OTIF: 95% On Time, 95% In Full — So Why Is Your OTIF 83%?

Your monthly report says 95% of lines shipped on time and 95% shipped in full. Then the customer's scorecard arrives with an OTIF of 83%, and everyone assumes somebody made a mistake. Nobody did. All three numbers are correct. They just count different things.

OTIF (on time in full) is the share of customer deliveries that arrived on the agreed date and with the complete quantity. Miss either condition and the delivery fails. Customers care about it most, because it measures what they experienced, not what your departments achieved.

How OTIF is calculated

The rule is simple: a delivery is a success only if it passes both tests. That gives OTIF two properties worth remembering.

• OTIF can never be higher than the weaker of the two. If on-time is 95% and in-full is 95%, OTIF is at most 95%. That happens only when the late deliveries and the short deliveries are the same ones.
• If the two failures are unrelated, OTIF is roughly on-time × in-full. 0.95 × 0.95 = 0.9025, so about 90%.

There is also a floor. If every late delivery is a different delivery from every short one, you lose 5% + 5% = 10%, and OTIF is 90%. So when on-time, in-full and OTIF are all counted the same way, 95% and 95% must give an OTIF between 90% and 95%.

Which brings us back to 83%. That number is only possible if OTIF was counted on a different basis than the other two. And it almost always was.

Order level vs line level: the hidden multiplier

Most internal reports count order lines. Most customers count orders. The difference is enormous.

Take one order with 10 lines. Nine ship complete, one ships 40 pieces short. At line level, in-full is 90%. At order level it is 0%: the customer received one delivery, and it was not complete.

Now a full month. You shipped 200 orders with 5 lines each, 1,000 lines in total.

• 50 lines shipped late. Late lines cluster, because a delayed order is usually delayed as a whole. Those 50 lines sit in 16 orders.
• 50 lines shipped short. Shortages are more scattered. Those 50 lines sit in 20 orders.
• 2 orders were both late and short.

At line level: on-time = 950 / 1,000 = 95%, in-full = 950 / 1,000 = 95%.

At order level: failed orders = 16 + 20 − 2 = 34. OTIF = (200 − 34) / 200 = 166 / 200 = 83%.

Same shipments, same month. The more lines per order, the wider this gap gets.

Think of a restaurant table for four. Three plates arrive hot, the fourth ten minutes later. The kitchen served 75% of dishes on time. The table remembers a late dinner.

The myth: re-promising protects OTIF

Here is the habit that quietly ruins the number. An order will miss its Friday date, so someone changes the confirmed date to next Wednesday. Now it ships "on time". The customer is exactly as annoyed as before.

The rule most customers apply is blunt: measure against the first date you confirmed, not the latest one. If your internal OTIF uses the latest date and the customer uses the original, your numbers and theirs will never meet. Keep both dates on the order and report against the original. The re-promised date tells the customer what to expect. It doesn't get to rewrite history.

Also agree whether you measure dispatch or receipt, and what quantity tolerance counts as "in full".

Where OTIF is really won or lost

When OTIF drops, the instinct is to push production harder. Often production wasn't the problem. Look at where failed orders got stuck:

1. The promise. The date was confirmed without checking stock and capacity. How a planner checks before saying yes is covered in sales orders to production.
2. The last line. Nine lines were ready on Tuesday; the tenth waited for one component. Order-level OTIF is ruled by the slowest line.
3. The last 48 hours. Production finished, but inspection, packing or paperwork did not. Finished goods that can't ship is a common, fixable leak.
4. The sequence. The shop floor worked by what was convenient, not by what completes orders. The rule used to pick the next job matters, as dispatching shows.

Car makers' suppliers learned this early: a line at 100% of target means nothing if one part number is missing at the customer's gate, as told in automotive production planning.

How factory.online helps

In factory.online, sales orders create manufacturing demand, so each production order and each shortage stays linked to the customer order it serves. A planner looking at a 10-line order sees which line is holding the rest back, whether it is waiting on material or on capacity, and what the finite schedule says about its finish date, before the Friday truck leaves without it.

OTIF vs OTD

They're not the same. On-time delivery (OTD) checks only the date. OTIF adds the quantity test, so it is always equal to or lower than OTD on the same basis. For how OTIF fits with other measures, see manufacturing KPIs.

If your internal numbers and your customers' scorecards keep disagreeing, book a demo and we'll trace a few of those orders together.