Chapter 4
Breaking Down Complex Problems with Logic Trees
Problem disaggregation is the heart of the seven-step process. Complex problems must be broken down into logical parts to understand drivers and causes. This critical step reveals potential solution pathways and helps determine what not to focus on. While the process may seem daunting, anyone can begin with literature reviews and brainstorming to create an initial logic tree structure.
Logic trees provide clear visual representations of problems, ensure comprehensive coverage of all relevant factors, and lead to testable hypotheses. They range from simple to highly complex but all begin as sketches on whiteboards or notepads, making intricate problems more manageable through visualization.
Different types of logic trees serve different purposes. Component trees break problems into constituent parts, while hypothesis trees organize specific testable statements. Deductive logic trees use top-down reasoning, arguing from general principles to specific conclusions. These are most useful when we understand a problem's logical structure, especially with mathematical frameworks.
Using a return on invested capital (ROIC) tree, Charles uncovered fundamental differences between Hechinger Hardware and Home Depot. Despite similar returns, Home Depot operated with 15-18% lower prices through superior inventory management, direct shipping from suppliers, and higher asset productivity. While Hechinger had spotless stores with extensive backrooms, Home Depot's warehouse model used 20-foot retail floor racks, eliminating restocking costs. Despite Hechinger's attempts to adapt, they failed to make the deep business model changes required and went out of business by 1997.
Inductive logic trees work in reverse of deductive trees, moving from specific observations toward general principles. We use them when we lack knowledge about general principles but have insights into specific cases. In practice, many problem solvers work both inductively and deductively, starting with some understanding of major drivers while also examining specific cases, iteratively refining the tree structure.
After creating comprehensive trees, good problem solving hinges as much on what you don't do as what you do. A useful 2x2 prioritization matrix weighs potential impact against ability to influence. In the Pacific salmon conservation case, climate impacts on ocean productivity were eliminated despite their significance because they couldn't be influenced, while hatcheries were initially deprioritized despite their impact because of political challenges.
Chapter 5
Building Effective Workplans and Team Processes
While some problems require only basic tools-a whiteboard, calculator, and internet connection-even straightforward projects can quickly become complex. Without proper planning, analyses waste time and effort. Effective problem solving means finding the critical path-continuing the prioritization process to focus on analyses with the highest probability of yielding insight.
The McKinsey approach addresses common workplanning issues with several key "twists": never conducting analysis without a hypothesis, visualizing the desired output form before starting ("dummying" the chart), ordering analyses logically so early work can eliminate unnecessary later steps, and being explicitly clear about who does what by when. This approach prevents wasted effort and ensures all analysis serves a clear purpose.
Rather than exhaustive project management documents, the authors advocate for "chunky workplans and lean project plans." They recommend short but highly specific workplans covering 2-3 weeks of critical analyses, coupled with rougher Gantt charts that track fixed milestone dates. This approach allows for constant revision as new insights emerge without getting bogged down in excessive detail.
"One-day answers" provide clarity by stating what you know about your problem at any point in the process. This approach helps establish emerging understandings and identify remaining unknowns, allowing teams to redirect resources to areas with the biggest gaps. Unlike typical approaches that gather enormous datasets before problem-solving begins, one-day answers follow a three-part structure: (1) a brief description of the initial situation, (2) observations or complications creating tension, and (3) the current best idea for resolution.
Effective problem-solving teams require specific structures and processes. While every team needs a coordinator (somewhere between a musical conductor and air traffic controller), the best teams have relatively flat structures during brainstorming and ideation. This can be challenging within hierarchical organizations, which is why temporary teams that operate outside normal reporting structures with limited lifespans often encourage greater creativity.
Successful problem-solving teams share three key behaviors: First, they're hypothesis-driven and end-product oriented, using strong hypotheses to guide work without becoming attached to them. Second, they "porpoise" frequently between hypothesis and data, remaining flexible and ready to pursue promising new lines of inquiry when others reach dead ends. Third, they seek breakthrough thinking rather than incremental improvements by examining performance limits and best practices from unexpected sources-like a mining company learning about faster tire changes from Formula 1 racing pits.
Chapter 6
Analytical Approaches from Simple to Sophisticated
Good problem solvers start with simple heuristics to understand directional relationships before employing complex models. Heuristics serve as powerful analytical shortcuts that help size different elements of a problem and determine efficient paths forward. Key heuristics include Occam's Razor (favor the simplest solution that fits the facts), order-of-magnitude analysis, the 80:20 Pareto Principle, compound growth calculations using the Rule of 72, S-curves for estimating adoption rates, expected value calculations, Bayesian thinking, reasoning by analogy, and break-even point analysis.
After using heuristics to understand problem scale and direction, deeper analysis can often be accomplished through straightforward questioning rather than extensive number crunching. The Sherlock Holmes approach of asking who, what, where, when, how, and why serves as a powerful root-cause analysis tool. Root cause analysis using the "5 Whys" technique helps identify the underlying sources of problems rather than just addressing symptoms. Beyond its origins in production settings, this technique applies to business problems like market-share loss and social issues like homelessness.
When simple heuristics and root cause thinking aren't sufficient for complex problems requiring robustly quantified solutions, more sophisticated analytical tools become necessary. Advanced analytical tools have become increasingly accessible with the rise in computing power and decline in data storage costs. However, despite greater accessibility, complex modeling tools are often used prematurely or inappropriately. First-cut data analysis to establish causality and impact size remains essential before employing complex models.
The choice of analytical tool depends primarily on whether you're trying to understand causation or predict outcomes. Understanding problems typically require statistical approaches like correlation and regression tools, while prediction problems may need forecasting models, machine learning algorithms, or game theory.
Machine learning techniques function similarly to traditional statistical tools but excel with large datasets and complex patterns. Unlike traditional regression where analysts select variables, machine learning algorithms identify the most predictive features themselves. This approach has proven valuable in diverse applications: predicting sleep apnea in children with 20% better accuracy than doctors using simple questionnaire data; optimizing Boston school bus routes to save $5 million; and detecting sharks at Australian beaches through drone surveillance with 90.4% precision.
Game theory provides powerful tools for analyzing situations where your actions influence and are influenced by competitors. The Australian government research organization CSIRO's WiFi patent case demonstrates sophisticated game theory application. After sending licensing requests to 28 companies with no positive responses, CSIRO made the strategic decision to pursue litigation despite facing formidable tech giants. Their approach was methodical: they first invested $1 million in preliminary expert advice, then committed $10 million to initial court action based on potential returns exceeding $100 million. Rather than confronting all infringers simultaneously, CSIRO strategically sued Buffalo Technology first-a smaller Japanese company with less US patent litigation experience. Their persistence paid off when HP settled first, followed by the other 13 companies for $205 million. A subsequent action against ATT, Verizon, and T-Mobile yielded an additional $220 million.
Chapter 7
Crafting Compelling Narratives from Analysis
The final steps in the problem-solving process-synthesizing findings and crafting compelling narratives-transform analytical work into actionable insights. When done well, this synthesis attracts support for causes, secures funding, builds family harmony, or advances careers. However, many projects fizzle at this stage due to unfinished models, excessive unfocused data, conflicting information, or team disagreements.
Effective synthesis begins with visualizing findings through graphics that highlight key insights. The most powerful approach arranges these visuals as branches on a revised tree structure, creating a comprehensive picture that reveals cross-cutting insights. Team synthesis sessions where members present their analyses to the group provide crucial checkpoints for testing hypotheses and often reveal breakthrough insights that weren't apparent in branch-level analysis.
As problem-solving progresses, reasoning evolves from early hypotheses to evidence-based action plans. Initial "one-day answers" using situation-observation-resolution structures help test hypotheses and focus data gathering. As analyses yield results, the focus shifts from understanding implications to marshaling evidence that drives specific actions.
After synthesizing findings into convincing graphics, the final step is structuring communication for your audience. The traditional pyramid structure-essentially a logic tree turned sideways-organizes arguments with the governing statement at the top, supported by evidence below. This structure can follow either deductive reasoning (from general principles to specific cases) or inductive reasoning (from specific observations to general conclusions), depending on the nature of your answer and audience needs.
When facing resistant stakeholders, consider using a revealed approach rather than the traditional pyramid structure. In one case involving a remote refinery with management resistant to change, consultants used a decision tree structure that gradually revealed competitor data layer by layer, helping the local team accept difficult conclusions about becoming a niche operator with substantial cost cuts. This approach walks decision makers through yes/no branches with supporting evidence at each step, revealing the answer rather than leading with it-particularly effective when recommendations contradict what audiences want to hear.
Chapter 8
Navigating Uncertainty and Wicked Problems
Problem solvers facing high uncertainty must systematically work through potential strategies, edging into risk while seeking opportunities to learn before acting precipitously. Uncertainty ranges from Level 1 (reasonably predictable futures addressable with good team processes) to Level 5 (unknown unknowns). Level 1 situations might include launching a product extension in a stable market, where historical data and customer research provide reliable guidance. Level 5 scenarios could involve emerging technologies like quantum computing, where even experts struggle to predict implications.
The middle levels require specific approaches: Level 2 involves alternative futures with binary outcomes, such as whether a new regulation will pass or fail. Level 3 encompasses a range of possible futures without clear probabilities, like entering an emerging market with multiple competitors. Level 4 represents true ambiguity where potential outcomes cannot be fully conceived, such as the early days of the internet or artificial intelligence development.
When facing uncertainty, several strategic approaches can be employed. Information buying helps clarify uncertainty sources through market research, pilot programs, or expert consultation, though perfect information may be prohibitively expensive. For instance, pharmaceutical companies invest heavily in clinical trials to reduce uncertainty about drug efficacy. Hedging involves reasonable-cost investments that counter downside events, like fossil fuel companies investing in renewables or agricultural businesses diversifying crop varieties to manage climate risk.
Low-cost strategic options provide positioning for future opportunities. IBM's failure to maintain its stakes in Intel and Microsoft (which would have been worth billions) demonstrates the cost of abandoning such options. Similarly, Amazon's early investment in AWS as a side project proved transformative. No-regrets moves involve capability building that will be valuable regardless of outcomes, such as developing digital capabilities or investing in employee training. Big bets are taken when you have confidence about an outcome others don't share, like Netflix's early pivot to streaming or Apple's iPhone development, though these require strong cash positions and low debt to execute successfully.
The "growth staircases" framework helps companies expand into new businesses amid uncertainty by emphasizing learning, buying options, and building capabilities. This approach analyzes how companies successfully build valuable new businesses by taking sequential steps to enter markets, hire specialized talent, and make acquisitions. For example, Amazon's expansion from books to general retail to cloud computing followed this pattern. The method involves working backward from desired outcomes, taking incremental strategic moves as uncertainty decreases, and framing the timeline for building necessary capabilities.
Even the most challenging "wicked problems" can be approached using the seven-step methodology. These complex issues typically involve multiple causes, stakeholder disagreements, unintended consequences, and require substantial behavior change. Examples include obesity (involving food industry practices, lifestyle changes, and healthcare systems), overfishing (requiring international cooperation and economic restructuring), terrorism (demanding security, social, and political solutions), environmental degradation, and poverty. Solutions often come from relaxing assumptions (like questioning traditional business models), revising incentives to internalize externalities (such as carbon pricing), and finding novel ways to cleave problems into manageable components (as demonstrated by successful microfinance initiatives addressing poverty).
Chapter 9
The Problem-Solving Mindset
Creative problem solving has become an essential skill for the 21st century-no longer limited to specialized professions but expected across all sectors. From software developers debugging code to healthcare administrators optimizing patient care to entrepreneurs launching startups, the ability to systematically tackle complex challenges is now fundamental. Good problem solvers are made, not born, and the seven-step process provides a systematic approach applicable to business, personal, and societal problems without requiring advanced mathematics.
Ten essential principles guide effective problem solving: (1) take time upfront to truly understand the problem - rushing to solutions often leads to solving the wrong problem entirely; (2) start with just a problem statement using simple tools like issue trees and basic frameworks; (3) try multiple approaches to structuring the problem, as different perspectives often reveal hidden insights; (4) leverage teams to enhance creativity and reduce bias through diverse viewpoints and experiences; (5) invest in proper workplanning with clear milestones and deliverables; (6) begin analysis with simple statistics and heuristics to identify patterns quickly; (7) employ sophisticated analytical tools when necessary, but avoid complexity for its own sake; (8) put equal effort into synthesis and storytelling to ensure insights drive action; (9) treat the seven-step process flexibly like an accordion, expanding or contracting based on the situation; and (10) approach any problem with confidence, knowing that systematic methods yield results.
The best problem solvers share an attitude of "active openness" to new ideas and data while maintaining healthy skepticism toward conventional answers. They regularly read across disciplines, from psychology to technology trends, building a rich mental library of concepts and analogies. These individuals are deeply reflective, constantly questioning their own assumptions and biases. They demonstrate remarkable persistence, viewing setbacks as learning opportunities rather than failures. Their team orientation allows them to both lead and follow effectively, knowing when to contribute and when to step back.
Successful problem solvers also cultivate specific habits that enhance their effectiveness. They keep detailed problem journals, document their thinking process, and actively seek feedback from peers. They practice regular reflection sessions to identify patterns in their problem-solving approaches and deliberately build their pattern recognition capabilities through varied experiences. With reasonable investment in systematic problem solving, anyone can find insights into almost any consequential problem-making bulletproof problem solving feel like a "superpower" in our increasingly complex world.
