Assuming this is true…

Home / blog / Assuming this is true…
Facebooktwitterlinkedin

We make assumptions every day in almost every walk of life. We make assumptions about ourselves, about others, about situations. We make assumptions in business and in science.

It is part of human nature; it’s a way we make sense of the world around us and how we build our mental model of the way we believe things are and the way we believe things should be in the world.

One of the key ways we learn is by association and therefore we aggregate things that look similar.  By way of example, there are over 50 different makes of motor vehicles on the market today in Australia, they all have a similar recognisable form so we lump them together and call them cars.

However, if something is unfamiliar but it looks like something similar we have seen before we make the assumption that it’s the same, for example we choose our tomatoes from the vegetable section, they are not sweet, and so by association and conditioning we assume they are vegetables when in fact they are fruit.  If we didn’t learn by association we would have to learn and relearn every day.

In business, people make assumptions every day for the same reason but the stakes are higher and many of the critical assumptions made go untested. I suspect that one reason why upwards of 80% of business start-ups fail is that many untested assumptions are being made.

So it raises the question: are there ways to test assumptions? Yes there are.  Although I’m sure there are numerous ways, what I am going to share is one that I came across very recently which, while it is aimed at businesses, I believe it can apply in other walks of life.

Jon Fjeld writing in the MIT Sloan Management Review[1] offers us a way worthy of consideration, his formula is:

Severity x Probability the assumption is false
Cost of resolution

 

He describes severity as the impact on what you are planning to do if the assumption proves to be false.  Severity can be difficult to quantify as it is itself open interpretation and quite possibly assumption so it requires considerable discussion and thought. The probability of the assumption being untrue focuses us on two factors in the decision, time and money, because as Fjeld contends if you’re to minimise expenditure of both then the assumptions that are least likely to be true should be tested first.

The final area, the cost of resolution, is, according to Fjeld, the core question to be answered which is to define what constitutes a resolution.  It will differ for different situations. When there is clarity around this third area a ranking can be created by judging the severity of the unknown and multiplying it by the probability of the assumption being untrue and dividing the answer by the estimated cost expressed in money and time to arrive at a ratio for each assumption you want to test.  The ranking allows a path to be mapped to address the greatest risk for the lowest cost.

This solution probably won’t to appeal to everyone but then again I am making an assumption.  Still and all, I think it is worthy of consideration because we inherently want to make sense of things.

 

 Viventé Australia enables leaders and managers develop the bond between people and performance, creating a powerful advantage: the synergy between people, leadership, management, and culture, produces performance that allows your business to achieve its best. 

[1] https://sloanreview.mit.edu/article/how-to-test-your-assumptions/

Facebooktwitterredditlinkedin

RECENT POSTS