How to Measure On-Time Delivery | Zudo Works
  1. Home
  2. Answers
  3. Running the shop

How to measure on-time delivery in a small shop

On-time delivery is the share of order lines shipped in full on or before the date promised to the customer. OTD = lines delivered on time ÷ lines due in the period × 100. Measure against the first date promised, not a later revised one, count by line rather than by order value, and review every late line by cause each week.

By Arunkumar V, CTO · 03 November 2026 · 2 min read

  • how to measure on-time delivery
  • OTD formula
  • on-time delivery KPI
  • OTIF for small manufacturers

The formula and what counts as on time

In words: take every order line due in the period, count how many shipped complete on or before the promised date, divide by the total due and multiply by 100. In one line: OTD = lines delivered on time ÷ lines due × 100. Three decisions make the figure honest.

First, use the date you first promised the customer, so that moving the date does not improve the score; if the customer asked for the change, record that and use the new date. Second, a line counts as on time only if the full quantity shipped. Third, decide whether on time means the day shipped or the day received, and use the same rule every month.

Worked example

The numbers are invented. Say 84 order lines were due in October. Seventy-one shipped in full on or before the promised date. Six shipped in full one to three days late, four shipped partially on the due date with the balance a week later, and three had the date moved by the shop without the customer asking and shipped on the new date. On time is 71, late is 13, so OTD = 71 ÷ 84 × 100 = 84.5 per cent.

If the three moved dates had been counted as on time, the figure would read 88.1 per cent and hide the problem.

Finding why lines are late

Record a cause against each late line when it ships, from a short fixed list. Publish the percentage and the late lines together, review them weekly with the planner, buyer and supervisor, and act on the top cause. Track promised lead time too: hitting 95 per cent by quoting eight weeks when customers want three is not winning.

  • Material shortage or late supplier delivery.
  • Capacity: the work centre was booked with other jobs.
  • Quality: rework or scrap delayed the job.
  • Promised date was unrealistic when the order was taken.
  • Customer change after order, with the date agreed.
  • Subcontractor late with outside processing.

How to measure on-time delivery

  1. Step 1

    Record the first promised date on every sales order line when the order is accepted.

  2. Step 2

    Record the ship date and quantity shipped on each line when it leaves.

  3. Step 3

    At month end, list every line due in the month and mark it on time only if shipped in full by the promised date.

  4. Step 4

    Calculate OTD = lines on time ÷ lines due × 100, and record a cause code against every late line.

  5. Step 5

    Review the late lines weekly, fix the top cause, and track the percentage month by month.

Where ZMOS fits

ZMOS tracks work orders with status changes recorded in the audit log, holds customer records in the same workspace, and lets a shop add a promised-date or cause-code custom field to any form with workflow rules on status changes. Zudo AI can calculate from workspace data within the user's role and propose a chart, which a person approves. It suits sheet-metal, machining and fabrication shops of 10 to 50 people. If you need carrier integration and OTIF reporting to a retailer's standard, check that scope in a demo first.

What ZMOS does

People also ask

Short answers, each one complete on its own.

Related answers

Inside ZMOS

Zudo AIStock ledgerLow-code forms and rulesBarcode job cardsRoles, audit log, APISecurityFree trialWhy Zudo Works

What is the difference between OTD and OTIF?

OTD is on-time delivery: the line shipped by the promised date. OTIF is on time in full: by the date and in the full quantity. Most small shops use OTD but count part deliveries as late, which makes the two figures the same in practice.

Should OTD be measured by order, by line or by value?

By line is the most useful for a job shop, because one order can hold several parts with different problems. By order hides which part was late. By value lets small late lines disappear, even though a small part can stop a customer's assembly.

Which date should I measure against if the customer changes it?

If the customer asked for the change in writing, use the new date and record why. If the shop moved the date because it could not make the original, use the original. The difference is who caused the change.

What is a good on-time delivery percentage?

That depends on your customers and the lead times you promise. Rather than chasing a published benchmark, track your own figure monthly, aim for it to rise, and watch the late lines by cause.

See it on your own stock and jobs

ZMOS is free for six months for startups and registered partners, no card. Or ask an engineer anything on WhatsApp; we reply within 6 hours in working hours.