Transcription
I'm going to tell you about a big picture problem-solving framework that McKinsey and Company use as they work for their clients. You may never work for McKinsey or even external clients, but you can still learn from this framework for your own professional purposes. As a leader, you can recommend that your team use this problem-solving process to tackle whatever big issues you face. And if you're ever on a team that's working through a problem-solving process or steps, this is a big picture framework that will help you understand how to participate effectively in each key part. And that way, I'm not going to give you all of McKinsey's specific terminology. I'll explain it in a way that'll help you apply it to almost any professional setting. If you're keeping track, this is part four in a four-part series on McKinsey frameworks.
As part of my work as an outside consultant to McKinsey, I helped run practice problem-solving sessions and watched many of their new consultants work through the first half of these steps in problem-solving scenarios that we did live in the room. So, I'm going to emphasize the first half of the process where the group actually works together, but I'll share all the steps online. I've seen slightly different versions of this process, sometimes with six steps, sometimes with seven. My video is not to be definitive and authoritative. I'm grounding what I say in a podcast on McKinsey's website that was from 2019. That's the basis for selecting this version of the steps. And I'll put a link to that episode in the description below so you can listen for yourself.
The first step is defining the problem. Sometimes they call this problem definition. This is where the group asks questions like, "What is the actual problem we're trying to solve? What are the constraints? Is this a long-term problem or a short-term problem or both? Have we identified some root causes or are we still talking about symptoms?" I've seen a lot of groups in professional settings skip this key step. The trouble with skipping this step is that if you were to hypothetically ask each person in the room to write down what they believe the problem was at the beginning, it's very likely you would have completely different answers. But by the end of this conversation, using the power of the group, discussion, debate, analysis, give and take, the group should be able to agree on a basic definition of the problem and write out a concise statement of exactly what that problem is. For example, you might write out a problem like this: "How can we improve the effectiveness of online learning?" This is a completely hypothetical situation, just an illustration to walk you through McKinsey's problem-solving process. Either way, their problem should be worded in a way that calls for solutions to that problem.
Step two is to dissect the problem into its component parts. Yes, we already identified and defined the essential problem, but now through group discussion and the use of diagrams and logic trees and other tools, we're going to identify the most significant aspects of the problem. Many problems have multiple causes or symptoms. Some issues have moving parts. McKinsey might start, in fact, by using the pyramid principle that we talked about in an earlier video in this series to diagram the component parts of that problem. McKinsey uses the pyramid principle all the time, every day for everything. Engineers and other professionals might prefer to use a fishbone diagram. As an illustration of how this works, think about what your mechanic does when you bring your car in and tell them what is wrong with it. It's making a certain noise. It has a problem you can't quite identify. The mechanic asks you questions at first and then starts looking at the car himself to see if the problem might be, for example, mechanical, electrical, or a computer problem. If it's a mechanical problem, you break that down further. Is it part of the main engine, the transmission, the fuel system, etc.? The mechanic is using a mental logic tree to simplify the problem-solving process and figure out what part of the engine he should be looking more closely at. And we do the same thing when we're trying to improve, for example, the effectiveness of online learning. You might realize that one of the major problems with online learning is that the platform or the technology is nowhere near as interactive as in-person learning. Or you may decide from your analysis of the problem that teachers who were trained for the classroom may not necessarily have the skills to create a dynamic online class. So as we dissect the particular problem, we end up with two or three related problems that we need to solve. Dissecting the problem like this helps simplify the task.
Third is prioritization. This is where the group asks about how they can have the greatest impact on solving those problems. In the group problem-solving sessions I observed at McKinsey, the consultants often ask, "What are our levers?" For example, if you're trying to improve the effectiveness of online learning, what are the levers you can pull or the variables that you can control? You might be able to influence how to hire better instructors, the learning management system you're using, or training. Those might be good levers to prioritize, but you may realize there's not much you can do about other issues like budget or the short attention span of learners. This is a bit like seeing the issue through the Pareto principle, the 80/20 mindset. What 20% should we focus on that will create 80% of the results we want? And by identifying and prioritizing your most influential levers, you can focus your attention on actions that will have the largest impact. You don't want to waste your time too much talking about variables that are out of your control, levers you cannot pull.
Fourth, develop a work plan to look for data and answers that will help the team create an informed response to the problem in those later steps. There's no one-size-fits-all to developing a work plan. The plan depends upon the first three steps. You might have a work plan where you have a few subgroups on the team and that each of those teams are working their part of the problem. In terms of making online learning more effective, you may want to create a work plan that involves three parts. The first is a subgroup that researches better interactive learning management systems. Another subgroup on the team would spend their time identifying some effective training that existing teachers can go through. And the third group might work with human resources to figure out better ways to hire teachers who specialize in online learning. The details of this plan don't need to be all spelled out in advance because it's quite likely that as everybody looks into their tasks a little further and starts collecting information, you're going to learn things along the way that'll change your focus and lead to new questions. The key to a good work plan is that it has an initial focus, a clarity, and provides an actionable starting place to get to work.
Now, when it comes to actually working through this plan, it could take a day or two. It could take weeks or even longer. So, in the groups that I was working with and observing and coaching, this is typically where we'd stop our problem-solving simulation. So, on a day-to-day basis, I never got to see them tracking down data. But in your professional roles over the years, I'm sure that you have had to go out, ask questions, look for answers, collect data, and so forth. So once you've done that, let's move on to the next step.
Step five in McKinsey's process is to do an analysis. This work happens behind the scenes. So it's typically not part of an ongoing group discussion. This is where each subgroup that's gone off to work their part of the plan takes that information they've gathered and analyzes it as best they can. You might have tons of quantitative data to comb through or you might just have a few pieces of qualitative data. Either way, you want to make sure your analysis provides a healthy, well-rounded picture of the problem and likely action steps that could solve that problem.
Step six is this extra step of synthesizing your analysis into clear, distilled, understandable recommendations. This usually involves a well-developed presentation for the group and the client if there's a client involved. You develop a detailed deck of slides that show all of the key findings. You might think the data speaks for itself, but it doesn't. It's up to you to provide your recommended solution that you believe that data supports. The strength of your recommendation should reflect the strength of your analysis. So yes, you want to have confidence in your recommendations. But if that's not what the data shows from your view, then giving recommendations with qualifications is completely appropriate. Maybe you're the decision maker. Maybe your supervisor will decide. Maybe it's the entire team. You want to help whoever is making the decision make a fair and fully informed choice. Don't overpromise or tell an unnecessarily one-sided story.
Here's a summary of the six steps McKinsey uses in their problem-solving process. Like I said, this is not a definitive list. It's one version that McKinsey has promoted through its public online podcast. It's a great starting place to understand how they approach these conversations and you can use it to lead your own teams or at minimum this will help you participate more effectively as a team member when your group is solving problems. And since this is the end, part four, and this wraps up our series on these McKinsey frameworks, then I'd like to know if you made it to the end, let me know in the comment section, and I'll be sure to give you a heart and say congratulations for making it all the way. And feel free to check out my website. There's a 100% free course about the top five communication skills that all professionals should have in addition to other free resources. Until next time, thanks.