OTIF

How OTIF is worked out

Count the deliveries that passed both gates, divide by the total number of deliveries, multiply by one hundred. Two gates, and a delivery has to clear both: late but complete fails, on time but short fails.

Everything then hangs on one definition: which date. If you measure against the date the customer asked for, you are measuring your whole business, including whether sales promised something the warehouse could not do. If you measure against the date you confirmed, you are measuring only whether you kept your own word. Retail customers use the first. Suppliers reporting internally almost always use the second, and the gap between the two numbers is frequently the most useful thing in the report.

Two smaller definitions matter nearly as much. An early delivery usually counts as a failure, because a distribution centre works to booked slots. And a split delivery usually counts as not in full, however complete it eventually becomes.

There is no good number

A supplier to a grocery chain lives with a target in the high nineties written into a contract, with a deduction for every point below. A business-to-business distributor with flexible customers may never calculate OTIF at all. The targets are contractual rather than natural, so a figure from somebody else's trade tells you nothing.

What the figure is for is the trend against your own contractual target, and the split between the two gates. A month at 92 per cent made up almost entirely of late deliveries is a transport problem. The same 92 per cent made up of short deliveries is a stock problem, and fill rate is where you go next.

In practice

Of those 500 deliveries: 470 arrived inside the agreed window, so on time was 94 per cent. 482 arrived complete, so in full was 96.4 per cent. Counting deliveries that passed both gates gives 462, so OTIF was 92.4 per cent.

Now change one definition. Measured against the date each customer originally asked for, rather than the date the sales team confirmed, 420 of the 500 were on time, and OTIF falls to 84 per cent. Same month, same lorries, same warehouse, eight and a half points of difference, and the second figure is the one the customer is looking at. The numbers here are an illustration.

What moves the figure

• Order confirmation discipline. The cheapest improvement available to most suppliers is confirming a date the warehouse can actually hit, rather than repeating the date the customer asked for.
• Booked delivery slots. A missed window at a retail distribution centre can mean a rejected load and a fine, and it scores as a failure even when the goods were ready on time.
• Availability. The in-full gate is decided by stock, not by the warehouse, and that makes OTIF partly a buying metric.
• Cut-off times against transit time. A customer two days away needs the order picked a day earlier than one next door, and OTIF reports rarely separate the two.
• Split shipment policy. Shipping what you have looks like good service and scores as a failure. Decide which you are optimising for.
• Who owns the figure. OTIF spans sales, buying, the warehouse and transport. Where it belongs to one of them alone, it tends to be reported on the definition that department looks best under.