One answer I’ve been hearing on occasion was “I’d appreciate it if I could talk to people who are a bit less courteous but who know more about the services the company is providing.” After interactions with several CSRs and tech support people, I can relate with this experience on a personal level.

The general pattern is that people do prefer it if they can speak directly (over the phone) with a human being who speaks their native language very fluently and are able to spend as much time as it takes with them on the phone. Most people seem to believe that it is important to be able to speak to someone instead of dealing with the issue in an “impersonal” manner.

Sounds obvious. And it probably is obvious to many executives, when they talk about customer service. So email support, outsourcing, offshoring, time limits on customer service, and low wages given to customer service representatives are all perceived by customers as cost-cutting measures.

But there’s something else.

We need the “chunky spaghetti sauce” of customer service. Yes, this is also very obvious. But it seems that some people draw awkward conclusions from it. It’s not really about niche marketing. It’s not exactly about customer choice or even freedom. It’s about diversity.

As an anthropologist, I cherish human diversity. Think of the need for biological diversity on the level of species but through the cultural, linguistic, and biological dimensions of one subspecies (Homo sapiens sapiens).
Yes, we’re all the same. Yes, we’re all different. But looking at human diversity for a while, you begin to notice patterns. Some of these patterns can be described as “profiles.” Other patterns are more subtle, harder to describe. But really not that difficult to understand.

The relationships between age and technology use, for instance. The common idea is that the younger you are, the more likely you are to be “into technology.” “It’s a generation thing, you know. Kids these days, they’re into HyPods and MikeSpaces, and Nit’n’do-wee. I’m too old to know anything about these things.”

Yeah, right.

All the while, some children are struggling with different pieces of technology forced unto them and some retirees are sending each other elaborate PowerPoint files to younger people who are too busy to look at them.

To go back to customer service on the phone. Some people are quite vocal about their preference for interactions with “real human beings” who speak their native language and are able to understand them. Other people would actually prefer it if they could just fire off a message somewhere and not have to spend any time on the phone. On several occasions having to do with customer service, I do prefer email exchanges over phone interactions. But I realize that I’m probably in the minority.

Many people in fact deal with different situations in different ways.

One paragraph I personally find quite surprising in the News.com piece is about the decision to not only strengthen the phone-based support but to, in effect, abolish email support:

Netflix’s decision to eliminate the e-mail feature was made after a great deal of research, Osier said. He looked at two other companies with reputations for superb phone-based customer service, Southwest Airlines and American Express, and saw that customers preferred human interaction over e-mail messages.

Sounds like a knee-jerk reaction to me. (It’d be fun to read the research report!) I’m pretty sure that most business schools advise future executives against knee-jerk reactions.

One thing which surprises me about the Netflix move is that, contrary to Southwest Airlines and American Express, the Netflix business is primarily based on online communication and postal services. My hunch is that a significant number of Netflix users are people who enjoy the convenience of one-click movie rentals without any need to interact with a person. Not that Netflix users dislike other human beings but they may prefer dealing with other human beings on other levels. If my hunch is accurate to any degree, chances are that these same people also enjoy it when they can solve an issue with their account through a single email or, better yet, a single click. For instance, someone might like the option of simply clicking a button on the Netflix website to put their rental queue on hold. And it might be quite useful to receive an email confirmation of a “Damaged Disc Report” (SRC: DISCPROBLEM) instead of having to rely on a confirmation number given on the phone by a friendly CSR in Oregon or, say, Moncton.

Yes, I’m referring to the specific instances of my interactions with Netflix. While I’d certainly appreciate the opportunity to speak with friendly French-speaking CSRs when I have problems with plane tickets or credit cards, I like the fact that I can deal with Netflix online (and through free postal mail). Call me crazy all you want. I’m one of those Netflix customers who find it convenient to deal with the company through those means. After all, Netflix is unlikely to have such an influence on my life that I would enjoy spending as much as ten minutes on the phone with friendly Oregonians.

As an ethnographer, I have not, in fact, observed Netflix to any significant extent. I’m just a random customer and, as it so happens, my wife is the one who is getting rentals from them. What little I know about the Netflix business model is limited to discussions about it on tech-related podcasts. And I do understand that Blockbuster is their direct target.

Yet it seems to me that one of the main reasons Netflix has/had been succeeding is that they went into relatively uncharted territory and tapped into a specific market (mixed analogies are fun). Even now, Netflix has advantages over “traditional” DVD rental companies including Blockbuster the same way that Amazon has advantages over Barnes and Noble. It seems to me that Amazon is not actively trying to become the next Barnes and Noble. AFAIK, Amazon is not even trying to become the next Wal-Mart (although it has partnered with Target).

Why should Netflix try to beat Blockbusters?

What does this all mean for corporate America?