Fill rate is the share of customer demand you satisfy directly from stock on hand, without a stockout or backorder. A 95% fill rate means 5% of what customers wanted was not available when they wanted it. It is measured three ways, by units, by order lines, or by whole orders, and each tells a different story about the same warehouse.

What is the fill rate formula?

It depends on the type. Unit fill rate is units shipped from stock divided by units demanded. Line fill rate is order lines filled complete divided by total order lines. Order fill rate is whole orders filled complete divided by total orders. In each case you count satisfied demand over total demand in the matching unit, excluding anything filled late from a backorder.

What is the difference between unit, line, and order fill rate?

They differ by what counts as demand. Unit fill rate counts individual units and is the most forgiving. Line fill rate counts order lines filled complete. Order fill rate counts whole orders shipped with nothing short and is the strictest, because a single missing item fails the entire order. The same period almost always shows unit fill above line fill above order fill.

What is the difference between fill rate and service level?

Fill rate measures how much of demand you satisfied. Cycle service level measures the probability of not stocking out at all during a replenishment cycle. They are not the same: a cycle can have a stockout and still post a high fill rate, because a stockout usually affects only the last sliver of demand before the next delivery. A 95% cycle service level typically yields a higher unit fill rate.

What is a good fill rate target?

Most operations aim for unit or line fill rates in the mid-to-high 90s, commonly 95 to 98%, and reserve near-100% only for items where a miss is unacceptable. Chasing 100% everywhere is rarely worth it, because the safety stock required climbs steeply as fill rate approaches 100%. Tier targets by item value instead of setting one blanket number.

How do you improve fill rate?

The main lever is safety stock, sized against demand and lead-time variability, but adding buffer everywhere just ties up cash. Better fill rate comes from accurate demand signals, shorter and more reliable supplier lead times, holding suppliers accountable for their own shorts, and concentrating buffer on the high-value items where a miss costs the most rather than spreading it evenly.

Why does fill rate usually read higher than order fill rate?

Because order fill rate is the strictest lens. It counts an order as filled only if every line ships complete, so one short line out of many fails the whole order while barely denting the unit count. A warehouse can ship 97% of units and 96% of lines yet complete only 92% of whole orders, which is the number the customer actually experiences.

What Is Fill Rate? Formula and Targets

Defining unit, line, and order fill rate, working each formula, setting realistic targets, and untangling fill rate from service level.

Fill rate is the share of customer demand you satisfy from stock on hand, without a stockout or backorder. It is measured three ways: unit fill rate (units shipped over units demanded), line fill rate (order lines filled complete), and order fill rate (whole orders filled complete). Higher is stricter.

Fill rate is the number that tells a customer whether you kept your promise. They ordered; did you ship what they asked, from stock, now. It sounds like one metric, but it is really three, and they can tell very different stories about the same warehouse. This post defines all three, works the math, sets realistic targets, and untangles the constant confusion between fill rate and service level, two terms people use as if they mean the same thing.

What is fill rate?

Fill rate is the proportion of demand that is satisfied directly from available inventory, rather than lost, delayed, or backordered. It answers the customer-facing question, when someone asked for product, how much of it could we actually ship right away. A 95% fill rate means 5% of what customers wanted was not on the shelf when they wanted it.

Because demand can be counted in units, in order lines, or in whole orders, fill rate comes in three flavors, and choosing which one you report is not a detail. The same period can show a comfortable unit fill rate and an ugly order fill rate at the same time, because a single missing item can fail an entire order while barely denting the unit count. Knowing which lens you are looking through is the whole game.

What are the three types of fill rate?

The three types differ by what they count as the unit of demand: individual units, order lines, or complete orders. Unit fill rate is the most forgiving, order fill rate the strictest, and line fill rate sits in between, which is why the same operation almost always posts unit fill rate above line fill rate above order fill rate.

The same order that ships 90% of units and 75% of lines complete counts as a total miss on order fill rate, because one line came up short.

Type	Formula	What it rewards
Unit fill rate	Units shipped from stock / units demanded	Shipping most of the volume, even if orders are split
Line fill rate	Order lines filled complete / total order lines	Completing each line item on an order
Order fill rate	Orders filled complete / total orders	Shipping whole orders with nothing short

How do you calculate fill rate?

You calculate each type over a period by counting satisfied demand against total demand in the matching unit. Work it as a short procedure using a period where customers requested 10,000 units across 500 order lines on 1,000 orders.

Pick the demand unit. Decide whether you are measuring units, lines, or whole orders; the choice sets the whole calculation.
Count total demand. Here: 10,000 units, or 500 lines, or 1,000 orders, over the period.
Count what shipped complete from stock. Say 9,700 units shipped from stock, 480 lines filled complete, 920 orders shipped whole.
Divide, matching the units. Unit fill = 9,700 / 10,000 = 97%; line fill = 480 / 500 = 96%; order fill = 920 / 1,000 = 92%.
Exclude backorders from the numerator. Only stock shipped on time counts; a line filled later from a backorder is a miss for the period it was due.
Track the trend, not just the snapshot. Rerun each period so you can see whether service is improving or slipping, and against which lens.

Notice the ordering: unit fill (97%) sits above line fill (96%) above order fill (92%), the usual pattern. If a boardroom is celebrating a 97% number, ask which fill rate it is, because the order fill rate the customer actually experiences may be five points lower.

How does fill rate relate to service level?

Fill rate and service level are related but not the same, and mixing them up leads to over-stocking. Fill rate measures how much of demand you satisfied. Cycle service level measures something narrower: the probability that you do not stock out at all during a replenishment cycle. A cycle can have a stockout yet still post a high fill rate, because the stockout usually bites only the last sliver of demand before the next delivery.

Service level is the chance of any stockout in a cycle; fill rate is how much demand you served. Because stockouts hit only the tail, fill rate usually reads higher than service level.

The practical consequence: do not set a 99% fill-rate target and a 99% service-level target as if they were the same dial. They are driven by the same lever, safety stock but they respond differently, and conflating them buys buffer you do not need.

What are realistic fill rate targets?

Realistic targets depend on the item and the channel, but most operations aim for unit or line fill rates in the mid-to-high 90s, commonly 95 to 98%, and reserve near-100% only for the items where a miss is unacceptable. The reason you do not just target 100% everywhere is that the safety stock required climbs steeply as fill rate approaches 100%. The last percentage point can cost more buffer than the previous ten combined.

Buffer rises gently through the 90s, then near-vertically toward 100%. Set targets by item value, not one blanket number.

So the sensible move is to tier targets: hold the highest fill rates on high-value, high-consequence items and accept lower fill rates on the long tail, the same logic ABC analysis and inventory service-level tiering apply to control policy. A blanket 99% target across every SKU is how working capital quietly balloons into excess inventory.

Targets also shift with the channel. A retail customer may hold you to a case fill rate on every purchase order and charge back the shortfall, while a spare-parts operation may accept a lower fill rate on slow-movers but demand near-100% on line-down critical items. One more distinction is worth settling up front: whether you report fill rate gross, against all demand including items you never stocked, or net, against the demand you committed to serve. The two numbers can differ by several points, so agree on the definition before anyone signs a service target against it.

Fill rate rarely lives alone, either. It usually sits inside a broader on-time-in-full, or OTIF, measure and the perfect-order metric, where an order counts only if it is complete, on time, damage-free, and correctly documented. Fill rate is the completeness piece of that scorecard, which is another reason to be precise about which version you are quoting.

What do the numbers say?

Context and definitions from standards and primary sources:

Fill rate and cycle service level are defined, distinct metrics in the supply-chain body of knowledge maintained by the Association for Supply Chain Management (ASCM/APICS) which separates the proportion of demand filled from the probability of not stocking out.
Fill rate is one component of the broader on-time-in-full and perfect-order measures that ASCM and the wider logistics field use to score customer service, where an order counts only if it is complete and on time.
The statistical link between service level, safety stock, and fill rate, buffer rising sharply as fill rate nears 100%, follows directly from the mathematics of demand variability that underpins standard safety stock formulas.

The takeaway is to name your metric precisely: a fill rate is not a service level, and a unit fill rate is not an order fill rate. The words carry real money.

Where fill rate breaks in practice

The common failure is that fill rate is computed after the fact, from data no one trusts, in a unit no one agreed on. Demand lives in an order system, what actually shipped complete lives in a warehouse system, and backorders live in an inbox, so the fill-rate report is stitched together weeks late and argued over rather than acted on. By then the misses are history. Harmony is an AI-native layer that connects machines, software, and paperwork into one operational layer, with no rip-and-replace, so orders, shipments, and stock positions become one live record instead of three. AI search returns cited answers across those records, so a planner can ask which items are dragging the order fill rate or which shorts traced to a single out-of-stock component and get a real answer, and Harmony's digital workflows route each shortfall and expedite to the right person. It is not an inventory-optimization product; it keeps fill rate honest and current by keeping the data in one place, the same paper-to-digital move Harmony makes on the floor (see the CLS case study and the product overview). The same clean signal lets you size the safety stock that actually moves fill rate, hold suppliers to a supplier scorecard when their shorts become yours, and shorten order-to-delivery lead time so the promise the fill rate measures is one you can keep.