Quality
and the Lindy Effect: When Your Organization Discovers That the Quality
Practices Which Survived the Longest Are the Ones Most Likely to Survive
the Next Crisis — and the Trendy Methodologies Everyone Chased Became
the Fads Everyone Forgot
There is a bookshelf in every quality manager’s office that tells a
story nobody reads anymore.
On the top shelf, dusty and unopened, sits a copy of Six Sigma
for Software Development from 2003. Below it, Total Quality
Management: The Second Wave from 1998. Below that, a boxed set of
Business Process Reengineering materials from 1995, still in
its original shrink wrap. And on the bottom shelf, dog-eared and
spine-cracked, sits a copy of Out of the Crisis by W. Edwards
Deming, published in 1982, next to a well-worn What Is Total Quality
Control? by Kaoru Ishikawa from 1981.
The top shelf represents hundreds of thousands of dollars in
consulting fees, training programs, and implementation projects. The
bottom shelf represents the ideas that actually changed something.
This is the Lindy Effect in action, and it is the most underrated
concept in quality management.
What the Lindy Effect
Actually Says
The Lindy Effect is a theoretical principle originally proposed for
non-perishable things like technologies, ideas, and books. Its core
proposition is deceptively simple: the future life expectancy of
a non-perishable thing is proportional to its current age. A
book that has been in print for fifty years is more likely to be in
print fifty years from now than a book published last month. A
technology that has survived a century is more likely to survive another
century than a technology invented last Tuesday.
The idea got its name from Lindy’s delicatessen in New York, where
comedians would gather in the 1960s and discuss which performers would
have lasting careers. The observation was simple: the performers who had
already been successful the longest were the best bets to continue being
successful. Not the hot new act. Not the viral sensation. The ones who
had already proven they could survive.
Nassim Nicholas Taleb later popularized the concept and extended it
rigorously: for things that don’t have a natural expiration date —
ideas, technologies, methodologies, institutions — time itself is the
most reliable filter. Survival is evidence of fitness. The things that
have endured have already survived every crisis, every disruption, every
competitive challenge that their predecessors could not. They carry
within them the accumulated wisdom of having been tested by reality.
Applied to quality management, the Lindy Effect makes a prediction
that most organizations systematically ignore: the quality
practices that have been around for decades — statistical process
control, control charts, root cause analysis, standard work, the PDCA
cycle — are more likely to be relevant in twenty years than whatever
methodology was launched at last year’s quality conference.
This is not nostalgia. This is mathematics.
The Quality Graveyard
of Good Intentions
Let me walk you through a partial list of quality movements that
organizations adopted with evangelical fervor and then quietly
abandoned:
Quality Circles (1970s-1980s). Voluntary groups of
workers meeting to solve problems. The concept was sound — engage
frontline workers in improvement. Implementation was another matter.
Organizations adopted the form without the substance, mandating
participation, turning voluntary problem-solving into mandatory
meetings, and then wondering why enthusiasm evaporated. The companies
that genuinely embedded worker-driven improvement survived. The ones
that went through the motions moved on to the next thing.
Business Process Reengineering (1990s). Michael
Hammer and James Champy told organizations to stop automating bad
processes and start from scratch. “Don’t automate, obliterate.”
Companies took this as license to lay people off while calling it
transformation. The useful kernel — periodically questioning whether
your processes still make sense — got buried under consulting fees and
headcount reduction. BPR became synonymous with downsizing, and the term
became radioactive.
Total Quality Management (late 1980s-1990s). Before
you object — yes, TQM contained genuinely valuable ideas. But as a
branded movement, it became so diluted that “TQM” eventually meant
“whatever quality-related activity we’re doing this quarter.”
Organizations that treated TQM as a label rather than a philosophy moved
on when the label lost its shine. The underlying principles — customer
focus, continuous improvement, data-driven decisions — survived because
they were never dependent on the acronym.
Six Sigma as Management Religion (2000s). Six
Sigma’s statistical toolkit is valuable. But the movement’s tendency to
create parallel organizational structures — armies of Green Belts and
Black Belts speaking their own language, operating in their own
projects, disconnected from daily operations — became its undoing.
Companies spent millions certifying people in DMAIC methodology while
their basic process control charts gathered dust. The tools survived.
The movement became a punchline.
Digital Transformation as Quality Strategy
(2010s-present). The current obsession. Every quality system
must be digital, cloud-based, AI-powered, and integrated with IoT
sensors. And some of this is genuinely useful. But organizations are
currently spending fortunes on digital quality platforms while their
operators still cannot reliably follow standard work, their measurement
systems still produce unreliable data, and their corrective actions
still address symptoms instead of root causes. The technology is new.
The problems are ancient.
The pattern is consistent and the lesson is Lindy-compatible:
the core ideas survive. The packaging expires.
Statistical thinking survived Six Sigma. Worker engagement survived
Quality Circles. Process analysis survived BPR. Continuous improvement
survived TQM. What died were the brands, the consulting frameworks, and
the certification programs — the perishable wrappers around imperishable
ideas.
What Has Actually Survived
Let’s look at what the quality profession’s longest-surviving
practices have in common:
Statistical Process Control (1920s). Walter Shewhart
developed the control chart at Western Electric in 1924. One hundred
years later, control charts remain the most reliable method for
distinguishing between common cause and special cause variation. The
mathematics has not changed. The application has expanded. Every modern
SPC software package implements the same fundamental logic Shewhart
described a century ago. This idea has survived the Great Depression,
World War II, the computer revolution, globalization, and the rise of
AI. The Lindy Effect predicts it will survive whatever comes next.
The PDCA Cycle (1950s). Deming popularized
Plan-Do-Check-Act (originally developed by Shewhart as PDSA).
Seventy-five years later, every functional quality system on earth
operates on some version of this cycle. Organizations that actually use
it consistently improve. Organizations that complicate it with
additional steps, phases, and decision gates tend to produce bureaucracy
rather than results. The simplest form has proven the most robust.
Root Cause Analysis (1950s). The 5 Whys technique,
the fishbone diagram, the basic discipline of asking “why” instead of
“who” — these have survived every management trend since the Eisenhower
administration. They survive because they work, because they are simple
enough to teach in fifteen minutes, and because they address a
fundamental human tendency: the urge to treat symptoms instead of
diseases.
Standard Work (1950s). Toyota’s approach to
documenting the current best known method for performing a task. Not a
permanent standard — a temporary baseline from which improvement becomes
possible. The concept has survived because it acknowledges a truth that
every quality professional learns eventually: you cannot improve what
you cannot define, and you cannot sustain what you cannot replicate.
FMEA (1950s-1960s). Failure Mode and Effects
Analysis emerged from the aerospace and automotive industries. The basic
structure — identify what could go wrong, assess how badly, determine
how likely, decide what to do about it — has remained essentially
unchanged for sixty years. Software tools have made it easier to
document. The thinking process is identical.
Gemba (the practice of going to where the work
happens). Not a tool, not a technique, just the disciplined
practice of leaving your office and observing reality directly. Has
survived because it addresses the universal tendency of managers to make
decisions based on reports, assumptions, and abstractions rather than
direct observation.
Notice what these survivors share: they address fundamental,
unchanging aspects of human organizations. People will always
make mistakes. Processes will always vary. Managers will always be
tempted to manage from abstractions. Problems will always have deeper
causes than first appear. These are not problems that technology solves.
They are conditions that effective practices manage.
The Lindy Effect predicts that these practices will continue to be
relevant precisely because they address these unchanging conditions.
Why Your
Organization Keeps Falling for the New Thing
If proven practices are more likely to remain effective, why do
organizations keep chasing new methodologies?
Three reasons.
First, the survivors are boring. Statistical process
control does not generate excitement at a board meeting. “We implemented
better control charts” does not make for a compelling conference
presentation. “We’re launching an AI-powered predictive quality
platform” does. The old practices work quietly. The new practices
announce themselves loudly. Human psychology, particularly in
organizational settings, rewards the visible and the novel over the
proven and the quiet.
Second, the survivors require discipline, not
budget. You cannot solve a PDCA problem by throwing money at
it. You solve it by thinking clearly, observing carefully, and following
through consistently. This is hard, unglamorous work. New methodologies
often promise to substitute investment for discipline — buy our
platform, certify your people, adopt our framework — and that promise is
seductive precisely because it offers a shortcut around the hard
work.
Third, the survivors produce incremental improvement, and
organizations crave transformation. A well-run control chart
program might reduce your defect rate by 15% over two years. That is
genuinely valuable. But it is not the kind of number that gets you
promoted or featured in industry publications. “Transformational change”
sounds better than “we got slightly better at something we were already
doing.” So organizations abandon the slow, reliable improvement engine
in favor of the dramatic, unproven transformation promise.
The
Practical Application: A Lindy Test for Quality Investments
Here is a framework I have used with organizations for the past
decade. I call it the Lindy Test, and it works like this:
Before adopting any new quality practice, tool, or
methodology, ask three questions:
-
Does this address a fundamental, unchanging condition of
quality management? (Variation, human error, process drift,
measurement uncertainty, organizational learning, customer needs.) If it
addresses a temporary condition — a specific regulatory requirement, a
current technology limitation, a market-specific challenge — it may be
useful but it is perishable. Invest accordingly. -
Can I identify the core insight, separate from the
branding? Every new methodology packages one or two genuine
insights in layers of jargon, certification requirements, and consulting
frameworks. Strip away the branding. What is the actual idea? If the
actual idea is “use data to make decisions” or “engage workers in
problem-solving” or “check whether your measurement system is reliable,”
you already have proven methods for doing those things. Consider whether
the new packaging offers genuine value over the existing
approach. -
Has the core insight survived at least one previous
management cycle? If the idea is genuinely new, it might be
genuinely valuable — but it is also genuinely untested. Treat it as an
experiment, not a transformation. If the core insight has appeared
before under different names, that is Lindy-validated. The packaging is
new; the idea is proven. Implement the idea, skip the
packaging.
After applying the Lindy Test, allocate your quality
improvement resources using the 70-20-10 rule:
-
70% to proven, Lindy-validated practices. SPC,
PDCA, root cause analysis, standard work, FMEA, Gemba walks, measurement
system analysis. These are your foundation. Invest in them consistently,
deeply, and without fanfare. This is where sustainable improvement comes
from. -
20% to proven practices adapted for current
conditions. Using control charts with real-time data
collection. Applying FMEA thinking to software development. Adapting
Gemba walks for remote and hybrid work environments. The practices are
proven; the application evolves. This is where modernization happens
without losing rigor. -
10% to genuinely new experiments. AI-assisted
defect detection. Digital twin simulation for process optimization.
Whatever emerged from the latest research. But treat these as
experiments. Set clear success criteria. Evaluate honestly. And be
prepared to discover that the exciting new approach produces marginal
improvement over the boring old method that your operators actually
follow.
The Counterargument and Its
Limits
The obvious objection to the Lindy Effect in quality management is:
“But things have changed.” Manufacturing is more complex. Supply chains
are global. Products are software-defined. Regulatory environments are
more demanding. Customer expectations are higher. The pace of change is
faster. Surely old methods are insufficient for new challenges.
This objection is partially correct and mostly wrong.
It is correct that the application of quality principles must evolve.
A control chart for a high-volume automotive stamping line looks
different from a control chart for a low-volume pharmaceutical batch
process. Gemba walks in a fully automated factory require different
observation skills than Gemba walks in a manual assembly operation. The
practices must adapt to context.
But it is mostly wrong because the fundamental challenges have not
changed at all. Your operators still make mistakes. The
specific mistakes have changed — typing the wrong value into a digital
interface instead of misreading an analog gauge — but the underlying
reality of human error in complex processes is identical to what
Shewhart observed in 1924. Your processes still vary.
The sources of variation have multiplied — software versions, network
latency, sensor calibration drift — but the mathematics of
distinguishing signal from noise has not changed. Your managers
still manage from abstractions. The dashboards are prettier
now. The data is more abundant. The temptation to make decisions based
on aggregated reports rather than direct observation is, if anything,
stronger than it was in Deming’s time.
The environment changes. The principles endure. This is precisely
what the Lindy Effect predicts.
A Story from the Field
I worked with an automotive supplier in 2019 that had invested
heavily in a predictive quality platform. Machine learning models
trained on sensor data, real-time dashboards, automated alerts. The
system had cost north of two million dollars and had been featured in
multiple industry publications.
When I visited the plant, I asked to see their basic process control.
The control charts on the critical dimensions — the ones that had driven
customer complaints for the previous six months — had not been updated
in three months. The operators knew this. The shift supervisors knew
this. The quality engineers knew this. Nobody had updated the charts
because everyone was focused on the predictive quality platform.
The predictive platform was detecting anomalies. It was generating
alerts. But the alerts went to quality engineers who were already
overwhelmed managing the platform itself. Meanwhile, the basic process —
the one that was producing the defects — was not being monitored with
the simplest, most proven tool available.
We restarted the control charts. Within two weeks, the operators
identified a tool wear pattern that the predictive platform had missed —
not because the platform was technically inferior, but because nobody
was looking at the basic process behavior. The platform was monitoring
the forest. The operators, armed with control charts, noticed the
specific tree that was diseased.
The predictive platform was not useless. It provided genuinely
valuable insights. But those insights were layered on top of a
foundation that had been neglected. The two-million-dollar system could
not compensate for the failure to maintain the hundred-dollar
practice.
This is the Lindy lesson: build on the foundation. Do not
substitute for it.
What to Do Monday Morning
If you accept the Lindy Effect as a useful lens for quality
management, here are specific actions:
Audit your Lindy foundation. List the proven
practices your organization should be executing consistently: SPC,
calibration, standard work, corrective action follow-through, management
review, internal audits. Rate each one honestly on a 1-10 scale for
actual execution quality. Most organizations discover that their
fundamental practices — the ones with decades of proven effectiveness —
are executing at a 4 or 5 out of 10 while their new initiatives are
consuming 80% of leadership attention.
Rebalance your investment. Shift resources toward
making your Lindy-validated practices genuinely excellent. This is not
glamorous work. It involves training, retraining, auditing, coaching,
and following up. It involves holding people accountable for doing the
basics well. But it produces more sustainable improvement than any new
initiative.
Evaluate new investments through the Lindy lens.
Before approving the next quality technology purchase or methodology
adoption, ask: “Are we doing this because it addresses a real gap in our
proven practices, or are we doing it because it is new?” If the answer
is the latter, pause.
Protect your institutional memory. The Lindy Effect
works because survival encodes information. When experienced quality
professionals leave and take their knowledge of why things work the way
they do, you lose accumulated wisdom that cannot be replaced by any
methodology. Mentor. Document. Transfer knowledge deliberately.
The Uncomfortable Truth
The quality profession does not have a tool shortage. It has a
discipline shortage.
Every statistical technique you need was developed before 1960. Every
management principle you need was articulated before 1980. Every
problem-solving approach you need has been proven across thousands of
organizations over decades.
The challenge is not finding better methods. The challenge is
executing the methods you already have with sufficient discipline,
consistency, and attention to detail. The Lindy Effect tells us
something we do not want to hear: the quality practices most
likely to help you in the next crisis are the same ones that helped
organizations survive the last crisis, and the one before that, and the
one before that.
They are not exciting. They are not innovative. They are not going to
win you a speaking slot at the quality conference. But they work. And
“works” has a longer track record than “innovative” ever will.
Your dusty bookshelf already contains the answers. The question is
whether you have the discipline to implement them.
Peter Stasko is a Quality Architect with 25+ years of experience
transforming organizations across automotive, aerospace, and
pharmaceutical industries.
