Skip to content

Systems and records

Supplier lead time correction data: recording a delay after the customer promise

By SourceX Editorial · Reviewed by Noah Loul ·

9 sections

Short answer

Record a supplier lead-time change after a customer promise as its own dated event rather than overwriting dates. Link the original promise and its supply assumption, the supplier's notice and when it arrived, every affected customer order, the recovery options and the constraints that ruled some out, the decision and decider, the customer notice and response, and the outcome. That history shows how commitments were recovered, not just final dates.

System export

Supplier lead time

This page covers supplier lead time correction data: recording a delay after the customer promise.Illustrative

Key takeaways

  • Never overwrite a promised date; each revised promise is a new version with a timestamp and reason.
  • Map which dates your ERP keeps and which it replaces in place before extracting anything.
  • Record rejected recovery options with the constraint that ruled them out, such as an unapproved source for the current specification revision.
  • Keep four timestamps for every affected customer line: when the business knew, when it told the customer, the revised date offered and the customer's answer.
  • Label infeasible promises and overrides separately from genuine supplier changes.

How should workflow data capture a lead-time change after a customer promise?

As its own dated event, linked in both directions. The record should show the customer promise as it stood, the supply it relied on, the supplier's change with when and how it arrived, every customer order the change touches, the recovery options that were considered, the decision and who made it, what the customer was told and when, and what finally shipped. The original promised date is never overwritten; a new promise is a new version.

Steps

How should workflow data capture a lead-time change after a customer promise

Most systems store the opposite: a purchase order line with today's confirmed date and a sales order with today's promised date. That snapshot hides the event an AI system most needs to learn from, which is the moment a commitment became wrong and a person had to choose how to recover. Supplier lead time correction data is the history that keeps that moment intact.

A complete event answers six questions:

  • What was promised to the customer, on which version of the order, and on what supply assumption?
  • What changed on the supply side, according to which message or system update, and when was it received?
  • Which customer commitments depended on that supply line, and how was the dependency known?
  • Which recovery options existed, and which constraints ruled some of them out?
  • Who decided, with what authority, and what was the customer told?
  • What actually happened, and was the lead time master data corrected afterward?

Which dates does each system keep, and which does it quietly replace?

ERP systems hold several lead times and several dates per order, and they do not all behave the same way when something changes. Before extracting anything, map which fields are planning defaults, which are commitments, and which are overwritten in place. The table summarizes behavior documented by Odoo and Microsoft Dynamics 365 Supply Chain Management.

One documented behavior deserves attention. Dynamics 365 updates order promising information on a changed sales order only when the existing promised date can no longer be met; if a change makes an earlier date possible, the old date stays. An unchanged promised date therefore does not prove that nothing changed underneath it.

Which dates does each system keep, and which does it quietly replace?
System and objectDates and lead times keptBehavior on changeWhat to extract
Odoo sales orderProduct delivery time plus sales security lead time give an expected delivery dateOdoo warns when a set delivery date is earlier than the expected dateThe expected date, the date set, and whether the warning was overridden
Odoo purchase orderVendor lead time, counted from PO confirmation to receipt, and days to purchaseLead times are product and vendor settings, not per-event historyLead time values as of each PO confirmation date
Dynamics 365 sales orderDelivery date control method such as sales lead time, ATP or CTPPromising is recalculated only if the current date cannot be fulfilledPromised ship and receipt dates per order version
Dynamics 365 purchase orderRequested receipt date and requested ship date from lead time, transport days and vendor ship calendarShip dates can be calculated in the past, with a parameter deciding whether to allow or warnRequested and confirmed dates per line and the parameter setting
Dynamics 365 confirmed PO with changesOriginal and new quantity, original and new confirmed receipt dateA changed confirmed PO returns to Approved until reconfirmed; direct downstream impact is shownChange list, linked downstream orders, reconfirmation date
Dynamics 365 vendor collaborationVendor response: accept, reject, or accept with changes to dates or quantitiesChanges create a new PO version; the previously confirmed version remains listed until the new one is confirmedVendor confirmation history with each version and response

What should the dated commitment and recovery decision dataset include?

One record per change event, with child rows for each affected customer commitment. Every component carries its own timestamp and a pointer to the source record, so a reviewer can replay the decision. The structure below is a template a team can adopt as is; it reports no results.

What should the dated commitment and recovery decision dataset include?
ComponentFieldsSourceAcceptance check
CommitmentCustomer order ID, line, promised date, promise version, promise method, set bySales order and its historyPromise version predates the change event
Supply assumptionPO line, item, specification revision, confirmed receipt date, lead time value usedPurchase order and item masterLead time value is the one in force at promise time
Change noticeNew date or quantity, stated reason, channel, received timestamp, vendor response typeVendor portal, EDI message, emailReceived time recorded separately from entry time
Impact linkAffected customer lines and how they were linked: pegging, marking, projectPlanning output or reservationsEach affected line has one documented link
ConstraintsApproved sources for the spec revision, substitutes allowed, partial shipment rule, customer delivery windowApproved supplier list, customer termsEach constraint is dated and cited
Options consideredWait, expedite, split, alternate source, substitute, reallocate stockPlanner notes, ticketsRejected options carry a reason code
DecisionChosen option, decider role, authority basis, decision timestampApproval or ticket recordDecision follows the change notice
Customer noticeNotice timestamp, revised date offered, consent or cancel responseCRM activity, emailNotice precedes the original promised date, or the gap is labelled
OutcomeActual receipt, actual shipment, customer acceptance or cancellationReceipts, shipments, invoicesOutcome dates exist for closed cases
Master data follow-upWhether and when the vendor lead time was correctedItem or vendor master historyYes or no with date

Which constraint makes the fastest or cheapest recovery invalid?

The obvious recovery is often the wrong one, and the dataset is valuable precisely when it records why. A model that sees only the chosen option learns to imitate it; a model that sees the rejected options with their reasons learns the constraint. Capture the constraint at the time of the decision, not as it stands today.

These are the patterns that most often disqualify the attractive option:

Which constraint makes the fastest or cheapest recovery invalid?
OptionWhy it looks attractiveWhen it is invalidEvidence to record
Alternate supplierShorter quoted lead timeThe supplier is not approved for the current specification revisionApproved source list entry and spec revision on the decision date
Substitute itemAvailable in stock nowThe customer's specification or drawing names the original partCustomer specification reference and substitution rule
Split shipmentShips most of the order on timeThe customer requires ship-complete or the items only work togetherShip-complete flag or term on the order
Reallocate stockFixes this order immediatelyThe stock is reserved for another customer's earlier promiseReservation record and the other promise date
Wait silentlyNo extra handlingThe original promise will pass before supply arrives and the customer has not agreed to a new dateCustomer notice and response timestamps

When must the customer hear about the delay, and what proves they did?

For orders taken by mail, internet or telephone in the United States, the FTC's rule sets a clear sequence. If a seller cannot ship within the time it stated, it must offer the buyer the choice to consent to a delay or cancel; for a definite revised date 30 days or less past the original, silence can count as consent. If the seller then misses the revised date, a renewed notice is needed, and silence no longer counts as consent. When the buyer cancels or does not respond in the required cases, the seller must cancel and refund.

Business-to-business orders are also shaped by the contract and the parties' terms, and the evidence a dataset needs is similar: when the seller knew, when it told the buyer, what new date it offered, and how the buyer answered. Record those four timestamps for every affected customer line. Whether a particular notice met a legal requirement is a question for counsel, not a label the dataset should assign.

Worked example: a molded part slips by nine days after the promise (illustrative)

This example is invented to show the template; all counts and durations are illustrative. A distributor promises a customer 40 units of a molded housing for delivery in 21 days, based on a vendor lead time of 14 days and 3 days of transport. Six days later, the vendor accepts the purchase order with changes, moving its confirmed receipt date 9 days later and citing a tooling repair.

The change event links the PO line to three customer orders through planning output. For the first order, the planner considers an alternate molder with a shorter lead time and rejects it because that molder is not approved for revision C of the part. She considers shipping 25 units from stock reserved for a second customer and rejects that because the second customer's promise is earlier. The decision is to notify the customer the same day with a revised date 9 days later and to ask whether a partial shipment of 15 available units is acceptable.

The customer accepts the partial shipment and the revised date. The outcome shows 15 units shipped on the original date and 25 units shipped on the revised date, and the master data follow-up shows the vendor lead time raised from 14 to 18 days 30 days later. Every one of those facts is a separate, dated row; none of them is visible in the final sales order.

What if the promise was never feasible in the first place?

Not every late order is a supplier correction. If the original promise was made against a default sales lead time that ignores stock and planned supply, or if someone overrode an expected-date warning, the delay began at the promise, not at the supplier. Dynamics 365 documents that the sales lead time method uses a default number of days and does not consider stock availability, known demand or planned supply; Odoo shows a warning when a delivery date is set earlier than the expected date.

Label these cases differently: promise infeasible at creation, promise override, and supplier change after a feasible promise. Mixing them teaches a model that suppliers cause delays that were in fact created at order entry, and it hides the most correctable failure in the process.

How do you check the event set against the source systems?

Reconcile before anyone trains or evaluates on the set. Count purchase order changes after confirmation in the source system for the period and compare them with the change events extracted; a large gap usually means changes arrived by email and were keyed without a vendor response record. Confirm that every customer notice has a timestamp later than the change notice it answers, and that no promise version was rewritten in place.

Agree on units before computing any slip. Lead times may be stored in calendar days or working days, and a vendor ship calendar or a transport days table can move a calculated date without any message from the supplier. Compute every slip in one unit, record which calendar was applied, and flag cases where a date moved because a calendar or transport setting changed rather than because the supplier sent a new date.

Then sample by hand. Pick a few cases from each label, open the source records, and walk the chain from commitment to outcome. Supplier names, prices and customer identities usually need pseudonymizing, and supplier agreements may restrict sharing their quoted lead times or terms; check those contracts before the set leaves the company.

Can a distributor license its commitment and recovery history?

Distributors and manufacturers that kept purchase order versions, vendor responses and customer notices hold a record of planning decisions under pressure that is hard to reproduce synthetically. A distributor that chooses to license that history can do so through SourceX, which manages the rights review, the contract, data preparation, delivery and payment.

Qualification comes first, through a brief questionnaire about the business with no order data attached, and no record is handed over until the company has signed off on scope and terms and signed the agreement.

Frequently asked questions

Should the original promised date be overwritten when a supplier delays?

No. Keep the original promise as its own version and add the revised date as a new version with a timestamp and reason. Overwriting erases the event that shows a commitment going wrong, which is the part of the history an AI system learns from.

What counts as the time the business learned about the change?

The time the supplier's message or portal response was received, which is often earlier than the time a buyer keyed it into the ERP. Record both. The gap between them is itself useful information about how the process works.

How do you link a purchase order delay to the customer orders it affects?

Use the system's own dependency records, such as planning pegging, inventory marking or project links, and record which one was used. Dynamics 365's confirmed purchase order changes workspace, for example, shows direct downstream impacts but notes that indirect impacts are not yet considered.

Is a late shipment always a supplier lead time problem?

No. Separate cases where the promise was infeasible when it was made, cases where someone overrode a system warning, and cases where a supplier changed a confirmed date after a feasible promise. Each teaches a different lesson.

Does the FTC delay notice rule apply to business customers?

The FTC rule covers merchandise ordered by mail, internet or telephone, and its business guide describes the delay option notices sellers must send. Business orders are often governed by negotiated terms as well, so counsel should confirm which obligations apply to a given order.

Sources

  • Odoo computes an expected delivery date from delivery time and sales security lead time, defines vendor lead time from PO confirmation to receipt, and warns when a set date is earlier than expected. Source
  • Dynamics 365 order promising supports sales lead time, ATP and CTP methods, and recalculates a changed order's promise only when the current date cannot be met. Source
  • Dynamics 365 returns a changed confirmed purchase order to Approved and lists original and new confirmed receipt dates with direct downstream impact. Source
  • In Dynamics 365 vendor collaboration, vendors can accept with changes to dates or quantities, and changes create new purchase order versions with a confirmation history. Source
  • Dynamics 365 calculates requested ship dates from the requested receipt date, lead time, transport days and vendor ship calendar. Source
  • The FTC business guide explains delay option notices, when silence counts as consent, renewed notices and required cancellations under the mail, internet or telephone order rule. Source

See if your company qualifies

A short company assessment. No data uploads are needed.

See if your company qualifies