Service level reads one thing. The shelf does another.
Service level is the percentage of order lines the distribution center fills on time. The formula is honest. It only measures as far as the store’s back door.
For logistics, it is the right indicator. It says whether the DC delivered what the store ordered, in quantity and on the day. Ninety-eight percent service level is a DC that works.
What it measures well
Replenishment performance between the distribution center and the store. Comparable across DCs, across suppliers, across weeks. For negotiating with a supplier and sizing a fleet, it is the number.
What it hides
Service level ends where the customer’s problem begins. It says the product reached the store. It does not say it reached the shelf.
A DC can run 98% service level while the store has 89% on-shelf availability. The difference is product in the back room, on an unopened pallet, on a cart parked in the aisle. The truck delivered. Replenishment did not finish the job.
The high number becomes an alibi. Logistics shows 98% and the lost sale at the shelf goes orphan, because each area looks at a different piece of the same gap.
The pair that is missing
Service level alone measures half the path. The pair that completes it is OSA, availability measured at the shelf, at the moment of purchase.
Read together, they locate the fault. High service level and low OSA: the problem is the store, in internal restocking. Low service level and low OSA: the problem is the chain, upstream of the store. Without the pair, the chain fights the supplier when the bottleneck is its own aisle.
The implementation note
A replenishment project that targets only service level pushes stock to the store and declares victory. The stock arrives and stops in the back room. The acceptance criterion has to include the shelf, not the dock.
It is the same aggregate reading that misleads in DSI and in phantom inventory. The number measures where it is easy to measure, not where the sale is lost.
Take the replenishment indicator that has sat on the director’s slide for six months and barely moves. Does it measure the dock or the shelf? If it measures the dock, it is healthy and blind.
Send me its name and how it is measured today. In one hour I will send back a one-page audit: the measurement quality problem, the pair that should sit alongside, and the size of the work if the gap confirms. If it confirms, the two-week Diagnóstico scopes the rebuild.