Typically, the mission of strategy and innovation consultants is to grow a client’s business by doing the right things, smarter. Growing a business usually follows a rock-paper-scissor life cycle, i.e. moving from niche inception to market domination. Doing the right things smarter means excelling in complex problem solving, which requires the right mental models to combine a deep business, tech and human understanding as well as having the humility to acknowledge the so-called unknown unknowns. Part 1 of this post series explores why complex problem solving is quite difficult, while part 2 describes techniques to get better at it.
Solving for effectiveness vs solving for efficiency
Let’s first distinguish between solving for effectiveness (i.e. tackling complex problems; seeking validity) vs solving for efficiency (i.e. tackling complicated problems; seeking reliability). According to the World Economic Forum (WEF), the following top skills are needed to solve 21st-century business problems:
- Complex problem solving
- Critical thinking
The WEF defines complex problem solving as ‘The skill to see relationships between industries and craft creative solutions to problems that are yet to appear (…)’ and critical thinking as ‘(to) turn data into insightful interpretations (…) due to the complexity and interconnectedness of various fields like computer science, engineering and biology’. Hence, it’s not a surprise that there is a huge need for these kinds of skills, often summarised as complexity thinking, as many business challenges are becoming complex challenges due to faster cycles, globalisation, new technologies, automation and rapidly changing user loyalty. Solving for effectiveness thus builds on these skills and aims to produce outcomes that are valid (i.e. meet a desired objective). Collaborative and explorative models such as design thinking, job-to-be-done and lean startup are key approaches of this category.
In contrast, the aim of solving for efficiency is to reliably produce consistent, predictable outcomes. This type of problem solving prevails at most organisations today. In order to make things manageable, tried-and-tested theories and algorithms are applied in order to simplify and break problems down into smaller parts, which are easier to analyse and to solve by specialists further down the hierarchy (think: 3-year strategic plans, waterfall, specialised silos, CRM systems, performance marketing). This process strongly fosters reasoning by analogy, which means implementing pre-defined models for the situation at hand. If this activity sounds like a perfect use case for automation through AI, you’re right. It’s already happening that jobs relying on specialisation in a narrow range of routinised activities will be automated. It works best with predictable, linear problems and has been pretty well established by management consultancies, who have developed waterproof methodologies for solving any kind of organisational problem as lean as possible.
A global bias towards reliability over validity
Executives at large organisations solve problems for a living, and most do so by solving for efficiency, which is the result of organisational incentive structures and the way the human brain works. Anyone who has ever worked for a large corporation knows that speed, efficiency, swift execution and reliable, reproducible outcomes are highly rewarded. No one wants to hear about your 3-month innovation project without getting to know the exact deliverables of each sprint beforehand. Hence, there are strong incentives for managers to act confidently and to solve issues quickly due to time pressure and a predominant focus on productivity. Such organisational environments support jumping right into execution mode and implementing solutions as fast as possible. This often leads to dissatisfying outcomes, as no one stated and analysed the actual problem in the first place.
Furthermore, many executives are also inherently biased by their professional specialisation, the so-called ‘expert trap’. For a hammer everything looks like a nail and a CFO will always look at the world in a P/L-driven way (e.g. discounted cash flow analysis). Obviously, the resulting narrow mental framing leads to solutions no one ever asked for — they don’t fit the problem.
The same trap applies to siloed teams with deep expert knowledge. Too often, intricate execution plans are followed with the goal to deliver highly sophisticated solutions that are unfortunately already irrelevant when launched due to changing markets and missing real-world user input.
Next to this system-wide bias towards execution, the human brain is also hardwired to think fast. As proven by behavioural psychologist Daniel Kahneman, humans are really good at processing a tiny bit of data and weaving coherent patterns and stories around them. Our mind favours coherence over validity, so we constantly try to connect the dots with the data we have at hand to create a story and to come up with a solution that fits our story, not the actual problem. It’s the so-called system 1 (fast, intuitive) thinking mode, a term coined by Kahneman, that prevails most of the time. It is computationally less expensive and just more fun to come up with solutions right away than deeply analysing a problem. Again, this thinking mode fosters making surface level analogies: A manager draws on past experience and applies the same solutions over and over, failing to notice that most problems can have pretty different root causes. An often-cited example is ex-Apple Manager Ron Johnson’s failed retail turnaround strategy at JCPenney in 2011-2013. He confidently applied principles that worked well in Apple’s retail stores, but not at JCP, and thus destroyed over 50% of the company’s stock value.
The World Economic Forum, 02 July 2018: “10 skills you’ll need to survive the rise of automation”
Harvard Business Review, from the January-February 2011 Issue: “Reinvent Your Business Before It’s Too Late”
Harvard Business Review, from the January-February 2017 Issue: “Are You Solving the Right Problems?”