What your forecaster can't tell you

Forecasting is a tricky business. You're trying to pluck a number out of the ether and way too many things can go wrong.

You never have enough information, enough accuracy in measurements, the correct weighting for all the variables and always too much time since the data were gathered.

And in the auto industry, forecasts that determine how many thousand $30,000 products to produce, parts to order or factories to build certainly raise the stakes.

But forecasters sneer at such challenges. They love multiple variables, thrive on refining measurements and extrapolate everything else. No obstacle is too big for a forecaster.

Except one.

Client expectations. By client, I mean who commissioned the forecast, which can include bosses. You can't deliver a forecast your client isn't prepared to believe.

Deliver a forecast so high or low that your client doesn't believe it, then you lose credibility at least, and often future business.

In the forecasting biz, sometimes you're only as good as your client lets you be.

(Full disclosure here. I worked for two automotive consulting firms in the 1990s between stints at Automotive News. And yes, I helped prepare forecasts.)

That happened in recent memory. Back in 2007 and 2008, I know forecasters who took a lot of heat from clients for warning that sales could crash. One told me his original forecast was much lower, almost as low as 2009 U.S. sales actually fell to, but he modified it a bit to make it more believable to his client. Clients wanted only the upside.

That's the real danger in a big decline.

In the downturn, some forecasting firms stopped labeling three forecasts as best-case, most likely and worse case after a client growled “Forget the others. I only want to see the worst case. The new labels? “Worst-case” became “baseline” and “most likely” became “upside.” Clients were on the defensive and only willing to believe the worst.

Now the industry is in recovery, but clients haven't changed their caution yet. This week, forecasters say some clients aren't prepared for good news. Some most-likely forecasts are morphing into upside forecasts.

The caution is understandable, but some clients may not be ready if sales start to take off.