5 clues for solving complex problems
Reading some Sherlock Holmes stories recently reminded me of the process we go through to find solutions to complex business problems. Indeed, there are 5 helpful clues we can learn from.
Sherlock Holmes is a famous fictional, Victorian age, consulting Detective recently modernised by Benedict Cumberbatch. The stories range from complex murder mysteries like “the sign of the 4”, to the banal “blue carbuncle” where detailed inspection of a lost hat leads to the identity of the thief who stole a precious jewel. The character is famous for attention to detail (at a level way beyond where most people look), carefully assessing the facts in front of him to test for completeness and accuracy, using sound logic to create inferences and deductions, before solving cases consistent with the facts. Indeed his method is summarised in the quote “When everything that is impossible has been eliminated, what remains, however improbable, is the truth”.
I often encounter complex problems in my experience as a Digital transformer, Engineer and Business change leader. Some take many weeks/months and large teams to resolve, some less so. Often the challenges are paradoxical eg how can we increase productivity and reduce cost? Many are simply complex e.g. how can we change to a digital operating model whilst growing services volume and maintaining tight delivery milestones (eg in product development, telecoms and finance)? And some have high levels of risk e.g. taking new technology to market with fixed budgets and timescales without compromise to human safety (eg for hospitals and police).
So using Sherlock Holmes as inspiration, lets look for 5 clues to help us solve complex problems:
- Divide and conquer: Large problems are often comprised of lots of smaller problems. There are lots of ways to break something down e.g. analysis based on impact/urgency, cause and effect, sensitivity, time/cost/quality constraints, abstraction or hierarchical breakdown. Even these elements may be complex and interdependent so a meticulous approach is needed. Further, the know/unknown analysis often quoted to Donald Rumsfeld, ex-US Secretary of Defense, can be invaluable: we might know what we know and what we don’t know, but how do we find out what we don’t know we don’t even know?! Above all this clue is all about understanding the problem, characterising it, finding out what needs to be better understood and defining it carefully. Solving the wrong problem is too often a problem!
- Collect Good Data: From good data, insight and information can be drawn. This adds to our knowledge ultimately contributing to wisdom on a subject. This is the basis of good science – but in projects and business good facts are often abstract and scarce, but undeniably still crucial. For example on endeavors that are significantly off track, often the hardest question to answer is “where are we?”. Digital developments and large organisation changes can often appear to be more progressed than they actually are. What is needed are some hard facts: What specifically has been achieved? How much in each area specifically has been spent? What value are customers receiving? How many people are actually booking time?Systematically collecting data is key but there are some health warnings: “You can’t control what you can’t measure” is an Engineering maxim that applies equally in business. However, this needs to be balanced by the flip side that “you affect what you measure”. How often have you observed unbalanced goals drive the wrong behaviours? (eg sales measured just on revenue diminishes profit). It takes time and sound orgisation to gather good facts, especially where the source is not naturally enumerated, but the effort will save time in the long run.
- Analyse & Suspend Judgement Once some of the facts start to appear, especially in pressured situations, there is a strong desire to jump to conclusions. This clue is a reminder that cool heads are needed to carefully assess that all dimensions have been considered and sound data collected. Inevitably there will be gaps and for each gap an appropriate assumption will need to be made. Good assumptions need good insight – even a “finger in the air” guess can be based on something known. Further, each assumption also presents a risk that may need appropriate management.Knowing when to stop analysis should not just be a matter of time’s up. A good complex problem solver needs to be brave and hold true to the facts using insight, teamwork and experience to judge when/how to progress. This is where experience really counts.
- Creatively Solve: We often hear “I can’t see the wood for the trees” or “take the helicopter view” these phrases characterise that finding new perspectives to a challenge can unlock creative solutions. Mindmaps and hierarchy models are a powerful graphical means to capture the extent of the challenge faced and break it down, systematically into smaller pieces. Being free to capture elements and use techniques to group, assess inter-relationships and summarise often helps find ways to new ways to see a problem (eg stickies on a wall). Sometimes the solution lies around the corner – but how can we gain some insight to see around bends?Engaging all parts of the brain is key to this clue – stand up & walk around, use colours, post-its, cards and facilitated “brainstorming” all play their part. If the problem is very large and has been broken into several parts the solving may take some time. Teamwork, as ever, is key to good outcomes.
- Rigorously review: So the team have creatively come up with some solution options after mapping out the problem, collecting data, suspended judgment in the analysis. But how well do the solutions meet the defined the attributes of the problem? Perhaps no solutions meet the criteria – so what are the tradeoffs and who is best placed to choose? Complex problem solving is not necessarily a deterministic or linear process – things change. The underpinning data and assumptions set may need regular re-evaluation to check that the selection solution path is still optimal.