The Seven-Step Problem Solving Process

This post summarises the seven-step problem solving process from this article, which is basically a transcript of a McKinsey Podcast about problem-solving strategies hosted by Simon London, and featuring guests Charles Conn, CEO of venture capital firm Oxford Sciences Innovation, and McKinsey Senior Partner Hugo Sarrazin.

Summary: How do the McKinsey folk define problem solving?

1. Problem definition. It is important to take a step back and ask the questions, “What are we trying to solve? What are the constraints that exist? What are the dependencies?” There’s some similarities here with Problem Restatement techniques which I shared briefly in a previous article. The key is really breaking down the problem’s context. One useful example of this via the problem statement, “Can we grow in Japan?” Some other spin-off questions to dig deep into the context include, “What, specifically, are we trying to uncover in the growth of a product in Japan? Or a segment or channel in Japan?” Other questions to consider are: “What are the forces acting upon your decision maker? How quickly is the answer needed? With what precision is the answer needed? Are there areas that are off limits or areas where we would particularly like to find our solution? Is the decision maker open to exploring other areas?”

2. Disaggregate the problem. In other words, to take the problem apart into logical pieces. Conn introduces the idea of using logic trees to help break down the problem further. For example, a profit tree breaks down the components of revenue (price & quantity) and cost (cost & quantity). This provides some clarity on the business in question and enables analysts to organise problem solving efforts.

Simple profit tree (source:

3. Prioritisation. To focus efforts, we need to identify and address only the high impact levers in a problem. We ask the questions, “How important is this lever or branch of the tree in the overall outcome that we seek to achieve? How much can I improve that lever?” It is also important to focus on just the levers that are movable and can be changed. Conn used the example of the Moore foundation wanting to save Pacific salmon. Ocean conditions were a big lever, but also very difficult to adjust. Instead, Conn’s team shifted their attention to fish habitats and fish-harvesting practices, which were relatively easier to tackle.

4. Work plan. The next is to draw up a plan to address the problem with (somewhat like an ‘answer’ to the problem). The key here is to craft a plan that reflects the level of precision, the time frame you have, and the stakeholders needed for the exercise. Depending on these factors, a problem can be answered in “one hour or one day”. If the stakes are high, the model might need to undergo a more rigorous validation. However, a work plan should not stretch too long (Conn gives an illustration of a 50-page work plan over three months being over-the-top long) as plans can get outmoded quickly. Plans are also a good stage to deal with biases. It is important for a team to have diversity to avoid groupthink. Availability bias (i.e. thinking that we’ve seen a problem before and matching it with our previous conception of it) and sunflower bias (i.e. where people alter their approaches after hearing more senior people speak) are also critical to watch out for.

5. Analysis. Before applying advanced tools using computation and analytics, it is important to start with simpler heuristics and explanatory statistics. One example is A/B testing, which is a simple test-and-learn feedback loop that is repeated till satisfactory and its best for contexts for maximising learning. While modern-day tools and technology can bring new insights to problem-solving, they are no substitute for sharp problem-solving skills. Good problem definition, as well as an understanding of current algorithms and models, are a must before progressing to big data sets and unknown algorithms.

6 and 7. Synthesis and storytelling. Conn takes the final two steps together as they go hand-in-hand. One must avoid the pitfall of stopping at data and analysis, and progress to synthesising the pieces from there and weaving them into a story that helps the decision maker answer the question: “What should I do?” I like these quotes from the interview:

…Until you motivate people to action, you actually haven’t solved anything.

Simon London

…decision-making theory, without a bias to action, is a waste of time. Everything in how I approach this is to help people take action that makes the world better.

Charles Conn

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create your website at
Get started
%d bloggers like this: