The Squawk Point

Organisational Mechanics

  • Home
  • Blog
    • People
    • Data
    • Process
    • Wild Cards
    • Index
  • Podcast
  • Book

Better to Be Approximately Right Than Precisely Wrong

12 July, 2018 by James Lawther 4 Comments

Spurious accuracy

Thirty years ago I sat in a factory office with my head bowed low, looking at the dirty grey lino floor.  My boss, a middle-aged, overweight man wearing a white coat smeared with ice-cream was berating me in a broad west country accent.

Apparently I was wasting both my time and his.  I would be far better off if I stopped faffing about with my computer and did some proper work (my words not his).

This was the late 80’s, my boss wasn’t big into computers.

The spreadsheet

I had just taken possession of my first copy of Lotus 1-2-3.  Some of you may remember it, it was the predecessor to the ubiquitous Microsoft Excel.

I was busy working out how much ice-cream we could get on a lorry.  I’d worked through the calculations:

  • Weight of an ice-cream
  • Ice-creams per carton
  • Cartons per pallet
  • Pallets per lorry

So far so good…

Then I’d refined and improved my calculations adding:

  • Volume of air in each ice-cream wrapper
  • Cardboard thickness
  • Carton filling pattern (how many ice-creams to a carton)
  • Carton stacking pattern (how many cartons to a layer)
  • Layers per pallet
  • Thickness of interleaves (sheets of cardboard that hold pallets together)
  • Number of interleaves
  • Weight of interleaves
  • Pallet shrink-wrap weight (think cling-film)
  • Shrink wrap length

Then I’d created a whole host of combinations of carton filling and stacking patterns.  I was very proud of myself.  I’d worked out how to squeeze 1,032 extra choc-ices on a truck.

All calculated to three decimal places.

Precision is not accuracy

By boss took one look at the spreadsheet and rolled his eyes towards the ceiling. He pointed out that for all my variables, data and fancy calculations, the only thing we could be sure of was that my final number was flat-out wrong.

He gave me a thermal coat and sent me and my mate into the cold-store (minus thirty-two degrees centigrade, bloody chilly). We spent the next two hours stacking pallets then went back to tell him what we’d found out.

For some reason he didn’t fancy coming into the cold store to have a look himself.

His point was simple

Pulling together a spreadsheet and running some numbers is all important. But…

  • I didn’t know what information I needed.  How about the thermal expansion coefficient of chocolate?
  • If I did know, then I couldn’t collect it all without months of research.
  • I was wasting my time searching for incremental data.
  • I was giving myself a false sense of security. Thinking I knew everything, when I didn’t.

Worst of all I was wasting time not coming to a conclusion.

Analysis paralysis

As he put it.  My obsession with accuracy was stopping progress (that’s not what he said, his terminology was cruder).  Cutting hairs over and over again was not going to get me a more accurate outcome.  Nor was it going to get any more ice-creams on the lorry.

Analysis by itself is a pointless exercise

It doesn’t take you anywhere unless you act on it.  Even then it isn’t enough to act, you have to learn from your actions.

A few freezing hours later we had come to a much more workable solution.  It transpired I’d forgotten to include the weight of the pallet…

I went away to refine (and simplify) my spreadsheet and persuade the factory manager to change the stacking pattern.

It is better to be approximately right than precisely wrong

If you are analysing something, first create a simple, straight forward spreadsheet that gives you some insight into what is going on.  Then test your assumptions and learn from them.

Running endless spreadsheet scenarios just consumes time.  After all that work, the number you end up with will be precisely wrong.

The greatest works of fiction are written on spreadsheets ~ Anon

If you enjoyed this post click here to receive the next

Ice-cream van

Read another opinion

Image by Beverley Goodwin

P.S.  I prefer Microsoft Excel to Lotus 1-2-3.  It has so many more options for making your spreadsheet look pretty.  You can even colour code it to match it to your PowerPoint. After all, if you have nothing insightful to say, it is always best to make it look impressive…

Filed Under: Blog, Operations Analysis Tagged With: analysis paralysis, assumptions, over-processing, spurious accuracy, statistics, test and learn

About the Author

James Lawther
James Lawther

James Lawther is a middle-aged, middle manager.

To reach this highly elevated position he has worked in numerous industries, from supermarket retailing to tax collecting.  He has had several operational roles, including running the night shift in a frozen pea packing factory and carrying out operational research for a credit card company.

As you can see from his C.V. he has either a wealth of experience or is incapable of holding down a job.  If the latter is true this post isn’t worth a minute of your attention.

Unfortunately, the only way to find out is to read it and decide for yourself.

www.squawkpoint.com/

Comments

  1. Bob Spencer says

    13 July, 2018 at 10:09 am

    Lotus 123… you always were posh. Some of us started out on Supercalc… without a hard drive. Twin Floppies was the future!

    Reply
    • James Lawther says

      13 July, 2018 at 6:16 pm

      Not Posh Bob, just young :)

      Reply
  2. Steve Brett says

    13 July, 2018 at 6:10 pm

    Absolutely – what is the decision we need to make and what is the minimum we need to know to decide which way? Oh – and remember it might be easier to try it before doing any analysis!

    Reply
    • James Lawther says

      13 July, 2018 at 6:16 pm

      Nice and succinct. Rather than what data can we lay our hands on and what does it tell us…

      Reply

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Explore

accountability assumptions beliefs best practice blame bureaucracy capability clarity command and control communication complexity continuous improvement cost saving culture customer focus data is not information decisions employee performance measures empowerment error proofing fessing up gemba human nature incentives information technology innovation key performance indicators learning management style measurement motivation performance management poor service process control purpose reinforcing behaviour service design silo management systems thinking targets teamwork test and learn trust video waste

Receive Posts by e-Mail

Get the next post delivered straight to your inbox

Creative Commons

This information from The Squawk Point is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
Creative Commons Licence
Customer Experience Update

Try This:

  • Fish Bone Diagrams – Helpful or Not?

  • Regression to The Mean

  • Should You Punish Mistakes?

  • Brilliance Alone Won’t Take You Far

Connect

  • E-mail
  • LinkedIn
  • RSS
  • YouTube
  • Cookies
  • Contact Me

Copyright © 2025 · Enterprise Pro on Genesis Framework · WordPress · Log in