The engine

You do not need to watch 240 SKUs. You need the four that moved.

Set a tolerance per SKU. Planster compares actual sales against the forecast and tells you which ones ran outside it, and by how much — early enough that lead time is still on your side.

A tolerance per SKU, not one number for the catalogue

A 15% swing on a slow mover is noise. The same swing on the SKU that carries a third of your revenue is next quarter’s problem. Tolerances are set per SKU so the list you get is the list that matters.

By how much, not just that it happened

The exception carries the size of the miss and the direction. Running 40% above forecast for three straight weeks is a different decision from a single spike, and you can see which one you are looking at.

Early enough to matter

An exception you find at month end is a post-mortem. Planster surfaces it against your lead times, so the question is still "do I move this order up" rather than "how long are we out for".

Overnight

What the agent does with exceptions

Every night

Exceptions are what the agent ranks the morning digest around. It re-checks every SKU overnight and the ones outside tolerance are what lands at the top of the email.

What it can’t do

It cannot change your tolerances or decide an exception is not worth telling you about. You set the thresholds and you can silence any check yourself.

Forty percent up, for three weeks, unnoticed

Amazon velocity on a mint SKU ran 40% above forecast for three consecutive weeks. Nobody was watching that SKU — it had never been a problem. The exception put it in front of the planner in week one with the size of the miss attached, which is the difference between moving a purchase order forward and explaining an eighteen-day stockout.

Find out what you are not watching

Connect your data and see which SKUs are outside tolerance right now. Most brands find one they did not know about.