Imagine a shop that sells about 100 boxes of cereal a week. One week it sells 110. The shop manager thinks demand is rising, so orders 130 from the wholesaler, to cover the extra and rebuild stock. The wholesaler sees its orders jump by 30% and orders 170 from the factory. The factory sees a 70% jump and buys a lot of grain. A month later shoppers are back to 100 a week and everyone is stuck with too much.
This is the bullwhip effect: small changes in real demand get bigger and bigger as they travel up the chain, like a flick of the wrist that becomes a big crack at the end of a whip. The factory sees wild swings even though shoppers barely changed what they bought.
There are four usual causes. Each stage forecasts from the orders it gets, not from real sales, so it reacts to the stage below's overreaction. Ordering in batches (a full truck once a month) turns smooth demand into lumpy orders. Price promotions make customers stock up and then buy nothing for weeks. And when goods are short and rationed, customers order more than they need to get a fair share, then cancel when supply recovers.
Long lead times make it all worse. If an order takes four weeks to arrive, you'll often order again before the first one lands, forget what's already on its way, and end up ordering twice. The classic Beer Game, played in business schools since the 1960s, shows this: even sensible people produce a bullwhip when they can only see their own stage.
The fixes follow from the causes: share real sales data up the chain, so every stage plans from what shoppers actually buy; order smaller amounts more often; keep prices steady instead of running big promotions; shorten lead times; and always count what you've already ordered, not just what's on the shelf. The Supply Line game on this site lets you try it.
Practice
In the bullwhip effect, where are the swings in orders biggest?