Birdseye view of blue shoes standing and analysis a chart on the ground

Is your KPI Analysis biased?

Published: 

Blue makes me a poor analyst.

I like the colour blue. But just to be clear, I don't drench my entire existence with it. My house is not blue. My car is a strange off-white and as I am writing this, I'm wearing some kind of grey knitted thing, although my favourite team (coincidently) play in a lighter shade of the stuff.

I've clearly never lost any sleep trying to understand why it appeals to me. However, I am acutely aware of its affects on my choice of décor, or the products I buy.

But why does my deep-rooted appeal for this colour affect my ability to be a good analyst?

Our pre-programmed bias

What sets us apart as individuals (other than our near infinite physical discrepancies) is our likes, dislikes, opinions, loves, hates, beliefs, superstitions, political views and so on - most of which were forged by positive experiences or external influences, such as our parents or the media.

This cognitive bias helps us navigate the world around us more efficiently. It encourages us to draw on those experiences to guide us through the decisions and choices we make.

Although these 'guide ropes' provide very practical uses like choosing a 'blue' umbrella, or paying that little extra for a blue LED instead of white – yes done that. When it comes to making rational decisions, or analysing something with complete neutrality they are an impairment.

Analysing without bias – a rare gift

We rely on fair and unbiased judgements in many areas of society. None more so than the role of a Jury to provide such impartiality. Yet, Jury selection has become a science based on which individuals are 'more or less' likely to sympathise with a defendant. So, when something so fundamental to the principals of fairness is clearly hampered by our cognitive bias - how can anything be really analysed with complete neutrality?

The exception: Konrad Kellen

In 1965 Konrad Kellen - whilst working as an analyst for an American policy think tank, concluded that the Vietnam war was unwinnable. Something that flew in the face of nearly every US Administration Analyst. Their almost universal view was established mainly from transcripts of interviews with captured North Vietnamese soldiers. They claimed that their morale was so low that victory was inevitable. Kellen was able to analyse through the almost suffocating fog of biases at the time. He asserted that, even though American military action was relentless, the conditions unbearable and provisions scarce - they were still far from demoralized.

Sadly, Kellen's analysis was quickly dismissed as coming from a small group of discerning voices. The war continued for a further 5 years until its eventual, foreseen conclusion.

What set Kellen apart from his contemporaries, was his unique ability to listen without bias, and to analyse and interpret data without prejudice.

This ability is indeed a rare gift. It's similar to that of photographic memories, or being able to recall unusually long lists of data. But if like me, your one of the 99.99% mere mortals - how can we try to limit the effects of cognitive bias?

Free your mind and insights will follow

KPI Analytics is generally difficult at the best of times. The underlying data needs to be gathered, evaluated, considered, and an impartial conclusion derived. So, when a KPI indicates a sharp deviation we dive straight in, but here's a few things to help restrain our prejudice.

View the latest posts