Blog · Board and management reporting
The honest answer to what on-time-in-full rate a supplier or a distributor should reach: the 95 to 98 percent figures retailers demand and the 85 to 90 percent many suppliers achieve are measured by different rules. This page gives the ranges by channel and the definition each uses, the three measurable things that set the right figure for one desk, the definition of on-time and in-full, the level per customer against its own history, and the cause split, and the table to compute before anyone quotes a percentage.
OTIF is the share of orders delivered on time and in full, and the question of what it should be has three answers depending on who measures it and how.
| Channel | Requirement or typical | On-time rule | In-full rule |
|---|---|---|---|
| Large grocery and mass retail | 95 to 98 percent, with chargebacks | Requested date, window of 0 to 1 day | Order level, cases shipped ÷ cases ordered ≥ threshold |
| Mid-size retail and foodservice | 92 to 96 percent | Requested date, 1 to 2 days | Order or line level |
| Industrial distribution | 90 to 95 percent | Promised date, 2 to 3 days | Line level |
| Construction and project supply | 85 to 92 percent | Agreed date, often renegotiated | Line level |
| Aftermarket parts | 90 to 97 percent, by urgency class | Promised date by class | Line level |
The same delivery can score 97 under one rule and 84 under another. The OTIF worked example computes ten lines under two rules.
On time = delivered date within [requested date − early tolerance, requested date + late tolerance] In full = quantity delivered ÷ quantity ordered ≥ threshold, at the stated level OTIF = orders (or lines) both on time and in full ÷ all orders (or lines)
Every term is a choice. Requested or promised date. Early tolerance, because early is a miss for a retailer with no dock space. Line or order level. The desk states its own, and computes the customer's beside it for every customer that has one.
| Customer | Rule | OTIF this month | Trailing 12 months | Change | Chargebacks |
|---|---|---|---|---|---|
| Retailer A | Requested, 0 days, order | 91% | 94% | −3 | $18,000 |
| Distributor B | Promised, 2 days, line | 96% | 95% | +1 | none |
| Retailer C | Requested, 1 day, order | 97% | 96% | +1 | $0 |
The company-wide 95 hides A's slide and A's chargebacks. The row per customer, against that customer's own history and rule, is the number to act on.
| Cause | Misses | Share | Owner |
|---|---|---|---|
| Stock-out at pick | 41 | 46% | Planning |
| Warehouse, late pick or short pick | 18 | 20% | Operations |
| Carrier | 15 | 17% | Logistics |
| Customer date not agreed | 9 | 10% | Sales, order entry |
| Order data error | 6 | 7% | Order entry |
Half the misses are stock-outs. Improving OTIF here is a planning problem, and the lead time variability piece is where it goes next.
| Measure | Formula | From |
|---|---|---|
| OTIF, own rule | On time and in full ÷ all, stated rule | Order and delivery files |
| OTIF, customer's rule | Same, per customer with a rule | Same, plus the customer's terms |
| Per customer, trailing 12 | Same, by customer, by month | Same |
| Cause per miss | One cause by a stated mapping from reason codes | Delivery and warehouse files |
| Cost of misses | Chargebacks plus expediting, by customer | Deductions and freight |
| Identity | Orders = OTIF + late only + short only + both | Order file |
One rule for everyone. The supplier at 95 charged for being at 88 under the customer's rule.
Company-wide figure. The slide at the one customer that charges, averaged away.
No cause. Operations blamed for stock-outs planning caused.
Early counted as on time. The retailer refuses the truck; the supplier's report says delivered.
A good OTIF rate is the one each customer requires, measured by that customer's rule, stable or rising against its own history, with the misses split by cause and owned. Company-wide, most suppliers are healthy at 92 to 96 under their own rule and need 95 to 98 under a large retailer's. Covirage computes both rules per customer from the order and delivery files every week, with the cause split and the identity checked.
Because the rules differ. The supplier measures against the promised date with a two-day window and a line-level in-full; the retailer measures against the requested date with a same-day window and an order-level in-full, and counts a partial as a miss. Both are correct by their rule. The supplier should compute both and show the customer's rule beside its own.
Only where a customer charges for less. Reaching 98 on a customer with a same-day window costs inventory and expediting; reaching 98 on a customer with a week's window costs nothing. The target per customer follows what the customer measures and charges, not a company-wide number.
Ownership. An OTIF miss is a stock-out, a warehouse pick, a carrier delay, a customer-requested date the supplier never agreed, or a data error in the order. Each has a different owner and a different fix. An OTIF rate without the split is a score with no way to raise it.