If you work in automotive or any industry that supplies to it, you
have probably lived through APQP. Maybe you called it “new product
introduction.” Maybe you called it “that giant binder of deliverables
the customer wants before sign-off.” Either way, Advanced Product
Quality Planning was supposed to be the framework that guaranteed
quality at launch — not after launch, not during production, but baked
in from the very first concept sketch.
Here is what usually happens instead: a cross-functional team kicks
off the project with enthusiasm. Phase one deliverables get done on
time. Then the timeline compresses. Engineering changes pile up.
Supplier parts arrive late. Suddenly, the team is racing through the
later phases, checking boxes to hit the launch date rather than using
each gate as a genuine quality decision point. The PPAP submission goes
out with a mix of real data and hopeful assumptions. Production starts.
Problems surface within weeks. And everyone sits in a war room wondering
why the same launch failures keep recurring.
This article is about why APQP fails in practice — and what it
actually takes to make it work.
What APQP Was Designed to Do
APQP was developed by the automotive industry, specifically the Big
Three (GM, Ford, Chrysler), in the 1980s and early 1990s. It was
formalized through AIAG (Automotive Industry Action Group) and has since
been adopted globally. The core idea is deceptively simple: plan quality
upstream so that by the time you reach production, the product and
process are proven capable.
The framework divides new product development into five phases:
- Plan and Define Program — Translate customer
expectations (voice of the customer) into specific design goals,
reliability targets, and quality objectives. - Product Design and Development — Engineer the
product to meet those targets. Assess feasibility. Conduct design FMEA.
Build and test prototypes. - Process Design and Development — Design the
manufacturing system that will produce the product consistently. Conduct
process FMEA. Develop control plans. - Product and Process Validation — Run production
trials at volume. Validate capability. Confirm that the process produces
conforming product under realistic conditions. - Launch, Feedback, Assessment, and Corrective Action
— Begin serial production. Measure performance. Close gaps. Feed lessons
learned back into the system for the next program.
Each phase has defined inputs and outputs. The intent is that you do
not proceed to the next phase until the deliverables of the current one
are genuinely complete and reviewed.
Where APQP Breaks Down
The framework is sound. The failure is almost always in execution.
Here are the patterns I have seen across dozens of programs over 25
years:
Timeline Pressure
Overrides Quality Logic
This is the most common failure mode. APQP is supposed to follow a
logical sequence — design verified before tooling committed, process
validated before production approved. But program timelines are often
set by commercial priorities, not engineering reality. When the customer
says “we need parts by Q3,” every gate review becomes a rubber stamp.
Phase three deliverables get compressed into a fraction of the time they
need. Phase four trials run once, fail partially, and get signed off
anyway because there is no time for a second attempt.
The result: you enter production with unresolved risks. Those risks
do not disappear. They show up as scrap rates, customer complaints, and
warranty claims within the first few months — exactly what APQP was
designed to prevent.
FMEA Treated as
Paperwork, Not Thinking
Design FMEA and Process FMEA are central APQP deliverables. Done
well, they force the team to systematically identify potential failure
modes and address them before they occur. Done poorly — which is most of
the time — they are retrospective documents filled out after the design
is frozen, listing risks everyone already knows about with RPN scores
that never drive action.
I have reviewed hundreds of FMEAs in my career. The pattern is
remarkably consistent: the severity-occurrence-detection ratings are
calibrated to produce acceptable RPNs, not to reflect actual risk.
High-severity, high-occurrence failure modes get detection scores of 2
or 3 (almost certain to catch) when the actual detection mechanism is a
visual inspection by a tired operator at end of line. The FMEA says
“risk addressed.” The reality says “risk deferred.”
Supplier Development Left
Too Late
Modern products depend on dozens of suppliers. APQP requires that
supplier parts are validated to the same standard as in-house
components. In practice, supplier APQP status often becomes a blind
spot. Tier 2 and Tier 3 suppliers submit PPAP packages that are
incomplete or based on different test conditions. Issues with material
properties, dimensional conformance, or process capability surface
during phase four — sometimes during launch — when corrections are
expensive and slow.
Strong programs assign supplier quality engineers to monitor Tier 1
APQP progress monthly and audit critical Tier 2 suppliers directly. Weak
programs send a spreadsheet asking for status updates and accept
whatever comes back.
Cross-Functional
Involvement That Isn’t
APQP demands involvement from design engineering, manufacturing
engineering, quality, purchasing, and supply chain. In theory, these
functions collaborate throughout the program. In reality, each function
works in its own silo and presents finished work at gate reviews.
Manufacturing engineering sees the design for the first time at phase
two gate. Quality sees the control plan at phase three. Purchasing has
already committed to suppliers based on price, not capability.
Genuine cross-functional collaboration means manufacturing and
quality are involved in concept reviews. It means purchasing sits in on
design FMEAs to flag supplier feasibility concerns. It means the team
operates as an integrated unit, not a relay race where each runner hopes
the previous one did their job.
What Effective APQP Looks
Like
I have also seen programs where APQP worked — where launches were
smooth, capability was proven, and early production matched
expectations. Those programs shared certain characteristics:
Gate reviews with teeth. The phase gate is not a
status update meeting. It is a decision point. If the design
verification testing is incomplete or if critical-to-quality
characteristics are not capable, the gate does not close. This requires
leadership support — someone with authority to say “we are not ready”
and enforce it, even when the timeline is at risk.
FMEAs that drive action. A meaningful FMEA
identifies failure modes that the team genuinely had not considered. It
generates action items — design changes, process controls, additional
testing — that are tracked to completion. The document is updated when
new information arrives during testing or trials. It is a living tool,
not a one-time deliverable.
Realistic timelines. Programs that build quality in
allocate sufficient time for each phase, including buffer for iteration.
If the design verification fails (and it often does on the first pass),
there must be time to investigate, modify, and re-test before committing
to tooling. Compressed timelines eliminate this learning loop.
Supplier quality integrated from day one. Critical
suppliers are selected during phase one, not phase three. Their APQP
progress is monitored with the same rigor as internal work. Site visits,
capability studies, and trial runs at supplier facilities happen in
parallel with internal development — not as an afterthought two weeks
before launch.
Lessons learned actually applied. Phase five of APQP
includes capturing lessons learned and feeding them into future
programs. Most organizations skip this or reduce it to a bullet on a
closing checklist. Organizations that take it seriously maintain a
structured database of launch issues, root causes, and preventive
actions. New programs start by reviewing relevant lessons from prior
launches. This single practice can prevent repeating the same mistakes
across successive programs.
APQP in Non-Automotive
Industries
While APQP originated in automotive, its principles apply to any
complex product introduction. I have helped adapt the framework for
medical device manufacturers, aerospace suppliers, and industrial
equipment builders. The phases translate well:
| APQP Element | Automotive | Medical Device | Industrial Equipment |
|---|---|---|---|
| Phase 1 | Voice of customer, design goals | User needs, design inputs | Customer specs, performance targets |
| Phase 2 | Design FMEA, prototyping | Design controls, verification | Concept testing, feasibility |
| Phase 3 | Process FMEA, control plan | Process validation, IQ/OQ/PQ | Manufacturing plan, tooling design |
| Phase 4 | Production trial run, PPAP | Process qualification, validation lots | Site acceptance test, commissioning |
| Phase 5 | Launch, lessons learned | Post-market surveillance | Warranty tracking, continuous improvement |
The terminology shifts, but the underlying logic — plan quality
early, verify at each stage, validate before commitment — is
universal.
The Cost of
Skipping APQP (Or Doing It Poorly)
Organizations that treat APQP as bureaucracy rather than engineering
discipline pay a predictable price. Here is what the data consistently
shows across manufacturing sectors:
- Launch scrap and rework costs are typically 5-10x
higher for programs that compressed or skipped phase three and four
activities. - Warranty claims in the first 12 months correlate
strongly with the thoroughness of design FMEA and validation testing
during APQP. - Time to stable production — the period from launch
to achieving consistent Cpk > 1.33 — is 2-3x longer when process
capability was not proven during phase four. - Engineering change volume in the first six months
of production is a direct indicator of upstream APQP quality. Programs
with disciplined phase gates typically issue 60-70% fewer post-launch
changes.
The irony of skipping APQP to save time is that it costs more time
overall. A launch delay caused by thorough validation is measured in
weeks. A launch that proceeds on schedule but fails in production causes
months of firefighting, customer escalations, and recovery efforts that
dwarf the original time savings.
Digital Tools and APQP
A note on technology: many organizations have invested in APQP
software platforms — cloud-based systems that manage deliverables, track
action items, and provide dashboards. These tools can help with
visibility and version control. They cannot substitute for the
engineering judgment and cross-functional engagement that make APQP
effective.
I have seen companies implement sophisticated PLM (Product Lifecycle
Management) systems and still produce poor launches because the team’s
behavior did not change. The software tracked that a design FMEA
existed. It did not track whether the FMEA was any good. The dashboard
showed green status on all phase three deliverables. It did not show
that the control plan was copied from a previous program without
updating for the new product’s characteristics.
Tools support process. They do not replace thinking.
Building APQP Discipline
If your organization struggles with APQP — and most do — the path
forward is not a new template or a different software system. It is
cultural and operational:
Start with gate reviews. Make them meaningful. Require evidence, not
assertions. If a deliverable is not complete, the gate stays open. This
will cause discomfort the first few times it happens. Leadership must
support the decision to hold gates, or the entire system collapses.
Invest in FMEA capability. Train your engineers not just on the
mechanics of the form, but on the thinking behind it. Use
cross-functional reviews where people challenge each other’s
assumptions. Bring in external facilitators if internal reviews have
become echo chambers.
Integrate suppliers. Treat critical suppliers as extension of your
team. Share APQP expectations early. Monitor progress actively. Build
relationships, not just transactions.
And perhaps most importantly: close the loop. Every launch produces
lessons. Capture them. Categorize them. Build them into the next
program’s APQP checklist. Over time, this creates an organizational
memory that prevents the repetition of avoidable failures.
Final Thought
APQP is not complicated. Five phases, clear deliverables, logical
sequence. The difficulty lies not in understanding the framework but in
committing to it when timelines compress and commercial pressure mounts.
The organizations that launch successfully are not the ones with the
most sophisticated APQP documentation. They are the ones that use the
framework as it was intended — as a decision-making process that quality
comes before schedule, because quality problems always find the schedule
eventually.
I have spent over 25 years helping manufacturers implement APQP that
works in practice, not just on paper. The companies that get it right
are not smarter than the ones that do not. They are simply more
disciplined. And in quality engineering, discipline is the difference
between a smooth launch and six months of firefights.
About the Author: Peter Stasko is a Quality
Architect with over 25 years of experience in manufacturing quality,
process optimization, and product introduction. He has led APQP programs
across automotive, industrial, and medical device sectors, helping
organizations move from checkbox compliance to genuine quality
engineering. Peter writes about the realities of quality management —
what works, what fails, and what to do about it.
