Most safety stock is a round number someone picked years ago and nobody has dared to touch since. There is a better basis, and it is not complicated.
Ask most teams why their safety stock is set where it is, and the real answer is history. Someone chose two weeks of cover, or a 20% buffer, a long time ago; it felt safe, and it calcified. The figure is not wrong because it is round. It is wrong because it is disconnected from the two things that actually determine how much buffer you need.
Safety stock exists to absorb two kinds of surprise: demand that is lumpier than expected, and lead times that stretch longer than promised. The more variable either one is, the more buffer you genuinely need; the less variable, the less you should be carrying. A flat percentage across the whole catalog over-protects your steady items and under-protects your volatile ones, which is the worst of both worlds.
The honest basis for safety stock is your real forecast error and your real lead-time variability, item by item, set against the service level you have chosen to hold. Textbook as it sounds, the buffer scales with how uncertain you are and with the square root of the lead time; a 99% service target carries more than a 90% one, and it should. The point is not the formula. The point is that every buffer becomes a decision you can explain, instead of a number you inherited.
When safety stock is sized on evidence, two things happen. The buffer on your predictable items falls, releasing cash you did not know was trapped. The buffer on your volatile items rises, where it was quietly causing stockouts. Same service level, often less total inventory, and for the first time a reorder policy any planner can defend in a room full of skeptics.
A buffer you cannot explain is not caution. It is cash you have decided not to look at.
Picture your next S&OP meeting: one number on screen that nobody argues with. See it free, on your own data, in 45 minutes; and if it does not hold up, we part as friends.