Estimation accuracy is a skill: here is how you train it

One of the first things many of us do in the morning is stepping on the scales. You know, just to get the reassurance of how beautiful we are (ha ha, very funny!). I knew “tricks” for all of the scales I had. Because there are mornings, when the last thing you want is the harsh truth on the small LED screen under your feet. On those mornings, I knew that all I had to do was either lean forward or stand on my toes, or lean on something slightly using my hands.  I know, I know, I know… not fair, not accurate, “not” many other things. But what if I feel like it? And I am pretty sure many of you recognize what I am talking about. It’s called bias. Whether deliberate or unconscious – it is still bias. It will impact all of our estimations.

One thing is deliberately choosing to see the “lighter” version of the truth on the scales every morning. Another – making business decisions on inaccurate data. And I am not talking only about multi-million projects. Everything starts with small every day habits. Some examples:

  • Time planning. One option – being always late (arriving later than promised); second option – arriving always too early.
  • Budget planning. Making assumptions on spend, sales, growth, etc.
  • Selecting savings opportunities to be pursued – prioritising initiatives and projects.
  • Negotiation planning: estimating your own and your negotiation partner’s positions.

Making any assumptions or estimations at all means exposure to a human bias. I knew people have the tendency for overestimating or underestimating things, but it never occurred to me, that you can:

  1. Measure it.
  2. Calibrate it.

Yup, that is right. The rule “if there’s a will, there’s a way” works here, too. It is called confidence calibration. It is a skill and you can train and improve it. You can find some more detailed information here and here. The vast information pool on the internet offered a tool which I tried out on myself. Turns out, I am quite on spot if I know the subject and I am being rather “safe” (underestimating, leaving some space “just in case”), if I am uncertain.

This has helped me to increase the accuracy of my estimations. Furthermore, the old scales broke down and I am still working on finding ways to befriend the new ones. Would love to receive your feedback – if you try using the confidence calibration tool, that is. Oh, and any scale “calibration” tips would be highly appreciated.

Estimation accuracy is a skill: here is how you train it