Service Level, Fill Rate & On-Time Delivery

Holding stock and running a smooth process are means to an end, and the end is simple: customers get what they ordered, complete and on time. To know whether that is happening, operations leans on a few plain measures — service level, fill rate and on-time delivery. Each looks at the same promise from a slightly different angle, and together they keep “we’re doing fine” honest.

Fill rate: did the order arrive complete?

Fill rate is the share of demand you met from stock on hand. If customers asked for 1,000 units and you shipped 950, your fill rate is 95%. It is a blunt, useful number because it counts what actually went out the door against what was wanted. A high fill rate means few customers walked away empty-handed; a low one means stockouts are biting. You will often hear a target like “a 95% fill rate,” meaning the business aims to satisfy 95 of every 100 units demanded directly from stock. The remaining five are stockouts — sales delayed, lost, or turned into backorders. Watching the number week to week tells you instantly whether your stocking is keeping pace with what customers ask for.

On-time delivery: did it arrive when promised?

A complete order that shows up late is still a broken promise, which is why on-time delivery (OTD) is tracked separately. It is the share of orders delivered by the date you promised. You can have a great fill rate and poor OTD — everything ships, but always two days late — or the reverse. Watching both stops one good number from hiding a bad one. A courier that delivers every parcel but always a day late looks perfect on fill rate and dreadful on on-time delivery — and the customer remembers the lateness, not the completeness. Many businesses also combine “complete” and “on time” into a single stricter measure, often called on-time-in-full (OTIF): the order only counts if it arrived both whole and punctual.

Service level: the overall promise

Service level is the umbrella idea — the probability that you can meet customer demand when it comes. Stated as a target like “a 95% service level,” it sets how often you are willing to fall short. Notice the honesty in that number: it openly admits you will miss 5% of the time. That is deliberate, not a failure of ambition.

Why 100% is the wrong target

Chasing a 100% service level sounds noble but is usually a trap. To never, ever run out, you would need enormous safety stock for the rarest demand spike — and the cost of that last sliver of reliability climbs steeply. Going from 95% to 99% can cost far more than the first 95% did, for a benefit few customers notice. So businesses pick a service level that balances the cost of holding stock against the cost of disappointing a customer. The right answer is high, but rarely perfect.

A 95% service level is a deliberate choice, not a shortfall: the last few percent of reliability usually cost more than they are worth.

Spot the performance measure

Read each metric and decide which one it is — fill rate, on-time delivery, or service level. Tap a card to check.

Sort the metric uses

Drag each scenario into the bucket it belongs to — improve fill rate, improve on-time delivery, or reconsider service level. Hit Check placement when you’re done.

How to use it

These measures only help if you read them together and remember what each one quietly ignores.

• “Our fill rate is 95%, but on-time delivery is only 88% — we ship complete, just late.” (stops a good number masking a weak one)
• “Let’s track OTIF, so an order counts only if it’s whole and punctual.” (raises the bar to the customer’s real experience)
• “Pushing service level from 95% to 99% would cost more than it’s worth.” (weighs reliability against stock cost)
• “Demand spiked, so our fill rate dipped — expected, within target.” (reads a number against its goal, not zero)

Use the trio to keep performance claims grounded in what the customer actually received.