The Lab: Issue № 2

Outlining a technique for approaching complex or novel problems

Most problems we encounter resemble those which already exist. Typical solutions to these challenges apply established frameworks in a reliable way. Complex or novel problems—those without standard methodologies—call for an alternative approach. How should one approach this complexity, when starting points are unclear? I have found a lot of success embracing a counterintuitive idea: ‘do it poorly’.

By ‘doing it poorly’, progress begins on something, even if it fails to yield a solution. This concept translates to the physical world; if you’re unsure how to get fit, go for a walk. Forget how far or how fast—these are of secondary importance when beginning. This first walk gives you a performance benchmark and key metrics to build upon in the future. An attempt, even a poor one, identifies stumbling blocks and informs next steps.

Even in subjective or open-ended work, this approach is still effective. By refining the first attempt, the author engages in a form of ‘error reduction’. This bears strong resemblance to the operation of some machine learning models which progressively remove Gaussian noise to realise a sample of interest, such as an image. Compared to the typical notion of creation—one with a focus on adding to a blank page—this technique subtracts the incorrect from an initial solution. This original work is a best guess; an honest response to a question, unburdened by issues of correctness or style. It is from this courageous first shrug that more substantive work can flow. Without it the process has not truly begun.

Undertaking ambitious efforts with the mindset that failure is not only accepted but encouraged is a surprisingly useful heuristic. The power of such incremental progress cannot be understated, particularly in long-term endeavours. Starting with a clear goal and the most basic of initial attempts, repeated cycles of change and reflection can lead to extraordinary results.

While comparable to the adage of how to ‘eat an elephant’ (one bite at a time), this method seeks to combat more than the scale of a problem. It is also about taking control of an unwieldy issue which requires novel outcomes. One which requires legitimate problem-solving under conditions of uncertainty. It combats the tendency to replace difficult, uneasy progress with the safety and certainty of planning and strategy. Without a poor start, it becomes easy to never begin at all. Though the final product may not contain any of the elements of the initial attempt, the value of such a start lies in edging towards perfection.