How to improve OTIF: check the measure first | TransFormX Skip to content
TransFormX
Insights

Your OTIF figure is probably flattering you

Most on-time in-full figures measure the promise, not the request. Before anyone works on the performance, it is worth checking what the number is actually counting.

Shane O’Regan · Principal Consultant · CQP MCQI · CMILT
· 3 min read

Five questions to ask about your own operation

  1. 01Is OTIF measured against the date the customer asked for, or the date you confirmed back to them?
  2. 02Does a line count as in full when part of it ships and the rest follows later?
  3. 03Can you name the ten lines that cause most of your misses, without running a report?
  4. 04Does your customer’s figure for your performance match your own?
  5. 05When OTIF dips, does anything change in how the following week is run?

If you answered no, or were not sure, to two or more, the figure on your dashboard is unlikely to be the figure your customer holds.

Why a green dashboard and an unhappy customer can both be right

An OTIF figure is only as honest as its definition. The most common gap is the reference date. When performance is measured against the date the business confirmed, rather than the date the customer requested, every renegotiated delivery becomes an on-time one. The dashboard improves and the customer’s experience does not.

The second gap is partial shipment. If a line counts as in full once anything has left the building, a backlog can grow while the headline holds steady. The third is averaging. A site-level figure hides the handful of products and customers where almost all of the failure sits.

None of this is usually deliberate. Definitions drift as systems change and as people under pressure find ways to show progress. The effect is the same: the business works hard on a number that is not measuring the problem.

What the recovery looked like

86% → 97% On-time in-full across 40 sites, within one financial year

In a defence supply portfolio covering 40 sites, on-time in-full stood at 86%. The work started with the measure: agreeing the definition with the customer, then reading performance line by line rather than as a site average. That showed where the misses were concentrated.

From there the changes were operational. Service levels were redesigned around what the customer needed, a weekly performance forum took the place of monthly review, and operating routines were resequenced so failures were caught before they reached the customer. Within one financial year, on-time in-full reached 97% across all 40 sites.

Read the full case, with the limits of its evidence →

What to check this week

  • Recalculate last month’s OTIF against the date each customer requested, not the date you confirmed.
  • List the twenty lines with the most misses and work out what share of the total they account for.
  • Ask your largest customer for their figure for you, and compare it with yours.

If the recalculated figure is materially lower, that is where to start. It is usually where the quickest recovery sits.

Bring the number you are least sure of.

Thirty minutes on what is actually happening in your operation. Free, and no obligation either way. Bring the problem as your team describes it.

Book a confidential 30-minute discussion

Not ready to talk? Run the two-minute diagnostic →

More insights