I own this book and took these notes to further my own learning. Taking notes, publishing them and re-reading them allow me to flatten my forgetting curve. If you enjoy these notes, I highly encourage you to do the same, buy this book here and take your own notes.
INTRODUCTION
How to solve big problems and sell solutions like top strategy consultants. A practical introduction to problem-solving tools and techniques
CORE
- The most important skill you never learned
- Thinking Fast and Slow: humans have 2 minds, system 1 thinking is largely involuntary, automatic and unconscious vs system 2 thinking slow, voluntary, it requires effortful attention and conscious deliberation.
- Knowing this allows us to not jump too fast to a solution but to question sufficiently the information we have, search for more and finally overcome our assumptions to generate better solutions
- Originals, Adam Grant, people are always looking to transform information into written stories. Even in noise, people are looking for signals and plausible story
- Shakespeare, Othello: A handkerchief leads Othello to kill Desdemona after thinking this was hard evidence his wife cheated on him. Hamlet: Hamlet wants to be sure that Claudius is guilty but his endless hesitation paralyzed him and led him to make terrible mistakes. To think and to act too fast vs. to get mired in analysis paralysis and to be stuck in a loop of endless questioning that is the question.
- Experts are better problem solvers than novices thanks to their experience, but be aware of the expertise trap and how it could dig people to view the world in a very narrow way. An analogy can also lead experts to develop the wrong solution in front of unfamiliar situations.
- Experts are more likely to fall into the trap of facing “unknowns-unknowns”, “things that we don't know that we don't know” (2002, US defense secretary Donald Rumsfeld)
- The 5 pitfalls of problem-solving
- Pitfall 1: Flawed problem-solving definition.
- Example, Sony against piracy for the music industry
- Pitfall 2: Solution Confirmation
- Rather than beginning with the problem of child nutrition and analyze the solution, Danone started with a solution they already had
- Pitfall 3: Wrong framework
- It's tempting, if the only tool you have is a hammer, to treat everything as it were nails.” “The law of the instrument” Abraham Kaplan and Maslow's hammer. Example: using machine learning for every problem
- Pitfall 4: Narrow Framing
- In front of long and complex problems, it could be tempting to frame the problem narrowly to something we may understand and come up rapidly to the solution. (better approach: design thinking path)
- Pitfall 5: Miscommunication
- “Fat vs. Sugar”, Yudkin was a real scientific but never was a good communicant, no wonder why the theory of Ancel keys, very good communicator, won and for years only the fat was accused of being harmful. Only years later, thanks to Lustig's masterpiece presentation (video) we got better learning on how harmful was sugar
- Pitfall 1: Flawed problem-solving definition.
- The 4S method
- State - Structure - Solve - Sell
- MECE problem decomposition
- Hypothesis-driven problem solving, Highly efficient method: formulate a hypothesis and then test them, but avoid exploring all the possible (“boiling the ocean”)
- Because of confirmation bias, we may be tempted to seek evidence that confirms our hypotheses. Francis Bacon “Man prefers to believe what he prefers to be true”. William James “a great many people think they are thinking when they are merely rearranging their prejudice”
- Consultants are not immune to this bias but they have developed safeguards: 1) being an outsider make them neutral, 2) working in team allow to challenge each other ideas, 3) experience in problem-solving method
- Design thinking is a powerful problem-solving tool that ally both creativity and analytical approaches. Today design thinking is part of some companies’ culture (IBM and intuit). Design thinkers suppress their assumptions about the problem but use insights generated from observations of users to develop hypotheses about solutions. These hypotheses will then be tested in the form of prototypes to converge toward the best fitting solution.
- 4S CHART
- A problem well-posed is half solved: “If I had 1 hour to solve a problem, I'd spend 55 min thinking about the problem and 5 min thinking about the solutions” - Einstein. The first question is not “what is the problem?” but “do I know enough to state the problem?". Stating the problem requires empathy.
- Structure the problem: hypothesis pyramid (highly confident in the solution), issue tree (no strong hypothesis but problem easily decomposable), with ideation
- Solve stage: 1) hypothesis-driven: test a hypothesis with hard facts to back it up, 2) issue-driven: using issue tree that breaks down the problem into parts to be separately analyzed (disaggregating a problem does not mean a solution will emerge), 3) design-driven: easier to get fresh ideas
- “The mission of design thinking is to translate observations into insights and insights into products and services that will improve lives” - Change by Design, Tim Brown
- Limit of Hypothesis-driven problem solving: confirmation bias > limit of problem disaggregating > design thinking
- Selling the solution: focusing on the answer and your audience
- State the problem: TOSCA Framework
- Trouble (what makes the problem real and present), Owner (whose problem is this), success criteria (what will success look like and when?), Constraints (what are the limits in the solution space), Actors (who as a say in the way we solve the problem and what do they want?)
- Trouble is a gap bwn an aspiration and the current situation. Ask your self “why now?".
- Success criteria: ask why (5)
- Write the problem, write the core question (no statement)
- A good problem definition or question: does the question address the trouble that got you to consider the problem in the first place? Is the question phrased from the perspective of the owner? Would the answer meet success criteria? Does this recognize the constraints and relevant actors?
- Consider immersing yourself in the actors’ shoes to state the problem
- Structure the problem Pyramids and Trees
- The hypothesis-driven approach
- Starts with a candidate solution to your core question, try to confirm or disconfirm. The hypothesis pyramid is built with a candidate solution on top, sub-hypothesis to confirm the hypothesis, sub-sub-hypothesis…
- The hypothesis Pyramids are built with necessary and sufficient conditions. Having one sufficient condition right is generally not enough to validate the leading hypothesis but having one necessary condition wrong is enough to reject it.
- Each level of your pyramid should be MECE
- Pros: fast, easy to sell, pattern recognition, good if strong hypothesis and expertise
- Cons: 5pitfalls mentioned above, logical challenge (necessary vs sufficient, cause vs consequence, correlation vs causal)
- The issue-driven approach
- Starts with the core questions and disaggregate it into questions, not hypotheses (MECE + 80-20)
- Pros: most accurate method, generate insights
- Cons: not always possible, time, not always providing answers
- The hypothesis-driven approach
- Structure the problem: analytical Frameworks
- Use a framework to break down the problem but don't rely on too much of them, frameworks are just tools among others. Don't be a hammer in search of nails
- Use multiple frameworks and combine them
- Use industry frameworks, formulas framework and use logical decomposition to break down problems
- Your issue trees and hypothesis pyramid will integrate multiple frameworks
- Solve the problem: 8 degrees of analysis
- Picking the right data, picking the right time frame of time series
- Benchmark your assumptions
- Biased sample, selection bias
- Test sensitivities: Would your conclusion still hold if your key input was wrong by 20%, 50%…?
- Be aware of confirmation bias ( - have peers challenging your analysis)
- When a correlation is statistically significant, it does not mean we can observe a causality. It just means we can be confident (in the level specified) that the correlation actually exists in reality
- Redefine the problem: the design thinking path
- Doug Dietz, ted talk, designing medical scanners
- The principal advantage of the design thinking past is that rephrases the problem statements and open-up new opportunities
- When to use a design thinking approach? The problem is human-centered, the problem is complex, the cause of the problem is unknown, the problem cannot be stated clearly.
- Empathize > Define > Ideate > prototype > Test
- Always begin the process with fresh eyes and learn from users as if you were a beginner
- See the problem from the perspective of the users, the user is the problem owner. Shadowing / immersion / interviews… Consider also to look at extreme users to not fall into the confirmation bias trap
- Define the problem from the perspective of the user. Define activities phase with: Empathy maps, journey maps, insight cards, how might we….
- Structure and Solve the problem using design thinking
- “the best way to have good ideas is to have lots of ideas and throw away the bad ones”
- Generate as many concepts as possible: diversify the team, defer judgment (Psychological Safety), look for quantity, be visual, be focused on the problem.
- Ideation tools: Analogical thinking (model T assembly chain was inspired by the Chicago butchering slaughterhouse), brainstorm, brainwriting, morphological analysis, SCAMPER,
- Select concepts: create a selection matrix and rank the concept
- Prototype to communicate, to manage risk, to learn through a visual,
- Test and iterate
- Sell the solution: Core message and storyline
- Tell the story of the solution, not the story of the research
- Truman: “give me a one-handed economist… all my economists say on the one hand… on the other hand”
- Pyramid principle: Core message, key lines that support the core message, supports
- Grouping or argument: easy to sell vs hard, simple message vs complex message, audience expect the answer vs going for an argument
- Grouping: core message with bullets points that support the core idea. Argument: core message supported by a logical sequence, each is connected to the prior one and lead to a conclusion
- Never forget that the goal of selling a solution is to trigger action - SO WHAT?
- Sell the solution: Recommendation Report and Delivery
- Schedule intermediary checkpoint with the problem owners to reconfirm problem alignment, to share intermediary finding and hear first objections
- PWP: Exec summary, key line A, message A.1, message A.2, message A.3, key line B, message B.1, …. Background page
- Add trackers to your presentation
- Data visualization is to know what you want to see jump off the page
- Waterfall chart
- Antoine de Saint-Exupery: “perfection is attained not when there is nothing more to add, but there is nothing more to remove”
- Let the audience absorb the content before the presentation (amazon narrative)
- Engage the audience by telling a story first
- The 4S method in Action
- Conclusion: becoming a master of problem-solving