Which late purchase orders actually matter?

The problem with the past-due report

Most ERP systems can print a list of open purchase orders past their promised date, grouped by supplier or sorted by how late they are. On a normal Monday that list can run to dozens of lines, and a buyer has time to chase only some of them properly.

Sorting a past-due purchase order list by days late feels logical, but it answers the wrong question. A purchase order that is three weeks late for a part with two months of stock on the shelf is a supplier performance issue, not an emergency. A purchase order that is two days late for a part that runs out on Thursday is the one that stops a line.

The question that matters is simpler: will this part run out before this order arrives?

Why this matters more right now

In the Institute for Supply Management's September 2026 Manufacturing PMI report, the Supplier Deliveries index was 59.0, against 59.3 in August. ISM reads that as deliveries still slowing, at a slightly slower rate than in August, and the index has pointed to slower deliveries for 10 months. Slower deliveries mean more open orders drifting past their promised dates, and longer past-due lists to sort through.

The data you need

You can do this with data you already have, in two exports. No integration is required.

• Open orders: PO number and line, part number, supplier, open quantity, promised date (blank if the supplier has not confirmed one) and status.
• Stock by item: usable quantity on hand today and usage over the last 4 to 12 weeks, so you can compute an average daily usage.

Our guide to preparing an ERP extract lists the fields for the most common systems.

The method, step by step

1. Compute days of cover for every part with an open order

Days of cover = usable stock on hand / average daily usage. Use the same usage window for every part. If a part has no usage history, flag it rather than guessing.

2. Decide when each order will really arrive

• Order not yet late: use the promised date.
• Order already late: the promised date is no longer a date. Use the supplier's latest confirmed date if you have one; if not, treat the arrival as unknown.
• Order with no promised date: treat the arrival as unknown.

Unknown arrival dates are not a reason to skip a line. They are a reason to look at it first when cover is short.

3. Compare cover with arrival

For each late or undated order line, ask whether the part runs out before the expected arrival.

• Cover shorter than the days until arrival: the part will be short before the order lands. This is a shortage risk.
• Cover longer than the days until arrival: the order is late, but stock carries you. Track it, do not escalate it today.
• Arrival unknown and cover short: treat it as a shortage risk until the supplier gives you a date.

Check other open orders for the same part. A second order due before the stock runs out may cover the gap; one due after it does not.

4. Rank by days of cover, shortest first

Among the shortage risks, the part with the fewest days of cover goes to the top. That is your chase list for today.

5. Keep the evidence with each line

For each line on the chase list, write down the numbers behind it: stock, average usage, days of cover, promised date, days late, and any other open orders for the part. When you call the supplier, you are discussing facts, not urgency. Our supplier escalation playbook covers how to run that conversation.

Illustrative example, not a customer result

Two late orders on the same Monday morning:

• Order A, gasket kit: 9 days late, supplier now confirms delivery in 7 days. Stock on hand covers 3 days of usage. The part runs out 4 days before the order arrives. Chase today, and look for an alternative source or a partial shipment.
• Order B, steel bracket: 21 days late, no new date from the supplier. Stock on hand covers 50 days of usage. The order is badly late, but nothing runs out this month. Ask for a date, log it for the supplier review, and move on.

Sorted by days late, Order B comes first. Sorted by days of cover, Order A does, and Order A is the one that can stop production.

Where the method breaks

• Promised dates are only as good as what suppliers confirmed and what was entered. If confirmations are not recorded, many orders will show as undated.
• Average usage hides spikes. A part with steady usage and a large order next week will look safer than it is.
• Days of cover tells you when a part runs out, not which customer order or work order it affects. Linking the two still requires your planners and your bill of materials.

Lead times set once in the ERP make all of this harder; see why your ERP lead times are probably wrong.

How WERISCO does this

WERISCO reads your stock, usage and open purchase orders together and tells you which late or undated supplier orders will leave a part short before they arrive, ranked by days of cover, with the evidence for each one. Your team decides what to chase; nothing is written back to your ERP.

It starts from exports you already run. Before analyzing anything, it shows which fields are missing or inconsistent, and what that blocks. To see it on your own data, book a supply chain audit: 20 minutes, two exports, no integration.