The Accordion Theory of Decision Making

When are decisions like an accordion? That’s not a joke with a bad punchline; it’s a legitimate question. And the answer is, “Almost constantly.”

We saw this in last week’s article, and the initial information exploration stage of decision making. In general, information exploration has two key questions to address: “How can we expand our understanding?” and “How do we make sense of the information that we have?” There is an expanding out and seeking more, followed by a distilling down and making sense of what you have.

That dynamic is an incredibly important one, and it’s one that repeats itself regularly throughout the decision making process. This is no more true than in the second overall stage of the decision making cycle: options generation.

Considering options is a theoretically simple and straightforward activity. In evaluating the choices on the table, it is the process of identifying ways that the problem might be addressed. Simple in theory, but much more complex and convoluted in reality.

One of the problems with option generation is one we’ve already acknowledged: the race to the finish line. The irony of this reality is that most groups will tell you that they want to make good decisions. Their behaviour, though, is often contrary, and what they actually attempt to do is make fast decisions. In large part, this can be attributed to the psychological discomfort of sitting with uncertainty. The race to a decision is frequently an attempt to lurch back to the comfort of order and structure.

Identifying options is also limited by another force: our inner censors. Option generation is an act of creativity. It is very frequently a product of brainstorming, where good answers are often inspired by suggestions from out of left field. This isn’t to say that the off-the-wall suggestion is the answer to the problem, or that it survives unscathed. But it can frequently be a contribution that prompts alternative thinking that leads to an answer.

Answers from left field can only be considered if they are contributed. This is where meeting participants often get in their own way. Similar to the rush to the finish line, there is often a drive to get to the good ideas. That creates a couple of different challenges. For starters, we often get hung up on appearances. We don’t want to open ourselves up to ridicule by the group, so ideas that might be seen as crazy or off-the wall get quashed. We start censoring what we contribute, only identifying suggestions that we think would be supported by the group.

The larger challenge is that we don’t try to just contribute suggestions. We want them to be awesome suggestions. Amazing suggestions. Suggestions that miraculously solve the problem at hand and move the group forward exponentially. Doing so creates a cognitive roadblock that it is very difficult to get past. And it’s one that is likely to derail contributing most ideas before they are even fully articulated.

A simple exercise illustrates the problem that I’m talking about. Consider your answer to the question, “What’s the best restaurant in your city?” (This could also apply to music, books, movies…) It’s a simple question to ask. And it’s a complicated question to answer. What makes it complicated is one single word, “best.” Throwing that word in their exponentially increases your cognitive load, and virtually guarantees I won’t get an actual answer.

The complication in asking for “best” is how we have to solve the problem. We need to generate a list of options. We need to simultaneously invent a set of criteria to evaluate those options. We need to sift options through criteria to come up with the most positive option. We may also throw in some self-doubt about the effectiveness of the criteria. And all of this needs to happen in our heads, before we have a hope of articulating an answer.

A different question entirely (and a much easier one to answer) is: “What are all the really good restaurants in your city?” Or, even better: “What are all the restaurants in your city?” Answers to this question will result in everything from McDonald’s to that little romantic bistro downtown that you went on your first date with your partner. From there, it’s an easier exercise to answer: “What’s the best restaurant for a burger?” Or “What’s the best restaurant for a good steak?” Or “What restaurant makes the best martini?”

The cognitive barrier that “What’s the best restaurant” creates is the result of needing to use two very different thinking modes at the same time. Simply put, our brains don’t like doing that. Generating a list of options is a process that is known as divergent thinking. We are opening up and expanding options. We are deliberately seeking more choice. Whether listing restaurants or solving a strategic problem, we are looking for as many different selections as possible.

The key to successful divergent thinking is not editing while we do it. And every time we weigh the merit of an idea before we contribute it, we are editing. Divergent thinking is at its best when we are able to get to a place of pure creation, without worrying about the consequences or implications of the ideas. Go deep, go long, go crazy. Doing this requires two things: building trust in the group that contributions will not be criticized, and trusting in the process that we will sift and sort ideas later.

The sifting and sorting is the product of what is known as convergent thinking. It’s where we evaluate and reduce options. We hone in on the best (or at least, the better) options. Convergent thinking is about closing and narrowing down choice. It’s where we apply criteria to eliminate options and avenues that aren’t productive, and focus attention on ones that are more promising.

There are those that will view all of this as an abundant waste of time. Why bother generating options if you aren’t going to use them all? Why take the time to go broad, when you’re just going to sift down and sort out a few? The challenge with this is that we can’t get to the few unless we start with the many. Doing so leads us right back to the cognitive hurdle that we started with.

More importantly, really good divergent thinking is what feeds and supports better cognitive thinking. More directly: we make better choices when we have better options. It is having the crazy, far-fetched and potentially ridiculous options on the table that allows us to sort and see and create potential ways forward that all of a sudden seem much less ridiculous and much more promising. All of this depends on contributions that don’t start by worrying about whether an idea will be perceived as ridiculous in the first place.

A recent business case that I helped build illustrates this point amply. When I started working with the client, they had one option (which was to build a new facility). By the time we finished our initial conversation about the project, it was abundantly clear to all concerned that the option they had started with wouldn’t work. But we also had eight different alternatives on the table that were potentially much better and more relevant solutions to the problem at hand. Figuring out which option made the most sense still required analysis, but there were now meaningful options in play.

Interestingly, we are once again at a point in the decision making process where on-line meetings have an opportunity to outperform their in-person counterparts. Effective brainstorming is a difficult process to facilitate well. It is too easy for participants in a meeting room to self-censor and edit. It is hard to build the trust that allows creation and inspiration to emerge without fear that ideas might get judged. That judgement often occurs—sometimes with good natured intent, and sometimes not—when the options are explored and discussed in later meeting stages.

There have been attempts to address these challenges, often through the use of technology. What were known as group decision support systems attempted to manage process stages and anonymize input in ways that encouraged contribution. While these did a very good job of expanding options and encouraging idea generation, the later sifting and sorting and considering was often more complicated and awkward.

We also have better insight into how good brainstorming actually works. While the perception is that generating ideas in a group would produce more creative input, that’s not actually true. Brainstorming works most effectively when people are individually able to generate ideas, and then explore them further as a group. Not only does less editing occur, but the quality of generated ideas is often considered to be much better.

That once again creates interesting implications for how we facilitate meetings in a remote context. And we once again find ourselves valuing breaking up what would often be one continuous meeting into more narrowly defined and better chunked periods of meeting activity. What defines those chunks is a product of the thinking modes that are required, and the process of working through the material that needs to be explored.

Options generation happens after information exploration. The process of options generation optimally starts when we’ve got clarity on what we are trying to accomplish, and agreement on the critical elements of information related to the problem at hand. It’s important to clarify that this information doesn’t at this stage need to align or agree. There may be conflicting perspectives, unresolved viewpoints and tension between known facts. All of that is just fine. But there is information on the table, available to everyone, that it is possible to work through.

From there, you might start with a simple and individual brainstorm. That’s not to say that you start working on possible options right away. If there are indeed different viewpoints and perspectives of the problem, you might start with a brainstorm of the different realities or potential futures where the agreed upon information is all true. Within each reality, you could then further brainstorm possible options. The point initially, though, is to treat it as an individual exercise of the divergent production of ideas and suggestions, and to encourage them to be as far-ranging as possible.

From there, you might bring the group together to share and sift the contributions and makes sense of what has already been produced. Depending upon the circumstances, that could occur a few different ways. You might have each participant individually present their perspectives and ideas, and talk about the influences and thoughts that came to them in generating their options. Or you might start by anonymizing all of the input and attempting to make sense of the contributions as a group exercise.

Depending on where you get to, you may break out once again for further individual—and then group—consideration of additional options. This is where you may also build on the options as an act of creativity, taking elements from several different ideas and combining them to form and contribute new alternatives. Or you may decide that the quality of input is such that you can start working through what has been identified to figure out meaning and opportunity.

Convergent thinking can also be an individual—as well as a follow-on group—exercise. This starts with gaining agreement on the criteria that you are going to use to actually make choices. Criteria generation can also be its own brainstorm, with its own cycle of divergent and convergent thinking. Given the influence that criteria will have in identifying and selecting further choices, defining them is an important exercise on its own.

The act of applying the criteria can also be segmented and chunked. You might have individuals score and rate each option based upon the criteria that are defined. From there, you can compile the overall results. Those results can also prompt additional exploration, identifying and considering in particular those options where there was a range of perspectives and views. These differences may be a product of how the options are perceived, how the criteria are interpreted or the values that are at play within the group. Sorting to the most promising options can then lead to a further stage of exploration and analysis.

Once again, this may seem more formal and structured than many of the strategic decision making processes we have observed in the past. In theory, it is simply making explicit that activities and actions that should be in play in making those decisions. Once again, it’s highlighting how decision making processes can be structured, and how individual and group contributions can be channelled and supported. And most promisingly, there is every potential for the process to work more effectively in a remote meeting context than it does in person.

What is important to recognize in thinking about structure is the role that modes of thinking play in identifying and sectioning the discussion. Divergent and convergent thinking are at play in everything that we have explored so far. They shape how we collect and make sense of information, and they significantly influence how we cast about and sift through options. Expansion and contraction is the natural rhythm of decision making, and breathes into every step of the process. Like an accordion, we need to first open up, and then contract back down to land on a single note.

3 Comments to “The Accordion Theory of Decision Making”

  1. Michael Hilbert says:

    It has been my experience that company and team culture are a large part of how the options generation phase plays out. As you indicated, rushing to a conclusion simply to get an end, or lack of physiological security to offer that “left field” suggestion are all dynamics on how the team reacts to determining the best path forward. This process is affected my overall company management and philosophy.

    I recently read an article by Lihn Tran (InLoox Blog, 7/2015) about Project Management at the Disney Corporation that speaks directly to your point. Their creative strategy (and I believe their culture) is “If you can dream it, you can do it” (Walt Disney.) I will quote the relevant section from the article .

    “Blue Sky”-Brainstorming > At WDI not even the sky’s the limit. During the “blue sky”-brainstorming phase, coming up with outrageous ideas is explicitly welcome. In other companies, brainstorming sessions might last a few hours or, at most, a day or two. At WDI, this phase can last a few years. What’s more, they never completely throw away an idea. It’s common that Imagineers go back to an idea after a few years and then realize it. Project managers are often afraid to be creative or to put forward crazy ideas as they can pose risks, but also because cost and time constraints put pressure on them to be efficient rather than use their imagination. But it’s actually because PMs are under so much pressure and project success depends on so many variables, which are sometimes unpredictable, that they need to be creative. For some issues, you won’t find a suitable solution in a text book, you need to have a little imagination to solve them.”…..

    Would this not be a great environment to develop ideas and get to that elusive On-Time Delivery date with complete project success?

    Thanks for your thoughts Mark… Stay Safe All..


  2. Laura Lagendyk says:

    Hi Mark,

    Really enjoying this decision-making series and have shared it with colleagues who have also indicated their appreciation. We do a lot of work with groups and decision-making, often virtually (even before the pandemic).

    Two things struck me as I read this week’s blog. One – although we cannot separate professional training from the culture in which the professionals work, I have also wondered if certain professions are prone to – trained to? – move quickly to convergent thinking. In the health context in which I work, for instance, I have noticed that physicians are often impatient with divergent thinking. They simply want to make the decision and move on. When I think about it, physician training is all about hearing a lot of information (e.g., symptoms), sifting out what’s relevant (at least, from the physician’s point of view) and making a diagnosis – often as quickly as possible. There is no arguing that this is an essential skill for physicians in medical emergencies, or even in medical practice in general. But when I have worked with physicians in other non-clinical contexts, the tendency to cut to the chase, the sense of “hurry up and make the decision already” is often part of the group context. Not always easy to facilitate the group through that impatience. Having the confidence to lead through that impatience is bolstered by being able to articulate the benefits of doing so. Your blog helps with that.

    Second point – one good resource on decision-making that I enjoy and aligns with lots of what you have been discussing is Chip & Dan Heath’s book: “Decisive”.

    Thanks for the blog – always appreciate it.


    • Mark Mullaly says:


      Thanks so much for the insightful and extensive thoughts! And thanks for sharing the posts; glad you are finding them of value.

      The bias to convergent thinking is absolutely real, and I love your analogy of working with physicians. I think the bias to closure and decisiveness exists for many of us naturally, but it’s interesting how career and professional training can further reinforce that. Findings ways to pull back from this impulse is a significant challenge, and one of the essential ones in facilitating meaningful problem solving sessions.

      I had not read Decisive, so I appreciate the recommendation. I shall have to go find that.

      Thanks for continuing to read and contribute!


Leave a Comment