It’s The Decision, Stupid

We’re going to take a bit of a tangent this week.

I’ve been contemplating how to facilitate strategically important discussions online. That started with an exploration of the problems that we have to solve and continued on with a consideration of the priorities in encouraging participation involvement. A logical proposition, because without participants you don’t have a meeting.

From there, the intention was to explore the process and the technology and the intersection between those things in moving to facilitating online, rather than in person. And therein lies the problem.

It is easy to treat this as an exercise in replicating online what we do in person. And I have come to realize that is a trap, and a dangerous one. Firstly, it presumes that the way we meet in person is effective (it is often not). Secondly, it imagines that translating in person meeting approaches to online equivalents will produce the same results (they very likely won’t). What you will likely wind up with (and I suspect current experience is playing this out frequently) is either meetings conducted quickly that produce unsatisfactory agreements, or no agreement at all.

There are more than a couple of problems at play here, especially in our current environment. One of the biggest is uncertainty. Simply put, as human beings we don’t like to make decisions in the face of uncertainty, no matter how much we protest otherwise. We tend to default to the safe and accepted approach, or we push off the decision. In both instances, we aren’t addressing the substance of the uncertainty; what we are doing instead is largely avoiding and ignoring it, and substituting the complexity of the situation with a much easier problem to solve.

We also tend not to make decisions well in groups, despite that often being their express purpose. Theoretically, the reason that we try to make decisions in groups is that we get better input. We try to leverage the diversity of expertise, responsibility, insight and experience that is represented to get a better outcome. What research has demonstrated actually happens, though, is something quite different. Rather than emphasizing the differences of experience in expertise, groups quickly coalesce around the lowest common denominator of what everyone can agree on. We throw out the differences and accept what is simple and uniform.

Technology doesn’t help in all of this. Apart from basic issues of connectivity and technology familiarity (and the need to constantly remind people to mute or unmute themselves), the current state of meeting technology results in delays and warbles, the smallest of which complicate interactions, create frustration and rob us of the social cues that we are normally used to experiencing in real life. That can lead to limiting input, shutting down participation and attempting to get through the meeting as quickly as possible. This again defers complexity of discussion in favour of what we can all agree on.

For those paying attention at home, there’s an underlying theme emerging here: when faced with uncertainty and frustration, normal human nature is to take something difficult and complex and substitute in its place something simple and straightforward. It is an innate human bias. We do it almost without recognition—although if you are paying attention and looking for it, you can see it when it happens. What’s worse, though, is we take the answer to the simple question as given, and accept the complex problem as solved.

This where we come to the crux of how to think about moving strategic conversations into an online format. If we think that technology-mediated meetings can work the same as normal ones, we will very likely fail. And if all we do is find technology substitutes for in-person interactions, we will also fail. This is because what we are doing is exercising the same bias: we are taking a difficult problem and substituting in its place a more simple solution. We are taking the question of “how do we make good strategic decisions when they are mediated by technology?” and replacing it with “how do I do this in-person exercise online?” And while there is likely an answer to the second question, it will not satisfactorily address the first one.

That was a humbling realization to come to, but it’s an important one. It is about the decision, stupid. Not the meeting. The meeting is supposed to be a vehicle to decision making; to do that well, though, requires figuring out first how to arrive at the decision, and then figuring out how to structure the meeting—virtually or otherwise—to get there. So what we are going to tackle, today and going forward, is the actual question of how to do strategic decision making in a meaningful and effective way via technology.

As an example of the kind of decision—and the kind of problems—I am talking about, the Carter Race Case is a decision-making scenario I use often when I am facilitating workshops on decision making. Without getting into the details specifically, participants need to make a decision on whether or not to race in a situation where there is an undiagnosed problem with the car, and disagreement amongst the team as to whether or not the problem exists.

The way I normally facilitate this is to have participants make an individual decision first. They are asked to review the case, make a decision, and answer three fundamental questions:

• What was the data that was particularly helpful within the case in making your decision?
• What data was not particularly helpful or relevant in making your decision?
• What information did you not have that you would have liked access to?

The result is that there has been enough consideration to say not just that a decision was made, but also how. They have considered how they looked at the decision, and have given some critical consideration to what was missing or uncertain that they would like to have clarified.

In my experience facilitating this case—and I have probably done that about one hundred times now—how the decision plays out is surprisingly consistent. In a room of 40 people, between five and ten will make the decision not to race. The remainder of the room is a solid thumbs up. They are divided into groups that include both those who want to race and those who do not, and then asked to make a group decision. Despite starting out the exercise knowing there are different viewpoints and perspectives—and not everyone has come to the same decision the same way—the result is surprisingly uniform. At the end of the exercise, virtually every group decides to race. To date, I have had two groups that have not.

Think about the implications for that for a minute. In making individual decisions, there are differences in choice, and differences in reasoning. In groups, though, differences are smoothed over—or opposed or ridiculed or ignored—and the result in a group context is consistent and predictable uniformity. The purpose of the group is that it is supposed to explore divergent viewpoints. The actual reality is that the group squashes differences.

Substitution happens in this scenario, as well. What is presented as a decision making about uncertainty—we don’t know what the problem is with the car—gets resolved as a decision about racing. “Racing is risky.” “You don’t win a race in the pits.” “You lose 100% of the races that you don’t enter.” We start with complexity, and what we get back in return is rhetoric. The facts get ignored, and what emerges instead is simplification and slogans.

If this is how decisions work in person, just imagine how the same scenario might play out online. The results would probably be similar. While discussion can get quite heated in person, via technology people are likely to back off and not engage. There is likely to be even less expression of differences and alternatives—and less willingness to explore them—even though we know that they exist. The awkwardness of meeting online—and the consideration of each other required to allow for full participation—means that again we opt towards simple, straightforward, painless and fast. While we may have arrived at a decision at the end, that’s not to say it is a good decision. It is very likely a decision without agreement and commitment, and with relatively lukewarm support.

If we want better decisions, we need to take a different look at the process we follow. That includes considering not just the approach that is taken in arriving at the decision, but also how that sequence of steps is managed. That highlights an important point: most people consider decision a result, a thing. It is the answer at the end of it, and the answer is what people in general—and groups in particular—rush towards in scenarios like the one that I have described. What gets emphasized is the choice, not the means of getting there.

Decision as process takes a different view, in that it sets up and structures the conversation to consider specific perspectives as a way of building towards a resolution. If that sounds slow and deliberate, it arguably is. But if we care about how the result plays out, and view the decision as an important one, then investing that time and effort should be reasonable.

Looking back at the example, the initial step is a reasonable one. Individual participants are asked to make a choice, explain the data they used and how they used it, and to identify additional information that would be valuable. We are mapping out what we know, what we care about and what we don’t know. Groups are given the same task, with the same question, but it’s here that the process breaks down. If we care about how to facilitate strategic decisions in groups, this is where we still have work to do.

Part of the problem here is that when groups are given the assignment, they are given the whole assignment. They are not asked to return until they’ve done all the work. When they do, they present the choice they made, and why. If we are going to get group differences to play a role, then we need to take steps to highlight and explore the group differences.

How this discussion might be approached is as a series of conversations, not just one. The goal would be to first explore and identify what differences in viewpoint and perspective exist. We know that they are there, so let’s get them on the table. Questions that I might ask to explore this particular scenario are:

• What are all the reasons that exist to support racing?
• What are all the reasons that exist not to race?
• What are the things that we are certain about in making this decision?
• What are the things that we are not certain about in making this decision?

What all of the above help to do is to move us away from a decision, not towards it. Rather than focussing on the answer first, and rationalizing why we made it, we are focussing on what we do and don’t have to support finding an answer.

Continuing to expand the conversation and grow perspectives and options, I might then focus on the certainties and uncertainties. What’s important to recognize here is that—in this case, as in many complex decisions—what is known for sure is going to be a relative short list, and often an inconsequential one. The uncertainties are going to be much larger and much more relevant. That’s why simplification and substitution happens: uncertainty is messy and complicated and we don’t like it, so we put it in a box.

Questions that I might ask to explore the uncertainties are:

• How significant an impact is the uncertainty?
• How much don’t we know?
• What would need to be done to resolve the uncertainty?
• What are the consequences of not knowing and proceeding with a decision?

Some of the uncertainties can be resolved with investigation. Some of them will be unknowable and unresolvable in advance. That’s where understanding the consequences becomes important, and exploring the impact of those consequences. What we are doing now is starting to think about choices, but in doing so to think beyond those choices to how they might play out.

To do this, we might now explore the decisions themselves. In this particular scenario, it’s a binary choice. You either race or you don’t. But it’s the consequences of those choices that matter. And so for each of the choices I might ask:

• What’s the best thing that can happen if we made this choice?
• What is likely if we were to make this choice?
• What is the worst thing that can happen if we make this choice?

The insertion of the question of “what is likely?” is deliberate. Asking what is the best and worst that can happen forces a discussion of consequences, and the range of those consequences. But they are both at the extremities: potential consequences with low probability. Calibrating these against what is probable has some value.

The final part of the exercise before getting to a decision would be to go back to the consequences that were identified, and ask one fundamental question: “What would need to happen for us to be okay with this outcome, if this is the decision that we make?” For the best possible outcome, the answer is “not much.” For the worst possible outcome, the answer might be “That’s not acceptable.” Alternatively, for it to be successful, there might need to be some additional work, due diligence or investigation that is required.

In the racing case, most groups decide to race. In doing that, as I’ve pointed out, most ignore the data and substitute it for a generalized rationalization. Most groups don’t consider the worst possible scenario, or in any way give it credence. If they did, they might not be quite so enthusiastic, or at least temper that enthusiasm with the need for some preliminary research and exploration. That realization may be what is needed to reframe a theoretically unpalatable “no” decision as “not now.” The popular, easy decision is not necessarily the optimal, considered decision. But realizing that takes more discussion and participation than we might be used to.