Why Forewarned is Forearmed

“Forewarned is forearmed; to be prepared is half the victory” : Don Quixote by Cervantes

I’ve just been given a book. A nice book about mastering business forecasting. Having listened to one of the co-authors, Steve Morlidge, talk about the subject, i’m looking forward to reading it.

Amongst a gaggle (or is it a google) of IT and Finance professionals taking time out of their busy days to gain valuable insights I sat down to listen to Steve’s talk at the IBM Finance Forum, entitled “Future Ready : Mastering Business Forecasting” (which funnily enough is the title of the book as well).

IBM brought together key note speakers and subject matter experts to explore business challenges, and shed light on the impact on Finance departments of better working methods for Business Intelligence.

Steve’s views and presentation were a very illuminating and insightful about the reasons for forecasting, and the impact on business.

And I’ve got to say he didn’t say anything anyone in the audience didn’t already know.

Yes, it seems a strange thing to say. The fact is, he only spoke about things that were almost blindingly obvious, but unfortunately many businesses simply don’t do the things they should.

Rather than go into the nitty-gritty of technicalities he concentrated on the underlying reasoning behind his proposed methodology, taking a number of basic, common sense ideas and joining them all together. His examples were excellent, ranging from the mundane – driving a car round a bend – to the erudite- Plato’s analogy of government being like seafaring for example – and all served their purpose well.

A good example of the common sense he spoke about was his five characteristics of a good financial forecast:

• Timely – what’s the point of having a weekly forecast that takes a week to prepare?
• Actionable – you have to say what the information can be used for, and make sure that’s possible.
• Reliable – it has to be accurate enough for the forecast.
• Aligned – it has to be aligned with other stakeholders.
• Cost Effective – what’s the point of it costing more than it saves?
• All very obvious, but someone has to say it. TARAC if you want a mnemonic.

One thing he spent time on was reliability, explaining the effects of Bias and Variation on forecasts. I’ll come back to these in a more detailed blog soon, but for now...

Bias - the name he gave to systematic errors in the process, either positive or negative. For example, a sales force that habitually understates their targets to appear to exceed them introduces bias that invalidates the forecast. As does the Executive who wants the sales targets to be higher than the forecast says. (Again, common sense when you think about it.) Bias is unacceptable.

Variation - his term for unsystematic errors. If you have an unforeseen event, then, as long as it’s not a recurring one, it’s not bias. Your forecast should allow for this, and you can build in an acceptable level of variation.

For these he used a very informative slide of a golfer, who was

aiming at the flag (the target)
• happy to get the ball anywhere on the green (variation)
• wanted to avoid slice or hook (bias)

Anyway, all bias should be cut out of the forecasting process, but some variation is allowed. Like the golfer, he’s happy enough to get the ball on the green. A professional will have a narrower field of variation, with only getting the ball in the hole or within say four feet being acceptable.

Neither golfer will want to slice the ball.

The quote at the top is from “Don Quixote”. It’s on the back of the book Steve Morlidge wrote about forecasting. And I understand why he’s used it.

If you don’t want your business forecasting to become some quixotic tilting at windmills, you must allow for acceptable variations, because they will happen. Conversely, it’s imperative to get rid of your internally created errors. Do these and your forecasting will be much more valid.

Julian Shaw