The
Number Every Executive Wants and Nobody Can Agree On
Here is a conversation that has played out in quality management
conferences, boardrooms, and plant floors for the last sixty years. The
CFO turns to the Quality Director and asks a reasonable question: “How
much does poor quality actually cost us?” The Quality Director, who has
been waiting for this moment, clears their throat and says something
like “industry studies suggest the cost of quality ranges from fifteen
to twenty percent of revenue.” The CFO’s eyes widen. The Operations VP
shifts uncomfortably. And then, predictably, nothing happens. The number
is too big to believe, too vague to act on, and too politically loaded
to survive contact with departmental budgets.
This is the fundamental problem with Cost of Quality analysis. Not
the concept — the concept is sound, well-established, and genuinely
useful when applied with intellectual honesty. The problem is that
almost every organization that attempts to measure quality costs
produces a number that serves a narrative rather than informing a
decision. The quality department inflates failure costs to justify its
budget. The production department deflates them to protect its
performance metrics. Finance discounts the whole exercise because the
numbers don’t tie to the general ledger. And the executive team, sensing
that nobody is telling the unvarnished truth, files the report and moves
on.
Meanwhile, the actual costs of poor quality — the rework that
everyone has normalized, the scrap that shows up as a line item called
“yield adjustment,” the customer complaints that the sales team absorbs
as “relationship management,” the engineering change orders that pile up
because nobody validated the design before committing to tooling — these
costs continue to accumulate, invisible, unmeasured, and completely
unmanaged. The organization pays for them every single day. It just
doesn’t know it, because nobody has built the infrastructure to see them
clearly.
The PAF
Model: Simple in Theory, Dangerous in Practice
The traditional framework for Cost of Quality is the PAF model, which
divides quality costs into four categories. Prevention costs are what
you spend to avoid defects — training, process design, quality planning,
supplier qualification. Appraisal costs are what you spend to detect
defects — inspection, testing, audits, calibration. Internal failure
costs are what you spend when you find defects before they leave your
facility — scrap, rework, re-inspection, downtime. External failure
costs are what you spend when defects reach your customer — warranties,
returns, recalls, lost business, legal liability, reputational
damage.
The theory, articulated by Joseph Juran in the 1950s and refined by
Armand Feigenbaum, is elegant and still valid. The optimal quality cost
curve suggests that as you invest more in prevention, your failure costs
decline at a faster rate, producing a total cost curve with a minimum
somewhere well to the right of where most organizations currently
operate. The strategic implication is clear: most companies underinvest
in prevention and overpay for failure, and shifting the balance toward
prevention produces net savings.
So far, so good. The danger lies in what organizations actually do
with this model, which is usually one of three things.
First, they attempt to assign a precise dollar figure to every
quality-related activity in the organization, a exercise that sounds
rigorous but produces numbers of such questionable lineage that they
undermine their own credibility. Is the quality engineer’s salary a
prevention cost or an appraisal cost? Is the production operator’s time
spent on in-process inspection a production cost or an appraisal cost?
Is the customer service representative handling a complaint a sales cost
or an external failure cost? These allocations require judgment calls
that different stakeholders will make differently, and the resulting
number becomes a subject of debate rather than a basis for action.
Second, they focus exclusively on the costs that are easy to measure
— typically internal scrap and rework, which show up in production
reports — while ignoring the costs that are hard to measure but
potentially much larger, such as the revenue lost from customers who
quietly switched suppliers after a single bad experience without ever
filing a complaint. The resulting analysis systematically understates
the true cost of poor quality and therefore undervalues investment in
prevention.
Third, and most damaging, they use the COQ measurement as a weapon in
internal political battles rather than as a diagnostic tool for
improvement. The quality department presents the failure cost number to
demonstrate that production needs to do better. Production counters with
the appraisal cost number to argue that quality is over-inspecting.
Suppliers are scored on metrics they don’t understand and can’t
influence. The exercise generates heat without light.
The
Hidden Factory: Where Your Real Quality Costs Live
There is a concept in quality management, coined by Feigenbaum,
called the “hidden factory.” It refers to the unofficial, unmeasured,
undocumented system of rework, repair, sorting, expediting, and
workaround that exists alongside the official production process in
almost every manufacturing organization. The hidden factory is where
parts go when they don’t conform. It’s where operators spend time fixing
things that should have been right the first time. It’s where engineers
write deviation reports and suppliers send replacement shipments and
expedite fees get buried in freight budgets.
The hidden factory typically consumes between fifteen and forty
percent of total manufacturing capacity, depending on the industry and
the maturity of the quality system. This is not a speculative range — it
is consistently observed in organizations that undertake serious process
mapping and activity-based costing studies. But because the hidden
factory’s activities are distributed across departments, cost centers,
and budget categories, it doesn’t appear as a line item on any financial
statement. It is hiding in plain sight, absorbed into standard costs,
overhead rates, and variances that everyone has learned to live
with.
The most insidious aspect of the hidden factory is that it becomes
institutionalized. Rework routes are formalized in routing instructions.
Sorting operations become permanent headcount. Expediting becomes a job
title. The organization builds infrastructure around its poor quality,
and that infrastructure develops its own constituencies, budgets, and
defenses. When someone eventually suggests that perhaps the rework cell
shouldn’t need six full-time operators, the pushback comes not from
quality or production but from the rework cell supervisor who has built
a career on managing the consequences of a problem nobody is
solving.
This is why Cost of Quality analysis, done well, can be
transformative. Not because it produces a number — any number — but
because it forces the organization to see the hidden factory, measure
its capacity, and calculate its operating cost. Once you can see it, you
can decide whether you want to keep paying for it. Most organizations,
confronted with the true cost of their hidden factory, are genuinely
shocked. They had no idea they were spending that much on doing things
wrong and fixing them afterward.
Activity-Based
Costing: The Bridge Between Quality and Finance
The most practical approach to COQ measurement borrows from
activity-based costing, the management accounting methodology that
assigns costs to activities rather than departments. The advantage of
ABC for quality costing is that it sidesteps the political battles over
departmental allocation by focusing instead on what people actually do
during their workday.
The approach is straightforward in concept, though it requires
discipline to execute. You identify the specific activities that consume
resources due to quality issues — reviewing nonconformance reports,
sitting on material review boards, performing rework, sorting suspect
lots, investigating customer complaints, responding to supplier
corrective action requests, preparing for and hosting quality audits,
writing and revising procedures because the previous version didn’t
prevent a problem. For each activity, you estimate the time consumed and
multiply by the loaded labor rate. You add material costs for scrap and
rework, equipment costs for re-inspection and re-testing, and external
costs for warranties, returns, and field service.
What emerges is not a single number but a categorized picture of
where quality-related effort and expense are concentrated in the
organization. And this picture, almost without exception, reveals a
pattern that contradicts the organization’s stated priorities. Companies
that talk endlessly about quality are typically spending eighty percent
of their quality budget on detection and correction — appraisal and
failure — and less than twenty percent on prevention. They are paying
for quality after the fact rather than investing in it beforehand, and
the more they talk about quality, the more they are usually spending to
compensate for not having it.
The CFO who dismissed the “fifteen to twenty percent of revenue”
number as inflated suddenly becomes interested when you can show that
three operators spend four hours a day sorting parts, that the rework
cell runs two shifts, that the customer complaint investigation process
consumes twenty engineering hours a week, and that expedited freight for
replacement shipments cost a hundred and forty thousand dollars last
quarter. These are concrete, verifiable, tied to the general ledger.
They don’t require belief. They require action.
The
Prevention Investment: Why It’s Harder Than It Sounds
The logical conclusion of any honest COQ analysis is that the
organization should shift resources from failure and appraisal into
prevention. Spend more on design for manufacturability, process FMEA,
operator training, mistake-proofing, supplier development, and
preventive maintenance. Spend less on inspection, sorting, rework, and
warranty. The total cost goes down, and quality goes up.
This conclusion is correct and has been validated in virtually every
longitudinal study of quality economics ever conducted. But executing
the shift is far harder than recommending it, for three structural
reasons.
The first reason is that prevention investments are upfront and
certain, while failure savings are downstream and probabilistic.
Training twenty operators in statistical process control costs money
this quarter. The reduction in scrap and rework that results from better
process control arrives over the next several quarters, may be
influenced by many other factors, and cannot be attributed with
precision to the training investment alone. In a world of quarterly
budgets and ROI calculations, the certain cost consistently wins over
the uncertain benefit.
The second reason is that prevention requires cross-functional
coordination, while failure correction can be handled within a single
department. Training operators is a production responsibility. Designing
training content is a quality responsibility. Validating training
effectiveness is a compliance responsibility. Deciding what level of
process control capability justifies the training investment is an
engineering responsibility. Any one of these functions can block the
initiative, and in most organizations, at least one of them will.
The third reason is that prevention challenges the organizational
status quo in ways that failure correction does not. A rework cell with
six operators is a manageable problem — it costs money, but everyone
understands it, and the supervisor is incentivized to keep it running. A
root cause analysis that reveals the rework cell exists because of a
design flaw that engineering chose not to fix, or a supplier that
purchasing selected over quality’s objection, or a process that
production decided to run outside its validated parameters — this is not
a manageable problem. This is a political problem, and most quality
managers have learned that political problems are best left alone.
Cost of Quality in the
Digital Era
The ability to measure and manage quality costs has improved
significantly with digital manufacturing technologies, though adoption
remains uneven. Modern manufacturing execution systems can capture
rework and scrap data in real time, tied to specific operations, part
numbers, and root cause codes. Quality management systems can track
nonconformance reports, corrective actions, and customer complaints
through their full lifecycle, accumulating cost data along the way.
Enterprise resource planning systems can tie quality costs to the
general ledger with a level of granularity that was impossible a decade
ago.
The organizations that have invested in this digital infrastructure
and, crucially, in the analytical capability to use it, are achieving
levels of quality cost visibility that previous generations of quality
managers could only dream of. They can see exactly which processes
generate the most nonconformances, which suppliers drive the most
incoming inspection costs, which product lines carry the highest
warranty burden, and which customer segments generate the most complaint
investigation effort. They can model the financial impact of a process
improvement before implementing it, and they can validate the savings
after implementation with actual data.
But the technology is an enabler, not a solution. The organizations
getting value from digital quality cost analysis are the ones that
started with a clear definition of what they wanted to measure, why they
wanted to measure it, and what decisions the measurement would inform.
The ones that bought software expecting it to automatically produce a
credible COQ number are still waiting, and the software vendor has moved
on to the next customer.
The Right Way to Start
If your organization has never conducted a formal Cost of Quality
analysis, the worst thing you can do is attempt a comprehensive,
enterprise-wide study that attempts to quantify every quality-related
cost to the penny. These projects typically take six months, produce a
report that nobody reads, and join the proud tradition of quality cost
analyses that were performed once and never repeated.
The right approach is to start small, start concrete, and start with
a question that someone actually wants answered. Pick a single product
line or a single process that everyone agrees is problematic. Map the
activities associated with nonconformances in that area. Estimate the
costs using readily available data — labor hours, material values,
freight charges. Present the findings to the people who can do something
about it, using language they understand, tied to numbers they
recognize.
If the analysis reveals that a particular process generates fifty
thousand dollars a quarter in rework and scrap, and that a
twenty-thousand-dollar investment in mistake-proofing would eliminate
sixty percent of it, you have a business case that doesn’t require
anyone to believe in Cost of Quality as a concept. It requires them to
believe in a specific investment with a specific return, supported by
specific data. That is a decision they know how to make.
Build from there. Each successful analysis builds credibility,
refines the methodology, and creates demand for the next one. Over time,
the organization develops a practical, believable picture of its quality
costs — not a single number on a slide, but a living understanding of
where poor quality drains resources and where prevention investment pays
off. That understanding, embedded in the daily decision-making of the
organization, is worth more than any COQ report ever written.
The Real Question
Cost of Quality is not an accounting exercise. It is a leadership
question disguised as a financial calculation. The number — whether it’s
three percent of revenue or twenty percent — is less important than the
conversation it forces. Are we paying for quality upfront, through
deliberate investment in getting things right the first time? Or are we
paying for it afterward, through the hidden, distributed,
institutionalized machinery of detection, correction, and compensation?
And if we’re paying afterward — which almost every organization is —
what are we going to do about it?
The organizations that have answered this question honestly, and
acted on the answer, have consistently achieved quality cost reductions
of fifty percent or more within two to three years, while simultaneously
improving customer satisfaction, reducing lead times, and increasing
capacity without capital investment. The ones that haven’t are still
producing COQ reports that nobody reads, still running rework cells that
nobody questions, and still asking the CFO for a bigger quality budget
without being able to explain what the current one is buying.
The choice, as always, is between measurement that drives action and
measurement that provides comfort. Cost of Quality analysis can do
either. Which one it does in your organization depends entirely on the
honesty and courage of the people conducting it.
Peter Stasko is a Quality Architect with over 25
years of experience transforming manufacturing quality systems across
automotive, electronics, and industrial sectors. He specializes in
bridging the gap between quality theory and shop-floor reality, helping
organizations move beyond compliance theater to genuine operational
excellence. His work focuses on practical quality management that
delivers measurable financial results — not just audit pass
certificates.
