How to measure supplier OTIF from a plain PO export
Most small manufacturers and distributors know which suppliers are "usually late." Fewer can say how late, how often, or whether it's getting worse. That gap matters: a supplier who's late a little on every order behaves differently from one who's on time most of the time and occasionally weeks late, and the fix for each is different.
On-time in-full, or OTIF, is the simplest number that closes that gap. You don't need a new system to track it. A purchase-order export with promised dates and receipt dates is enough. Here's how to build it so the number holds up when you put it in front of a supplier.
1. Agree on what "on time" means before you calculate anything
OTIF arguments almost always come down to definitions, not math. Decide three things up front and write them down:
- Which date is the target. Use the supplier's promised or confirmed date if you capture it, because that's what the supplier committed to. Your internal need-by date is useful too, but it measures your planning as much as the supplier's delivery.
- The tolerance window. Same day only, or a day or two early or late? Many teams accept early deliveries and allow zero days late. Whatever you choose, apply it to every supplier.
- What "in full" means. Usually quantity received is at least quantity ordered on that line. If you allow a small short-ship tolerance, state it.
2. Get the right columns out of your ERP
Whatever system you run, you need one row per PO line with these fields:
| Field | Why it matters |
|---|---|
| PO number, line number, supplier | Grain and grouping |
| Order date | Actual lead time |
| Promised (or confirmed) date | The on-time target |
| Quantity ordered and quantity received | The in-full test |
| Receipt date (last receipt for the line) | When the line became complete |
| Line status | To exclude cancelled lines |
If your system stores several receipts per line, use the date of the receipt that made the line complete. Using the first partial receipt will make suppliers look better than they are.
3. Count only lines that were due
This is the most common mistake. If you divide OTIF lines by all lines, open lines that aren't due yet drag the percentage down. The denominator should be lines that are not cancelled and whose promised date is on or before today. A line that is past its promise date and still open counts as a miss. It hasn't arrived on time, and it may never arrive in full.
Due line = status is not Cancelled AND promised date <= today OTIF line = due AND qty received >= qty ordered AND receipt date <= promised date OTIF % = OTIF lines / due lines
In Power BI that's two calculated columns and one measure. The same logic runs in our live demo, and you can filter it by category, supplier region, and warehouse.
4. Split OTIF into on-time and in-full
A single OTIF number tells you there's a problem. The two parts tell you what kind:
- Low on-time, high in-full: the supplier ships complete orders late. Look at the quoted lead time versus the actual lead time. The quote may simply be wrong.
- High on-time, low in-full: they ship on the date but short. That usually points to allocation, minimum pack sizes, or a capacity problem at the supplier.
- Both low: time for a direct conversation and possibly a second source.
5. Pair OTIF with lead-time variance
OTIF says whether the promise was kept. Lead-time variance says whether the promise was realistic. For each received line, calculate actual lead time (first receipt date minus order date) and subtract the supplier's quoted lead time. Track both the average and the spread. A supplier with a small average variance but a wide spread is hard to plan around, because you can't set safety stock against a number that moves.
When you raise this with a supplier, bring the line-level detail, not just the percentage. "Your quoted lead time is 21 days. Your actual average over the last 40 receipts was 29, and 9 of them took longer than 35" leads to a much better conversation than "you're late a lot."
6. Look at a trend, not a single month
Small suppliers may only deliver a handful of lines a month, so one bad week can swing the percentage a lot. Show OTIF monthly over the last 12 months next to the line count, and don't rank suppliers on fewer than a few lines. Set simple color bands so people can read it at a glance, for example red below 80%, amber from 80 to 90%, and green above that. Adjust the bands to your own baseline.
7. Make it a weekly habit
The metric is only useful if someone acts on it. A practical weekly routine:
- Refresh the report on Monday morning.
- Review the open lines that are past their promise date, largest value first.
- Send follow-ups to suppliers with unconfirmed or past-due lines.
- Once a month, review OTIF and lead-time variance by supplier and update the quoted lead times in your ERP to match reality.
Steps 2 and 3 are where automation pays off. A Power Automate flow can email the past-due list every morning, and an agent can draft the supplier follow-ups for a buyer to approve.
Want this on your own data?
Our PO Watch Starter builds supplier OTIF, PO aging, and a daily late-PO digest from a single ERP export in about two weeks, at a fixed price. Or try the interactive demo first.