A retailer sees consumer demand rise five percent. They order ten percent more to rebuild stock. Their distributor, seeing a ten percent jump, orders twenty percent more from you. You, facing what looks like a twenty percent surge, buy forty percent more raw material and add a shift. Consumer demand moved five percent. Your raw material commitment moved forty. That amplification is the bullwhip effect, and it’s the reason supply chains oscillate between shortage and glut without underlying demand doing anything dramatic.

What Amplifies the Signal

Each tier reacts to orders rather than to end demand. You can only see what your customer ordered, which is already a distorted version of what their customer ordered. Distortion compounds at every step, and by the time it reaches raw material suppliers it bears little resemblance to what actually happened in the market.
Batching makes it worse. If a customer orders monthly to save on freight, you see nothing for four weeks and then a spike. That spike is a shipping decision, not a demand event, but it looks identical in your data.
Shortage behaviour is the most damaging amplifier. When supply is tight and orders get allocated proportionally, customers inflate orders to secure a bigger share. Everyone does it, apparent demand balloons, manufacturers expand, and when supply loosens the phantom orders evaporate. The resulting inventory correction is often more painful than the original shortage.
Promotions and price incentives add another layer. A quarter-end discount pulls demand forward. The following period looks weak. Neither period reflects consumption, and both go into your forecast history as if they did.

Reducing the Swing

Get closer to real demand. If you can see sell-through data from your customers rather than only their orders, you’re forecasting against something much closer to reality. Many customers will share this given a reason to — it’s in their interest for you to hold the right stock.
Shorten lead times. Amplification scales with how far ahead everyone has to commit. Every week you take out of your lead time is a week less speculation in the system.
Change allocation rules. Allocating scarce supply based on historical purchase volumes rather than current order size removes the incentive to inflate.
Smooth incentives. Quarter-end pushes create the lumpiness you then struggle to forecast. The cost of that distortion is rarely counted against the revenue the push generated.

Seeing it in Your Own Data

Compare the variability of your customer orders against the variability of your own purchase orders to suppliers over the same period. If yours swings more than theirs, you’re amplifying — and the amplification is coming from your own reorder policies, not from the market.
This is a genuinely useful diagnostic because it points inward. It’s tempting to treat volatility as something the market does to you, but a meaningful share of it is manufactured internally by reorder points that overreact, minimum order quantities that force lumpy buying, and safety stock rules that get adjusted upward after every shortage and never back down. Those are all within your control, and fixing them costs nothing but attention.
ticktick.ai separates genuine demand movement from ordering artefacts like batching and promotional pull-forward, so planning responds to signal rather than noise.