Going with the flow: Cascades can hinder team decision-making

My last post suggested that making decisions in diverse teams can help avoid decision-making traps, and there’s research evidence to support this view.¹ However, team decisions are often no better than individual decisions—and sometimes they’re even worse.² So what’s going on?

Essentially it has to do with the difference between having diversity on a team and managing the team environment and process in a way that enables the group to actually benefit from that diversity.

There are various mechanisms through which biases and poor decision-making can actually be heightened rather than diminished on a team, even a diverse one.

One way is through cascades. If you picture a waterfall, you’ll have a good idea about how a cascade works on a team. Once water starts flowing in a particular direction, downhill or over a cliff, there is little you can do to stop it. Similarly, once ideas, discussion, or decision-making on a team start moving in a particular direction, momentum often keeps them moving in that direction. Even if diverse views exist on the team, they’re unlikely to divert the team’s direction once it’s been set, in part because people are pretty unlikely to voice disagreement with an idea that gets early visible support. There can be two primary reasons for this:

• Reputational cascades occur when people don’t disagree because they fear looking bad or being punished in some way. These types of cascades are particularly likely to occur when a leader has stated a strong opinion or maybe even already reacted negatively toward someone who disagreed with them.
• Informational cascades occur when people don’t voice their disagreement because they assume early supporters must know something they don’t. In other words, they don’t offer their differing opinion because they suddenly suspect they might actually be wrong.

Either way, cascades lead people to self-censor, and when team members are doing that the team doesn’t benefit from their diverse perspectives. Moreover, decisions made by the team end up looking more unanimous than they actually are.

I’ve seen this happen on teams that are dominated by one or two Business Chemistry types, especially the more outspoken types, namely Pioneers or Drivers. They both tend to take charge in group settings and their styles—Pioneers tending toward being energetic and talkative and Drivers tending toward being competitive and direct—meaning they’re not likely to sit back and wait to hear others’ thoughts before offering theirs. So a group full of Pioneers and/or Drivers will tend to cascade in the direction they set with their early comments.

You can imagine being a Guardian or an Integrator in this situation. Maybe you’ve even experienced it. Guardians tend to be reserved, to speak and make decisions slowly, and to be non-confrontational. Particularly if they’re a minority in the group, what are the chances they’ll speak up at all or do so early enough in the process to divert the cascade? Likewise, Integrators, too, tend to be non-confrontational, and are also trusting and often focused on gaining consensus. If the group appears early on to be leaning in a particular direction, how likely do you think they are to offer a divergent perspective? Further, knowing that both Guardians and Integrators tend to be risk-averse, they may see very little upside personally, to sticking their neck out.

A second reason diversity on a team may fail to improve its decision-making is because of Hidden Profiles. In groups, we tend to focus on information that is shared by many and dismiss information that is exclusively held by a few. The majority tends to rule and the minority becomes virtually invisible (i.e., their differing profiles are hidden). As a result, we often benefit little from unique perspectives that could improve our decision-making.

Again, this problem is most likely to occur on a team dominated by one or two types. For example, a team of Guardians, who tend to be risk-averse and to stick with what’s tried and true, might specifically seek out some Pioneers to help them expand their horizons. But chances are, once those Pioneers have shared their thoughts, the Guardians will nod thoughtfully and revert back to their own perspective, because it’s shared among them and as such will overwhelm the different perspective that’s been offered.

Together, cascades and hidden profiles mean the presence of diversity on a team may not be enough to improve decision-making. That diversity must actually be managed in order to combat decision-making pitfalls. Here are some suggestions for doing so:

• When possible, create teams that are well-balanced across Business Chemistry types—representing a type with one or two tokens may not be enough—aim for a relatively even representation of all types.
• Acknowledge the importance of each type’s perspective and encourage active participation by all. This should be done in a way that makes more introverted types comfortable sharing. For example, asking team members to submit ideas or opinions in writing prior to a meeting gives more reserved or deliberate people the space they need to consider things and share them in a thoughtful way.
• We now know, based on decades of research, that individual brainstorming is more effective than group brainstorming, yet we persist with this group activity. Instead, ask people to brainstorm alone and then share with the group in round-robin fashion. This often leads to more, and better ideas.
• When a team is dominated by one or two types, encourage the minority types to speak first. This gives them at least some chance of influencing the direction of the conversation before a cascade takes hold.
• If a team is light on a particular type, ask others to “think like” that type. Do this early in the conversation.
• Raise awareness about decision-making pitfalls and explicitly ask “are we falling victim to these?”

Have you seen this kind of thing happening on your own teams? What have you done about it?

Endnotes

¹ Phillips, K. (2014). How Diversity Makes Us Smarter. Scientific American.
² Sunstein, C.R. & Hastie, R. (2014). Making Dumb Groups Smarter. Harvard Business Review.