Products change constantly: a cheaper supplier, a cracked bracket, a new regulation, a finish being retired. Engineering change is how a company makes those changes without losing track of what it's building today.
The three documents
| Purpose | Key question | |
|---|---|---|
| ECR (request) | Describe a problem and a proposed change | Should we change this? |
| ECO (order) | The approved change, with effectivity and disposition | What changes, from when, and what happens to what we have? |
| ECN (notice) | Tell everyone affected | Who needs to know? |
The part that takes the time
Approving a change is quick. Working out its impact isn't. A good change impact analysis starts with where-used: every assembly, configured product and rule that touches the part. Then it asks about open orders, stock on hand and on order, supplier lead times, documents and every downstream system, from CPQ to dealer price files.
At MillerKnoll I released more than 120 engineering changes a month. Speed came from boring things: a standard intake form, a fixed approval sequence, and prioritising by order value and risk instead of arrival time.
Where it goes wrong
- The ECO is released but the configurator isn't updated, so the old part is still sold.
- Effectivity is set to “immediately” with open orders still in the queue.
- Nobody owns the retirement half of a replacement: the new part arrives, the old one never leaves.
Questions people ask
Is an engineering change only for engineering?
No. Every change touches sales, purchasing, the plant, service and often dealers. That's why the ECN exists.
How do you speed up change without adding risk?
Standardise intake, sequence approvals, sort by impact rather than arrival, and make the configurator update part of the change, not a follow-up task.
Updated November 2026. Think I've got something wrong? Tell me.