Change is constant in electronics manufacturing. Components reach their end of life, regulations tighten, costs need trimming, and customers ask for improvements. The question is not whether you will make engineering changes. It's whether you can make them without introducing scrap, delay, or risk. Get change control right, and it becomes a source of resilience. Get it wrong, and it erodes margin, quality, and trust.
What is an engineering change note, and why does it matter?
An engineering change note is the controlled document that confirms an approved change and tells everyone involved to implement it. It is the definitive communication between you, the design authority, and your EMS partner. Once you outsource assembly, verbal instructions to production staff are no longer an option. The ECN becomes the single source of truth for what changes, when, and how.
We were taught years ago that a change is only an ECN if it affects the form, fit, or function of a product. That is a fine starting definition. But in practice, a manufacturing operation often uses the ECN process for more than that, including changes to the manufacturing or test process and the use of alternative components. That breadth is a feature, not a bug. The ECN records on file become a record of the product's life history and the processes that shape it, which is invaluable when you need to investigate a quality issue years later.
ECR vs ECN vs ECO: what's the difference?
An ECR proposes a change, an ECO authorises it, and an ECN communicates that the change is confirmed and must be implemented. The three documents form one workflow, but the industry uses the terms inconsistently, and smaller organisations often combine the ECO and ECN into a single document. This inconsistency is where OEMs most often get caught out, so agree on the vocabulary with your EMS partner before you raise your first change.
| Term | What it is | Who raises it | What it triggers | Where OEMs get it wrong |
| ECR (engineering change request) | A proposal to change a product or process, with the reason and justification | Anyone: engineering, quality, production, procurement, or the customer | Review by stakeholders and a decision to proceed or not | Treating an ECR as approval to build, before impact is assessed |
| ECO (engineering change order) | The authorised instruction that defines and approves the change, with all affected items and documents | Engineering, approved by a change review board | Implementation planning, updated drawings, BOMs, and work instructions | No formal approval routing, so changes proceed without sign-off |
| ECN (engineering change notice / note) | The notification that the change is confirmed and must be executed | Engineering or document control, issued to all affected teams | Execution on the line, in the supply chain, and in documentation | Sending it by email, with no acknowledgement or defined effectivity |
A concession is different. An ECN is usually a permanent change once implemented, whereas a concession should be temporary, covering a set period, batch, or range of serial numbers. A good EMS partner can manage both, but you must be clear from the outset which one you need.
What does a robust ECN actually need to contain?
A robust ECN removes ambiguity. It states exactly what changes, on which parts, from when, and what happens to material already in the system. If any of that is buried in a BOM or a PCB file rather than spelt out, it may be missed, and the whole build can end up starting again from scratch. That is when full engineering and tooling charges apply, instead of the modest or zero cost of a well-documented change.
A complete ECN should capture:
- Revision control: The up-issued drawing and part number, linked to the ECN for traceability. This is why revision level control matters so much in a manufacturing data pack.
- Affected part numbers and assemblies: Every item touched, not just the headline part.
- Effectivity: When the change applies. Immediate, from a date, from the next batch, or on run-out of existing stock (use-on-code).
- Disposition of existing stock and work in progress: Use as is, rework, scrap, or return.
- Requalification requirements: Retest, revalidation, or first-article inspection where needed.
- Documentation and drawing updates: Up-issuing drawings so no one builds to a superseded specification.
- Approval routing: Who signs off, in what order, with acknowledgement recorded.
The last point matters more than it might seem. An ECN should never live only in an email that can be deleted or lost in a junk folder. Your partner must be able to demonstrate how they capture acknowledgement and feedback once the change is complete, or you may not get the product you expected.
What does poor change control actually cost?
Poor change control is expensive, and the cost compounds the later a problem surfaces. The direct hits are scrap and rework, obsolete inventory write-offs, line-down events, and missed ship dates. The indirect hits are worse: field failures, warranty exposure, and regulatory nonconformance that can pull a product from the market.
There is a well-established principle here, usually called the rule of ten. It traces back to Barry Boehm's 1970s research on software defects, which found that a defect caught during requirements was trivial to fix, whereas the same defect caught after release could cost up to 100 times more.
In product lifecycle management, as engineering software firm CENIT AG puts it, the rule of ten says that "with each successive product lifecycle phase, the effort required to implement a change increases by a factor of ten." Treat it as a strong qualitative rule rather than a hard number, because the exact multiplier varies by part and product. The direction of travel is not in doubt: a change that costs very little to make on a drawing can cost a fortune once tooling is cut, stock is bought, or product is in the field.
Why is change control becoming harder?
Change control is harder now because the supply chain is generating more forced changes, while the tolerance for error has fallen. Three pressures stand out:
1. Component availability
According to independent component distributor GlobX, the tightest categories in 2026, including "power management ICs, MLCCs, MCUs, memory modules (DRAM/NAND), IGBTs, MOSFETs, and automotive-grade semiconductors," have lead times "ranging from 26 to 52+ weeks."
Much of the pressure is tied to AI-server power design and rising material costs, with demand for high-end MLCCs described by Holy Stone Enterprise chairman Bill Tang as reaching "levels not seen in more than two decades."
Commodity parts have largely normalised, but constrained categories such as high-capacitance and automotive-grade MLCCs, power semiconductors, and some memory remain tight, with Murata's new high-end MLCC capacity not expected before Q4 2026. Every constrained part is a candidate for a resourcing change, and every resourcing change needs an ECN.
2. Obsolescense
Hundreds of thousands of components reach end-of-life every year, and a large share arrive with no warning. Per Z2Data's 2024 Obsolescence Trends report, of the 473,910 parts that reached end-of-life in 2023, 142,173 (around 30%) "had no PCNs issued by manufacturers."
Z2Data's later analysis found the gap widened further, with less than half of the 621,909 parts obsoleted in 2025 accompanied by a product change notification. No notice means no planning window. Each substitution that follows is an engineering change that must be controlled.
3. Trade policy
Under the USTR Section 301 modification issued in September 2024 (Federal Register notice 2024-21217), tariffs on key semiconductor categories doubled "from 25% to 50%, effective January 1, 2025," and further action is scheduled: USTR's 23 December 2025 determination adds new Section 301 semiconductor tariffs starting at 0% and "increasing in 18 months on June 23, 2027."
Tariffs shift sourcing decisions, and resourcing driven by cost or trade policy, rather than design intent, is now a significant source of engineering changes. When a change is driven by economics rather than engineering, the discipline of the ECN is what stops a commercial decision from becoming a quality problem.
Nearshoring, co-shoring, and dual-sourcing are the strategic response. The EU Chips Act and European Commission proposals to mandate diversified chip sourcing are pushing OEMs toward a more regional, multi-source footprint. Diversifying sources is sensible, but it multiplies the number of qualified configurations you must control, which raises the stakes on change governance rather than lowering them.
How does a capable EMS partner de-risk change?
A capable EMS partner makes change a controlled, repeatable process. The tools are PLM and ERP integration, controlled document flow, revision-locked BOMs, component engineering, traceability to batch or serial level, disciplined stock disposition, and the ability to replicate a change consistently across multiple sites. During the move from NPI to volume, strong change governance keeps the ramp on schedule.
ESCATEC applies this in practice. Our supply chain team advises on evaluating and selecting components, and we have significant expertise in obsolescence management, using early warning processes to identify at-risk parts and perform obsolescence risk analysis before a part is discontinued.
We keep bills of materials accurate and complete using audit and analysis tools that surface component lifecycle information. Our counterfeit avoidance processes draw on industry guidelines, backed by a dedicated component test laboratory.
Our NPI process is a phased, cross-functional route from prototype to production, with defined PFMEAs, line parameters, quality control plans, and a supply chain risk plan covering alternates, obsolescence, and lead-time risks, informed by a Design for Excellence (DfX) approach from the earliest stages.
As a privately owned EMS partner with global sites in the UK, Malaysia, Switzerland, the Czech Republic, and Bulgaria, we support nearshoring, co-shoring, and offshoring, so change can be governed and replicated wherever your product is built.
Conclusion
Raising an ECN takes time, but skipping it costs far more. A robust change process protects your margin, your quality record, and your reputation, and in a period of constrained supply and tighter regulation, it is fast becoming a competitive advantage rather than a chore.
If you are reviewing how change is managed across your product range, or weighing up an EMS partner, it's worth asking hard questions about change control before you commit. If you'd like to talk it through, our team is happy to share how we do it.
FAQs
What is the difference between an ECR, an ECN, and an ECO?
An ECR proposes a change, an ECO authorises it, and an ECN confirms the change and instructs teams to implement it. Many organisations combine the ECO and ECN into one document, so agree definitions with your partner before you start.
Is an ECN the same as a concession?
No. An ECN is usually a permanent change once implemented, whereas a concession is temporary, covering a set period, batch, or range of serial numbers.
What should an engineering change note include?
Revision level, affected part numbers, effectivity, disposition of existing stock and work in progress, requalification requirements, documentation and drawing updates, and approval routing with recorded acknowledgement.
What does poor change control cost?
Scrap, rework, obsolete inventory write-offs, line-down events, missed ship dates, field failures, warranty exposure, and regulatory nonconformance. The cost of fixing a problem rises by roughly a factor of ten at each later lifecycle stage
Editor's note: This blog was first published in 2015 and has been updated in August 2026 for relevance and accuracy.
