Common information failures: wrong data, late updates and version confusion

Most order information problems are one of three patterns, and telling them apart decides the fix

This page is for anyone who has ever been told a number or a document was "wrong" and had to work out what actually happened. A merchandiser, a production planner, or a buyer spends much of the working day reading documents other people wrote: a tech pack, a purchase order, a costing sheet, a bill of lading. Most of the time these documents agree with each other. Sometimes they do not, and the order stops while someone finds out why.

At work, this is where an information failure costs real money: a cutting room starts work from a document that turns out not to be the one everyone should have used. Without knowing which pattern caused the problem, people argue about who typed something wrong, when often nobody did. This lesson names the three patterns this course tracks, wrong data, late updates, and version confusion, and gives you a method for telling them apart before you spend time fixing the wrong thing.

You need: lesson 9.1's idea that information about an order has to move between people and companies as reliably as the goods do. No calculation background is needed.

By the end of this lesson you can:

• name the three common information-failure patterns this course tracks;
• tell wrong data apart from a late update by asking whether the original information was ever correct;
• recognise version confusion, where two people each hold a different, both-plausible copy of the same document;
• explain why Cedar & Co., without PLM or ERP software, is especially exposed to version confusion;
• pick the correct fix for each pattern, instead of applying the same fix to all three.

Not covered here: what PLM and ERP systems each hold (lesson 9.2), how a tech pack's specification reference connects to a purchase order (lesson 9.3), and how a smaller company keeps information reliable without dedicated software (lesson 9.5).

Words you will meet in this lesson

Most order information problems fall into one of three patterns, and each needs a different fix

An order document can go wrong in three different ways, and they are not the same problem. Wrong data means the number or fact itself was never correct, someone entered it incorrectly at the source. Late update means a correct change was made, but it had not yet reached everyone who needed it when the problem showed up. Version confusion means two people each have a document that looks complete and correct, but the two documents are different revisions, and neither person knows the other's copy is out of date. Telling these three apart matters because the fix is different for each one: correct wrong data at its source, push late updates out faster, and give version confusion one clear current copy that everyone checks against.

The same failure pattern shows up across every kind of order document

None of these three patterns is rare, and a person new to the trade usually meets all three within the first few months. The table below shows the shapes each pattern actually takes, and where a smaller domestic brand carries more risk than a company running dedicated systems.

Order A's tech pack goes through revision 3 and revision 4, and Delta Garments' cutting room works from the older one

Order A's tech pack goes through several sample-round revisions as Northline Apparel and Delta Garments work through the fit and pre-production samples (lesson 5.2). Suppose Delta Garments' cutting room starts bulk cutting from revision 3 of the tech pack, while Northline Apparel's latest approval, sent by email a few days earlier, was for revision 4. Neither revision 3 nor revision 4 is a wrong document; revision 3 was correct when it was issued, and revision 4 is correct now. The problem is that revision 3 is no longer current, and the cutting room did not know that.

This is version confusion, not wrong data. Nobody typed an incorrect measurement into either revision. The tech pack itself is not at fault; what failed is that Delta Garments' cutting room had no single place to check which revision was current before cutting 15,000 pieces of fabric against it.

The physical-stage cost of a late catch is the same, whichever of the three patterns caused it

Order A's costing shows fabric at USD 1.15 per piece, and trims add USD 0.20. Separately, making the piece, meaning cutting and sewing, costs USD 1.10. Together these three cost components represent everything invested in a T-shirt by the time it has been cut and sewn, the risk scale for a mistake caught only at that stage.

Whether the cutting room worked from a wrong number, a late update, or the wrong revision, catching the failure only after bulk cutting risks the same USD 36,750. The pattern changes what caused the mistake and what prevents it happening again; it does not change the cost of catching it late at this stage.

Follow three questions, in order, to name the failure pattern correctly

1. Ask whether the original information was ever correct. Check the number or fact against its source at the time it was first entered. If it was wrong from the start, stop here: this is wrong data.
2. If the original information was correct, check whether a later, correct update had been made. Look for a change that was approved but had not yet reached every document or every person who needed it. If you find one, this is a late update.
3. If neither of the above applies, check whether two people are working from different, both-plausible copies. If both copies look complete and neither is flagged as outdated, this is version confusion.
4. Apply the matching fix, not a generic one. Correct wrong data at its source. Fix a late update by improving how changes are pushed out to every dependent document. Fix version confusion by establishing one clear current copy, the system of record (lesson 9.2), that everyone checks against.

A person treats every information problem as if someone simply typed something wrong

A common mistake is to treat every information problem as wrong data: assuming someone typed a number incorrectly, when the real cause is often a late update or version confusion instead. Correcting a costing-sheet number that was never actually wrong, while missing that two people are working from genuinely different tech pack revisions, does not fix the real problem. The cutting room in the case above does not need a corrected measurement; it needs to know which revision is current. Applying the wrong fix wastes time and leaves the true cause, an out-of-date copy still in circulation, ready to cause the same failure again on the next order.

• If the original number or fact was never correct, that is wrong data. Fix it at the source.
• If a correct change existed but had not reached everyone, that is a late update. Fix how updates are pushed out.
• If two complete, professional-looking documents disagree, and neither is flagged as outdated, that is version confusion.
• Version confusion needs a system of record, one agreed current copy, not a search for who made a mistake.
• A smaller domestic brand without PLM or ERP software is not more likely to enter wrong data, but it is more exposed to version confusion.
• The cost of catching an information failure late at a given production stage is the same whichever of the three patterns caused it.

Two cutting-room supervisors at Delta Garments each insist their own tech pack copy is correct

Two cutting-room supervisors at Delta Garments are preparing to cut Order A. Each one has a tech pack copy on their desk, and each insists their own copy is correct. Both copies look complete: both carry Northline Apparel's letterhead, both are professionally formatted, and neither is marked as outdated.

You are the merchandiser at Delta Garments. Does one of the two supervisors have to simply be wrong? Decide before you open the answer.

Information map sheet: add a failure-pattern note where Cedar & Co.'s documents disagree

The Information map sheet lists every document Order A produces. Imagine a similar mixup at Cedar & Co., a domestic brand without PLM or ERP software. Add a note against each row touched, naming the failure pattern.

Download the workbook, open "Information map", and add your own note for a real order.

Download the Information map sheet (Excel)

Lesson 9.5: Keeping information reliable without a full system

You can now name the three information-failure patterns and tell them apart. The next lesson looks at what a smaller company without PLM or ERP software, like Cedar & Co., can still do to keep its information reliable. Back to the course map