On-Time In-Full (OTIF)

A supplier delivers every item a retailer ordered, but a day after the agreed window, while another arrives on time with only half the units. Neither qualifies as OTIF under a strict order-level rule because each misses one condition.

DHL defines the combined delivery measure, and McKinsey explains how different definitions create disputes in consumer supply chains, so the sources support setting shared rules rather than assuming one universal calculation. Choose whether the unit is an entire order, a purchase-order line, a case or a shipment, because a split order can look different under each method.

Record the promised date or window when it is agreed, since a target changed after a miss should not quietly rewrite history. Specify whether on-time means arrival at the gate, unloading, receipt scan or shelf availability, because operational control differs at each point.

Define in-full against the accepted order quantity and permitted substitutions, so a partial delivery does not pass because the invoice was revised later. Agree which cancellations or customer-requested changes are excluded, separating supplier failure from legitimate order changes.

For an order-based measure, divide orders meeting both conditions by eligible orders due during the period, so if 920 of 1,000 eligible orders meet both requirements, OTIF is 92% for that period. Do not add separate on-time and in-full percentages, because the intersection, not an average, is the numerator.

Some contracts assess each order line instead, and a large multi-line order can be weighted differently, so label the method. Track the reason for each failure, such as stock shortage, production delay, bad address, carrier issue or receiving bottleneck, and report quantity-fill rate too so the size of shortfalls is visible.

A delivery may reach a distribution centre on time but miss store availability, so choose the handoff that matches the business promise. Proof of receipt and timestamps should come from agreed records, because carrier departure scans alone do not establish customer receipt, and time zones and cut-off times need checking for cross-border orders.

A late but complete shipment may still be useful to the customer, but it does not meet a strict OTIF promise, and a small shortage can fail an all-or-nothing order rule. A high OTIF figure can mask selective refusals or cancelled difficult orders, while a low figure may reflect unrealistic promises rather than only execution failures.

Segment performance by supplier, product, lane and customer, compare seasonal and promotional periods with context, and track lead time and forecast accuracy separately. Lock definitions across reporting periods, pair OTIF with damage rate and complaint data, never encourage unsafe driving or rushed quality checks, and use the measure to improve the promise and the handoff, not simply to rank partners without shared evidence.