Forecast Flaws

It was in 1997 that William Sherden exposed [the forecasters'] trade in his book The Fortune Sellers: The Big Business of Buying and Selling Predictions. His conclusion can be summarised in four brief sentences.
Private sector economists’ forecasts are no better than guesses.
Government economists’ forecasts are often much worse than that.
Increased sophistication does not improve accuracy.
There has been no improvement over the years.

As written in his recent Telegraph year review column from London by Dominic Lawson.

Pushing at my open door.

He used some delicious examples too.

From the augurs of Ancient Rome divining the future from bird behaviour, through Tolstoy on 1812 saying 'nothing was ready for the war which everyone expected', to our improving understanding of physics allowing for more accurate weather forecasts in the spirit of the Irish West Coast lady who spotted an incoming storm and so successfully delayed D-Day.

If only we could ever 'model' irrational humans the way of that last one.

Spoiler; we can't.

Enterprise Sales has more than its fair share of buyers bucking forecast.

Although the balance is not necessarily nice.

A bluebird may be a singular career highlight, whereas the sure thing failing to come home may seem perennial.

There is hope for us though. And it stops reliance on a hopecast, too.

Here's a trio of ideas for starters.

Abandon reporting on your ᴄʀᴍ 'gateways' passed through as main arbiter of deal likelihood.

These are too often tied to general deal signposts - longlist, shortlist, Board presentation - rather than meaningful components that place us as preferred.

Forget percentages.

I still see these. I cry inside. Actually, out loud.

You can be more binary.

But you must have embraced the pursuit of process refinement.

If that which would normally pretty much guarantee you gain the signature has not happened, then you cannot forecast it.

And if you want to be radical, how about following the root of world-leading Japanese mid-Twentieth Century manufacturing surge.

Get rid of forecasting your potential business.

For it was when (American) Professor Deming removed statistical process control from their production plants, did output improve.

There's a lot more around that, but for now, maybe that seed is sown.