A procurement leader’s guide to the fulfillment KPIs that show whether orders arrive on time, complete, and undamaged.
A promised delivery date slips by two days. A line item lands short: three cases instead of ten. A production run pauses, or a store shelf sits half-empty, while someone works the phones to find out where the rest of the order is. Most procurement teams know this scene, and most describe delivery performance the same way afterward—it’s usually fine.
“Usually fine” is a feeling, not a number, and you can’t manage a feeling. A supply chain key performance indicator (KPI) is a measurement of how reliably goods move from the moment you place an order to the moment you receive it. A handful of these metrics, the ones focused on fulfillment, tell you whether orders arrive on time, complete, and undamaged, and each comes with a formula you can calculate from data you already have.
This guide covers five of them, including what each supply chain KPI measures, how to calculate it, and how to read the result. That can mean the difference between believing delivery is fine and being able to prove it, or catch it early when it isn’t. Measurement is also the groundwork for any broader supply chain optimization effort, since you can’t improve what you’ve never quantified.
A supply chain KPI is a quantified measure of how reliably goods move from order to receipt, expressed as a rate or an average so you can track it over time and compare one period against the next. The point of the number is the trend it reveals, not the single reading.
Supply chain management KPIs cover a lot of ground, so it helps to be specific about which ones this guide is about. The fulfillment metrics here—fill rate, on-time delivery, order lead time, perfect order rate, and backorder rate—measure how orders arrive. They sit apart from supply chain cost metrics like total landed cost, inventory metrics like carrying cost or inventory turnover ratio, and working-capital metrics like cash-to-cash cycle time, which answer different questions and live elsewhere.
Sourcing and purchasing have their own metrics, like cost savings and negotiated terms. This piece focuses on what happens after the order is placed: tracking how reliably your suppliers deliver on those commitments.
Measuring fulfillment turns delivery from a gut feeling into a managed number, and organizations that measure tend to respond to disruption faster than those that don’t, which can help protect customer satisfaction. When you can see a trend forming, you can act on it before it becomes a shortage.
The link between measurement and resilience isn’t just intuition. In EFESO’s 2026 Global Supply Chain Survey, 48% of the top-performing companies had mature digital capabilities, compared with just 10% of weaker performers. The same research found that firms leaning on advanced analytics moved faster from spotting a signal to acting on it, and that better visibility tracked with a stronger ability to catch disruption early and keep operations running.
Measurement also tells you how much buffer is enough, which matters more as costs move. In McKinsey’s 2025 risk survey, 45% of those exposed to tariff impacts said they were building up inventory in response. Buffer stock can be expensive, and fill-rate and backorder metrics are how you can decide how much you actually need rather than padding every line. Reliable fulfillment numbers are also the foundation of wider supply chain efficiency, where the goal is to hold service levels without carrying more inventory than the risk warrants.
Five supply chain KPIs cover most of what a procurement team needs to watch: fill rate, on-time delivery, order lead time, perfect order rate, and backorder rate. Many teams track these for a clear read on whether orders arrive on time, complete, and undamaged.
Remember: benchmarks are general on purpose, and a “good” number depends on your industry, your items, and your risk tolerance. So treat the guidance as direction rather than a fixed target.
Formula: Order Fill Rate = (Orders Fulfilled Completely on the First Shipment ÷ Total Orders Received) ✕ 100
Why it matters: Industry guidance commonly describes fill rate as the share of orders or units fulfilled without a stockout and distinguishes order- and unit-based versions of the metric. A fill rate that drifts downward can signal a stock or supplier problem worth investigating before it turns into a stockout. Stock levels are the lever here, so pair the metric with inventory optimization to keep the right items available without overbuying, tying up working capital, or putting unnecessary pressure on cash flow.
Benchmark: Many organizations set a differentiated target by item criticality, demand volatility, substitutability, lead-time risk, and the cost of shortage. Avoid applying the same fill-rate target to a production-stopping component and a low-value, easily substituted indirect item.
On-time delivery (OTD) measures the share of deliveries that arrive by the agreed due date. On-time in-full (OTIF) is the stricter version: it counts an order only when it arrives both on time and complete, folding punctuality and completeness into one measure.
OTD formula = (Deliveries or Purchase Order Lines Received On or Before the Agreed Due Date ÷ Deliveries or Purchase Order Lines Due) x 100
OTIF formula = (Orders or Purchase Order Lines Received On Time and In the Required Quantity ÷ Orders or Purchase Order Lines Due) x 100
Why it matters: Many teams monitor OTD and OTIF because logistics disruption and supply risk can affect delivery reliability. BCG’s 2025 analysis found that 19 of the world’s 30 leading ports, which represent 35% of global throughput, face high risk from extreme weather and rising sea levels. In that context, measuring delivery reliability and lead-time volatility by supplier, lane, and critical item can help you assess where promised dates are least reliable.
Benchmark: A strong OTD or OTIF score indicates historical performance against the date definition chosen. It does not by itself prove resilience against future disruption, so many procurement teams pair it with lead-time variability, supplier capacity, geographic concentration, and contingency-source measures.
Formula: Average Supplier Lead Time = ∑(Receipt Date − Purchase Order Release Date) ÷ Number of Completed Purchase Order Lines or Orders
Why it matters: Order lead time is the elapsed time between a defined order-release event and physical receipt. For supplier performance, many organizations use the purchase order release or transmission date, not the earlier requisition-creation date.
Measure supplier lead time at the purchase order-line level and, when possible, segment results by SKU, location, and supplier, since whole-order averages can conceal partial deliveries and item-level variation. Supplier lead-time guidance recommends comparing actual lead time with promised lead time and reporting both central tendency and variability measures such as percentiles and standard deviation.
Benchmark: Shorter is better, but consistency often matters as much as the average. A supplier that reliably takes 10 days is easier to plan around than one that averages seven but swings between three and 14 days. Many teams watch the spread, not just the mean, and feed what they learn into demand forecasting so reorder points reflect how long replenishment really takes.
Formula: Perfect Order Rate = (Orders Meeting Every Perfect-Order Condition ÷ Total Orders) x 100
AHRMM, the Association for Health Care Resource & Materials Management, uses an alternative four-component multiplication method based on purchase orders sent to suppliers and distributors. Its components are on-time delivery, complete shipment, damage-free delivery, and successful three-way matching against the receiver and invoice.
Component-product formula: Perfect Order Rate = On-time % x In-full % x Damage-free % x Accurate-documentation %
Why it matters: Perfect order rate is the share of orders delivered on time, complete, undamaged, and with correct documentation—an all-in-one view of order accuracy. It’s the strictest fulfillment measure because an order has to clear every hurdle to count.
Because the components multiply, small dips compound quickly. For example, four components at 95% each land at roughly 81.5% overall, so a modest perfect-order score could hide several individually manageable problems.
0.95 x 0.95 x 0.95 x 0.95 = 0.8145 or 81.5%
Many teams report the direct count of perfect orders when the data is available. Multiplying component rates can be a useful standardized composite, but it can differ from the directly measured perfect-order percentage because failures are not always independent. For example, late deliveries may also be more likely to be incomplete or have documentation errors.
Formula (unit-based measure): (Units Placed on Backorder ÷ Total Units Ordered) x 100
Formula (order-line measure): (Order Lines Placed on Backorder ÷ Total Order Lines) x 100
Why it matters: Backorder rate is the share of ordered items that couldn’t be filled from stock and went onto backorder instead. It’s the flip side of fill rate, viewed from the shortfall.
Use the same measurement basis for fill rate and backorder rate wherever possible. They’re related, but they’re not necessarily mathematical opposites. For example, an order may be partially shipped, partially backordered, canceled, substituted, or shipped late without formally being marked as a backorder.
Benchmark: A rising backorder rate is a prompt for root-cause analysis. Possible causes can include:
Forecast error
Inadequate safety stock
Inventory-record inaccuracy
Demand spikes
Supplier capacity constraints
Long or variable lead times
Supplier delivery failures
You can measure supplier performance from your own order data by tracking:
On-time delivery
Fill rate
Perfect order rate by seller across your purchase history
Then compare the same seller over time and against others. The record you build can show what happened on your orders—nothing more and nothing less. The judgment about which suppliers to keep, grow, or replace stays with you, informed by your own numbers.
A vendor scorecard is a structured way to organize this, combining the fulfillment KPIs above into one view per seller so comparisons stay consistent. The sourcing and cost side of the same suppliers lives in your procurement KPIs, and the two views together can give you the full picture of a relationship.
To build that record from data you already generate, Amazon Business Analytics can help you review on-time delivery and fill rate from your own Amazon Business order and shipment history.
To start, you can pick the two or three fulfillment KPIs that map to your biggest risk, pull the order and shipment data behind them, and review them on a set cadence rather than whenever something goes wrong. Two supply chain metrics watched consistently beat five checked once and forgotten.
A simple sequence helps keep it manageable:
Define each metric and its formula so everyone reads the number the same way.
Agree on the data source—such as your ERP or purchasing solution—so the figure is reproducible rather than reconstructed by hand each time.
Set a review rhythm—many teams opt for monthly reviews, while high-risk categories often require a tighter review cadence.
Act on the trend, not a single reading, since one bad week doesn’t necessarily mean the same thing as a three-month slide.
Consistent tracking depends on seeing your own order flow clearly, which is what supply chain visibility provides: the order-level detail that turns scattered receipts into a metric.
Amazon Business Analytics and business order information can help you calculate on-time delivery, fill rate, and perfect order rate from your own Amazon Business order-level data, so the numbers come straight from your transactions. And where backorders and stockouts are the risk you’re managing, Amazon Business Restock can help keep fast-moving items replenished, which may reduce how often you fall short in the first place.
Five fulfillment KPIs give you a clear read on supplier performance. The idea that ties them together is a simple one: measured orders are manageable orders. KPIs are what help you catch a slide early, size a buffer honestly, and hold a supplier conversation on evidence rather than impression.
You don’t need all five at once. You can instrument one KPI this quarter—usually the one that maps to your biggest risk—get the data flowing and the review rhythm set, then add the next.
When you’re ready to track these metrics, Amazon Business can help you measure fulfillment performance using your existing order history.
Get started today
Was this helpful?