The Difficult Problem Framework is a simple tool that requires continuous learning and objectivity. Part II of a two-part series.
The framework above was developed to solve very difficult problems related to organizational change, particularly those now facing the legal field. I realize the framework looks laughably simple. That said, it’s harder to apply than you might think.
Part I (056) summarized my Deliberative Leadership course at Indiana Law, which gave me the opportunity to learn about and reflect on these topics. In Part II, I explain Difficult Problem Framework (DPF), starting first with the basic mechanics of how it works and then providing examples drawn from Deliberative Leadership and other materials.
One caveat: Initially, this post will seem more focused on decision-making than leadership. The goal, however, is effective leadership that has a chance at solving difficult problems. As you will see, the leadership part is already within our grasp. Effectiveness, however, requires more of a set-up.
Box 1: accurate assessments and root causes
Box 1 is the space where someone — a leader / innovator / change agent — seeks to understand a difficult problem and identify its root causes. Thus, Box 1 is primarily about fact-gathering and reasoning. To do it right, we observe the problem, locate relevant data and research, ask questions, listen, reflect over a period of days/weeks/months/years, write out our analysis in a clear and ordered way, and remain on the lookout for disconfirming evidence that reveals faulty assumptions or conclusions. In this respect, feedback loops are especially valuable. Cf. Chris Arygris, “Teaching Smart People How to Learn,” Harv. Bus. Rev., May-June 1991 (discussing double-loop learning).
Box 1 has two potential failure points. First, our assessment of the present is inaccurate due to insufficient fact-gathering, faulty unstated assumptions, lack of rigor, or something else (1A failure). Second, after constructing what we believe is an accurate analysis, we never leave Box 1, as we believe the hard part, or important part, is done (1B failure).
Readers may laugh when I say that Box 1 activities are similar to writing a peer-reviewed academic article. As we all know, however, academic journals are filled with symposia articles on problems that remain unsolved. Obviously, something is missing. That’s a 1B failure, the most common type for those of us in the academic crowd.
In contrast, practicing lawyers band together into firms that place a heavy emphasis on revenue generation. This leaves little time to read, reflect, and understand difficult problems that are one or more steps removed from the immediate demands of client work. Despite being liberated from timesheets, in-house lawyers are not much better. In both contexts, daily responsibilities thwart deep systems thinking. This dynamic keeps the entire profession stuck in a rut of 1A failure, as we are trying to solve our most systemic problems with after-hours resources.
The image below captures our dilemma (H/T Casey Flaherty).
Box 2: change strategy
Box 2 is about effective change strategy. Here, the operant conditioning of law school gets in the way, as we spend most of our time learning how to construct arguments that could prevail in a court of law. If we do it well, our reward is law review membership and a high-paying job. Yet, it’s also a problem-solving approach based purely on legal authority. That’s a big limitation.
Even if a client’s life or business problem might turn on question of law, most clients can’t afford to engage the wheels of justice. And even they can, few leave the courthouse feeling good about the experience. Thus, lawyers with large practices eventually build out a toolbox of nonlegal skills. In fact, Shultz & Zedeck documented 26 different tools. See “Identification, Development, and Validation of Predictors for Successful Lawyering,” LSAC Final Report, Sept. 2006.
If we move on to organizational problems — e.g., law firms, law schools, legal departments, court systems or regulators struggling to adapt to changing time — and we are paying any attention at all, we soon observe that stakeholders are seldom won over by reasoned arguments. In fact, they may not even show up for the meeting. If they do, their head may be elsewhere. For those who show up and listen, they’ll likely want to modify our ideas with some of their own. Suffice it to say, no one leaves these meetings with a quorum for change.
Several years ago, I learned this lesson the hard way, as I was part of a team building and selling evidence-based tools to lawyers. See Post 004 (discussing Lawyer Metrics); Post 016 (same). Straightforward presentation of data, even when connected to bottomline results, is not effective to win over a group of well-credentialed professionals. Cf. Daniel Kahneman, Thinking, Fast & Slow 227-29 (2011) (discussing hostility to algorithms, particularly when they demystify professional judgement). These experiences eventually drove me to diffusion theory and the insight that innovation adoption occurs through a social system where innovators and early adopters go first. When these two groups benefit, the rest of the social system follows. See Post 004 (presenting theory); Post 007 (providing detail).
Diffusion theory, however, is but one of many theories and frameworks that can improve our odds of desirable change. I’ll give a few examples below, many of which are not only connected with leadership, but also meaning, purpose and fairness. But for now, the core point is that Box 2 requires us to continuously learn new ideas and reflect on how they might connect to our difficult problems. This is no less time-consuming than the Box 1 analysis.
Difficult problems and decisionmaking
To summarize: We can only solve a difficult problem if we can accurately assess its root causes. This requires a major investment of time and resources to get right (Box 1). Thereafter, we need to formulate an effective change strategy that goes well beyond explaining/publishing our analysis (Box 2). The purpose of the DPF is to keep these two activities separate and analytically distinct.
Correct root causes + right change strategy = chance at success
Going a bit deeper, solving difficult problems is a thinking person’s game where the biggest risk factors are (a) self-deception that causes us to underinvest in learning, fact-gathering and reflection, and (b) bias and distortion in how we evaluate information. Through work ethic and mental training, we can mitigate these risk factors, but never completely. On this score, I’d recommend four “applied” resources: Charlie Munger, 24 Cause of Human Misjudgment (1995) (75-minute audio); Daniel Kahneman, Thinking, Fast & Slow (2011); Randall Kiser, How Leading Lawyers Think (2011); Ray Dalio, Principles: Life and Work (2017).
Let’s now move from the abstract to the concrete.
Box 1: what are my assumptions?
The example below is based on an assignment in Week 1 of my Deliberative Leadership class.
Imagine you are an executive at General Motors in 1984. For reasons of cost and quality, the company has been losing marketshare to the Japanese. You’ve given this a lot of thought and concluded that the root cause of GM’s woes is an old, expensive and undisciplined workforce protected by overly generous union contracts. Until that gets solved, the company cannot effectively compete with companies like Toyota, GM’s most formidable Japanese competitor.
This problem set is based on New United Motor Manufacturing, Inc. (NUMMI), which was a joint venture between GM and Toyota launched in the mid-1980s at an old GM auto plant in Fremont, California. Basically, this came about because Toyota was making great inroads in the U.S. market. To preempt a protectionist backlash, Toyota needed a plan to shift some of its production to the U.S. and learn how to adapted to U.S. workers. For GM, it was an opportunity to learn Toyota’s lean production methods, which combined world-class quality with world-class efficiency. This story is expertly told in a 1-hour podcast produced by This American Life. See NUMMI 2015, episode 561 (July 17, 2015).
The first part of the episode details the problems that existed at the GM Fremont plant prior to its closure in 1983 — pretty much non-stop drinking, drug use, absenteeism, and antagonism toward management. According to Bruce Lee, who ran the western region for the UAW, “It was considered the worst workforce in the automobile industry in the United States. And it was a reputation that was well earned. Everything was a fight. They spent more time on grievances and on things like that than they did on producing cars. They had strikes all the time. It was just chaos constantly.”
In negotiating the re-opening of the Fremont for the NUMMI joint-venture, the UAW demanded that GM and Toyota rehire a substantial portion of the old Fremont facility workforce (it would turn out to be 85%). Remarkably, Toyota was willing to go along. According to NPR automotive correspondent, Frank Langfitt, ” Toyota execs believed their system would turn bad workers into good ones.”
The rest of the episode tells the story of how, under the Toyota production system, the Fremont facility went on produce world-class quality on par with the rest of the Toyota system. What changed everything was the inclusion of workers in a team-based process of continuous improvement. For the first time on their careers, these old, tired workers were asked for their ideas on how the cars could be made better and more efficiently.
NUMMI is a vivid example of a 1A failure. The root cause of the problem was not the people; it was the system. For me, this lesson hit close to home because during the early 1980s, I was a college student living in Cleveland, Ohio. Pretty much the entire region blamed the workers and union for the decline of the industry.
The lessons of NUMMI are supported by others materials assigned during Weeks 1 and 2, such as Batman, This American Life, Episode 544 (Jan. 9, 2015) (expectations in our head have a profound effect on the physical and social world); Viktor Frankl, Why Believe in Others, Ted.com (video) (Jewish-Austrian neurologist who survived Nazi concentration camp and wrote Man’s Search for Meaning exhorting group of Americans to elevate their expectations of others and thus enable them to reach their full potential); Carl F. Braun, Management and Leadership (1948) (leader of C.F. Braun & Co., an international engineering and construction company, outlining the principles of human respect, dignity, and collaboration that underlie the company’s financial and technical success).
As these excerpts suggest, perhaps the root causes of organizational and institutional malaise are not exclusively gaps in logic or analytical rigor. Rather, a major root cause could be lack of clarity around purpose and, until that gets resolved, worry over status, hierarchy, and security.
Box 2: the missing link
I’ll admit that it wasn’t until that fourth year of Deliberative Leadership that I realized that there was a second box. The turning point was this spring when Alli Gerkman, Director of Educating Tomorrow’s Lawyers, visited by class. One of her selected readings was an article on an internal study by Google’s People Analytics group, dubbed Project Aristotle, that attempted to identify the attributes of high performing teams. See Charles Duhigg, “What Google Learned From Its Quest to Build the Perfect Team,” N.Y. Times Magazine, Feb. 25, 2016.
Google had long observed wide variations in team performance. If it could isolate the factors consistently associated with high performance, perhaps they could be scaled across the entire organization. Yet there were many false starts. In particular, Project Aristotle invested a lot of time and resources looking at how team compositions based on personality, skills or background affected team performance. “No matter how researchers arranged the data,” wrote Duhigg, “it was almost impossible to find patterns — or any evidence that the composition of a team made any difference.”
Eventually this led the Aristotle team to the social science on group norms. One line of this research suggested that norms within groups may produce a “collective IQ” that is distinct from the intelligence of any single team member. This hypothesis proved to be the missing link in Google’s research.
So, what is the cultural factor that explains high-performing teams at Google? Psychological safety.
“Google’s data indicated that psychological safety, more than anything else, was critical to making a team work.”
According to Professor Amy Edmondson of Harvard Business School, who conducted much of the group norm research relied upon by Project Aristotle, psychological safety is a ‘‘shared belief held by members of a team that the team is safe for interpersonal risk-taking. … [and] a sense of confidence that the team will not embarrass, reject or punish someone for speaking up.” Edmundson, “Psychological Safety and Learning Behavior in Work Teams”, 44 Admin. Sci. Quarterly 350-383 (Dec. 1999). From the outside, a psychologically safe group might appear free-flowing and chaotic. Yet, because of group norms, the members are very good at allocating airtime equally and truly listening to one another.
This past spring, I gave two talks on leadership, one to group of young lawyers and another to a group of students from several law schools. In both talks, I explained the Google research, presented the definition of psychological safety, and asked audience members to anonymously complete a notecard that said whether their workplace or law school was psychologically safe (“yes”, “no”, “something in between”). I then collected passed along the basket of notecards, letting each person draw a random card. Finally, I polled the results by having audience members raise their hands. In both cases, less than 1/3 reported feeling psychologically safe. That’s a problem.
Connecting it together
For all four years of Deliberative Leadership, I have assigned a well-known article on authentic leadership. See Bill George, et al., “Discovering Your Authentic Leadership,” Harv. Bus. Rev. (Feb. 2007). It’s an attractive thesis — that the most effective leaders “demonstrate a passion for their purpose, practice their values consistently, and lead with their hearts as well as their heads.” Yet, the Google article got me to think that perhaps the authentic leader’s effectiveness flows from the group norms they foster, especially psychological safety.
One of the repeat readings in my class (picked by more than one guest lawyer) is True North: Discover Your Authentic Leadership, which is basically the book version of the HBR article. It’s primary author is Bill George, former CEO of Medtronics who now teaches at Harvard Business School. True North was picked again this year, in close proximity to Alli’s article. Thus, I took the opportunity to to better understand the book’s methodology.
Below is a graph from True North that, the authors claim, fits the pattern of many of the 125 leaders in the study.
Often, according to the authors, authentic leaders are forged during a period of extreme hardship. Through the “crucible,” leaders finally develop the courage and confidence to live by their own values. Perhaps one way to establish a psychologically safe workplace is for the boss to explain difficult decisions in the context of their own learning, including painful failures and setbacks. I’ve had my own crucible moments in life. At age 55, I can say that crucibles really do burn away our allegiance to things that are stupid and really don’t matter.
In terms of Box 2 change strategy, the Google research and Bill George’s authentic leadership are connected together with the work of Chris Arygris, the late, great HBS professor who focused on organizational behavior, organizational learning, and change management. In a 1991 article in the Harvard Business Review that was later republished as a HBR Classic, see “Teaching Smart People How to Learn,” Arygris discusses his work with elite management consultants. The primary theme is engrained defensive reasoning that keeps very smart professionals from learning why many of their engagements continually fall short of desired results for them and their clients. Arygris reports many heroic efforts to change this dynamic that all end in failure.
Arygris then relates the story of a CEO of a large organizational-development firm who was so disgusted with the pattern that instead of preparing for an upcoming meeting, he decided to script out the failure in advance. He divided his work into two columns. On the right side, he wrote out the likely dialogue that would take place. On the left side, he wrote the thoughts and feelings that he would likely have during the meeting “but that he wouldn’t express for fear they would derail the discussion.” Then, instead of having the meeting, he used the time to analyze the scenario with his direct reports.
What happened next was an honest dialogue in which the CEO became privy to the honest but unspoken views of his entire team. Then they could finally hear him with a new set of ears. Finally, real progress could occur.
Below is a stylized version of Aygris’s recommended approach, which I use in my Deliberative Leadership class:
This is a potentially useful Box 2 tool. Do you have the courage to give it a try?
As legal organizations and institutions struggle mightily to adapt to rapidly changing times, there is renewed and growing interest in the topic of leadership. I am confident great things are going to happen as a result.