Why we built this

A forecast is never set in stone.

Which is strange, because almost every tool you've been sold treats it as though it were.

Accuracy is sold with an asterisk

Vendors sell accuracy the way economists sell theory: all other things being equal.

All other things are never equal. Thirteen major disruptions hit supply chains in five years. Every accuracy figure earned in calm water got quoted in a storm.

And the forecast that actually gets published has already been negotiated before it ever meets reality, nudged by targets and politics and somebody's optimism. Then it gets graded as though it were pure mathematics.

We don't think accuracy was ever really the point

What matters is the quality of the decisions you make, and how fast you revise them when the world argues back.

Accuracy tends to follow from a system that records what happened, learns from it, and acts. You can't buy it off a shelf.

This didn't start with research

It started with meeting the same person over and over.

A planner, in a real company, changing the forecast again. In a spreadsheet. For the fourth time that month. With no way of showing anyone why.

Only later did we go and read what the forecasting scientists had found. Most of it was already there.

What forty years of research says

8 in 10forecasts get changed by a person before anyone uses them.
~halfof those changes help. The other half quietly hurt.
Your next adjustment is predicted more strongly by your own last one than by anything the model got wrong.

The research described exactly what we were already looking at. What it never did was produce a product, because the change happens in a spreadsheet, and a spreadsheet sits outside every system anyone could have built one into.

Read the eight studies →

The problem was never that planners adjust the forecast. It's that the good changes were never written down anywhere you could learn from.

We're not here to fix you

If your manager wants the number up, the number goes up. That's your job and it isn't our business.

We won't block it. We won't lecture you about it. We are never going to build a leaderboard.

What we'll do is make the change quick, write down why you made it, and tell you honestly how it turned out.

Do that for a year and you'll have something planners have rarely had.

We're not trying to correct the real world. We're trying to nudge it.

Never trust a forecast vendor who grades their own homework

Every change made through Piwaka is scored against your actuals and your own baseline, by rules that were set before the change was made.

That scorecard lives in your ledger. Not in our marketing.

"A forecast is never set in stone.
It's an opening position."

Change a forecast and see for yourself.