Quality
and the Premortem: When Your Organization Imagines the Failure Before It
Happens — and the Disaster You Predicted Becomes the Disaster You
Prevented
The project launch meeting was going perfectly. The team had reviewed
every deliverable, confirmed every milestone, and nodded along to the
quality plan. The mood was optimistic. The schedule was aggressive but
achievable. Everyone agreed this would be the smoothest launch in the
plant’s history.
Then the quality director asked a question that made the room go
quiet.
“Imagine it’s six months from now. This launch has failed
spectacularly. What went wrong?”
Nobody wanted to answer. They had just spent an hour convincing
themselves — and each other — that everything would go right. Now they
were being asked to do the opposite. Not to hope for success. To
assume failure and explain why it happened.
That question saved the launch. Within thirty minutes, the team
identified three risks that hadn’t appeared in any FMEA, two supplier
vulnerabilities that the procurement team had been quietly worried
about, and a training gap that would have left an entire shift
unprepared for a new inspection protocol. None of these issues surfaced
during the normal planning process. They only emerged when the team
stopped trying to be right and gave themselves permission to imagine
being wrong.
This is the premortem. And if your organization isn’t using it,
you’re leaving your most preventable failures hiding in plain sight.
What Is a Premortem?
The premortem technique was developed by psychologist Gary Klein in
2007 as a formal response to a well-documented problem: people are
terrible at predicting what will go wrong with their own plans, but
surprisingly good at explaining why something did go wrong
after the fact.
The logic is elegantly simple. Instead of asking “what might go
wrong?” — which triggers optimism bias and defensive thinking — the
premortem asks participants to assume the project has already failed and
then work backward to explain why. This small reframing shifts the brain
from a protective posture to a diagnostic one. People who would never
volunteer a concern about their own plan become remarkably candid when
asked to explain an imaginary failure.
In quality management, this technique is a natural complement to
tools like FMEA, risk assessments, and control plans. But where those
tools operate in a structured, analytical register, the premortem
operates in a narrative, psychological one. It captures the risks that
don’t fit neatly into a severity-occurrence-detection matrix — the
political risks, the organizational risks, the “everyone knows but
nobody says” risks that live in the gaps between process steps.
Why
Traditional Risk Tools Miss What Premortems Catch
If you work in quality, you’re already familiar with risk
identification tools. FMEA, hazard analysis, fault tree analysis — these
are the backbone of preventive quality. They work by systematically
examining process steps, identifying potential failure modes, and
ranking them by risk priority numbers.
But here’s the problem. These tools are only as good as the team’s
willingness to be honest about what could go wrong. And decades of
cognitive science research tell us that teams are systematically
dishonest — not maliciously, but cognitively — about their own
vulnerabilities.
Planning fallacy makes us underestimate timelines
and overestimate our ability to deliver. Optimism bias
makes us discount negative outcomes and overweight positive ones.
Confirmation bias makes us seek information that
confirms our plan is sound while ignoring signals that suggest
otherwise. Groupthink makes teams converge on a shared
optimistic view and suppress individual doubts. Sunk cost
thinking makes us reluctant to acknowledge problems with plans
we’ve already invested in.
These biases don’t show up in an FMEA. An FMEA can’t capture the fact
that three team members privately believe the supplier won’t deliver on
time but won’t say so because the procurement manager is in the room. It
can’t capture the fact that the production supervisor is planning to
retire in four months and hasn’t told anyone, leaving a massive
knowledge transfer gap. It can’t capture the fact that the last three
similar launches all failed for the same reason, but nobody wants to
mention it because it would reflect poorly on leadership.
The premortem catches these. Not because it’s a better analytical
tool, but because it’s a better psychological tool. It creates
a safe space for pessimism. And in quality, pessimism is not the enemy
of progress — it’s the guardian of it.
How to Run a Quality
Premortem
The premortem process is deliberately simple. Complexity defeats the
purpose. Here’s how to run one that actually works.
Step 1: Frame the Scenario
Begin by clearly defining the project, process change, product
launch, or initiative you’re evaluating. Make sure everyone in the room
understands the scope, the timeline, and the expected outcomes.
Then deliver the framing: “Assume it is [specific future date]. This
project has failed. Not partially succeeded. Not delivered with minor
issues. Failed. Completely and spectacularly. Your job
now is to write the post-mortem report explaining what happened.”
This framing is critical. The word “assume” is doing important
psychological work. You’re not asking people to predict failure or
estimate probabilities. You’re asking them to treat failure as a given
and then explain it. This removes the social risk of being seen as
negative or unsupportive. Everyone is playing the same game: explaining
a failure that has already happened.
Step 2: Generate
Individual Explanations
Give every participant five to ten minutes to write down their
explanation of the failure. Individually. In silence. No discussion
yet.
This individual phase is where the premortem earns its value. When
people write their own explanations before hearing anyone else’s, you
get a diverse set of risks that haven’t been filtered through social
dynamics. The junior engineer who would never speak up in a group
setting writes down the real reason she thinks the new gage won’t work.
The supplier quality specialist who’s been nervous about a single-source
component puts it on paper without fear of contradicting the procurement
team.
Each person should generate three to five failure explanations.
Encourage them to think beyond technical risks. Ask them to consider
organizational, political, resource, timing, and human factors. The best
premortems capture risks that live at the intersection of these domains
— the technical problem that becomes catastrophic because of a political
decision, the resource gap that becomes critical because of a timing
constraint.
Step 3: Share and Cluster
Go around the room. Each person reads their failure explanations. A
facilitator records every item without judgment, evaluation, or
debate.
You’ll immediately notice patterns. The same failure will appear on
multiple lists, often described from different angles. One person might
say “the new inspection process wasn’t validated properly.” Another
might say “operators on night shift didn’t receive the training.”
Another might say “the quality engineer left the company two weeks
before launch.” These are three facets of the same systemic
vulnerability — and the premortem just revealed it from three
perspectives simultaneously.
Cluster similar items together. Don’t merge them — keep the
individual descriptions, because the nuances matter. The fact that three
people independently identified a training gap but described it
differently tells you something important about how that risk is
perceived across the organization.
Step 4: Prioritize and Act
Once all failure explanations are visible, the team can prioritize
them using whatever framework fits your organization. I’ve seen teams
use simple voting (dot stickers on the most concerning items), risk
matrices (plotting likelihood against impact), or direct integration
into existing FMEA documents.
The key deliverable isn’t a ranked list. It’s a set of
specific, actionable responses to the highest-priority
failure scenarios. For each top risk, the team should answer: What would
we need to do differently right now to prevent this from
becoming our actual post-mortem?
This is where the premortem transforms from an interesting exercise
into a practical quality tool. The failure explanations become leading
indicators. The preventive actions become control measures. And the
whole exercise becomes a living document that the team revisits at every
project gate.
Where Premortems Work
Best in Quality
Not every quality activity needs a premortem. But certain situations
benefit enormously from this technique.
New product launches. The classic use case. Before
launch approval, gather the cross-functional team and run a premortem.
You’ll be stunned by what surfaces.
Process changes. Before implementing a significant
process modification — new equipment, revised routing, updated
inspection criteria — a premortem can reveal risks that the change
management process missed.
Supplier transitions. Switching suppliers or adding
new sources is one of the highest-risk activities in manufacturing. A
premortem that includes procurement, quality, engineering, and
production will identify vulnerabilities that no single function can see
alone.
Audit preparation. Before a major certification or
customer audit, run a premortem assuming the audit resulted in major
findings. The failure explanations will direct your preparation efforts
toward the areas that actually need attention.
Organizational changes. Restructuring a quality
department, changing reporting lines, merging quality functions after an
acquisition — these changes carry quality risks that are invisible in
organizational charts but obvious in a premortem.
Corrective action implementation. After identifying
a root cause and designing a corrective action, run a premortem assuming
the corrective action failed. This catches the most common CAPA failure
mode: a technically correct solution that falls apart during
implementation.
The Psychology of Why It
Works
Understanding why the premortem works is almost as important as
knowing how to run one, because the underlying psychology applies far
beyond this single technique.
Prospective hindsight is the formal term for what
the premortem exploits. Research by Mitchell, Russo, and Pennington
(1989) found that when people imagine an event has already occurred and
then explain why, they generate 30% more reasons than when they simply
try to predict what might happen. The explanation task activates
different cognitive pathways than the prediction task. We’re simply
better at constructing narratives about events that have “happened” than
about events that “might” happen.
Social proof for pessimism is another mechanism. In
most planning meetings, optimism is the default social norm. Expressing
doubt feels like being disloyal or unsupportive. The premortem flips
this dynamic. Pessimism becomes the required behavior. Everyone is asked
to find problems. The social risk shifts from “will I look negative if I
raise this concern?” to “will I look unprepared if I don’t have a
failure explanation?” This is a profound shift, and it unlocks candor
that normal planning processes never access.
Temporal distance matters too. When you ask someone
“what might go wrong with this plan,” they evaluate the plan from their
current perspective, with all their emotional investment intact. When
you ask them to imagine being six months in the future looking back at a
failure, you create psychological distance from the current plan. It
becomes easier to be objective about something that feels like it
belongs to the past rather than something you’re actively committed to
in the present.
Common Mistakes That Kill
Premortems
I’ve seen organizations try premortems and abandon them, usually
because they made one of these preventable errors.
Treating it as a checkbox exercise. If you schedule
15 minutes at the end of a three-hour meeting and rush through the
process, you’ll get superficial results that confirm what you already
know. A proper premortem needs 60 to 90 minutes, dedicated space, and
genuine commitment from leadership to act on what emerges.
Inviting the wrong people. Premortems work best with
diverse perspectives — different functions, different experience levels,
different organizational positions. If you only include the project core
team, you’ll get risks that the core team has already considered.
Include the people who will actually execute the plan, the people who
interact with its outputs, and the people who have seen similar plans
fail before.
Not acting on the results. This is the most common
and most damaging failure mode. If you run a premortem, identify real
risks, and then ignore them because the schedule is too tight or the
budget is already committed, you’ve done something worse than not
running the premortem at all. You’ve taught your team that their honest
input doesn’t matter. The next premortem will be useless because people
will have learned that the exercise is performative.
Running them too early or too late. Too early, and
the plan isn’t concrete enough to generate specific failure
explanations. Too late, and you’re locked into decisions that can’t be
changed. The sweet spot is when the plan is detailed enough to evaluate
but flexible enough to modify — typically at a design review, gate
review, or similar decision point.
Confusing premortems with risk registers. A risk
register is a living document that tracks identified risks, their
likelihood, impact, and mitigation plans. A premortem is a time-bounded
exercise that generates risk insights. They complement each other but
serve different purposes. Don’t try to turn one into the other.
A Personal Observation
In twenty-five years of quality work across automotive, aerospace,
and pharmaceutical manufacturing, I can tell you that the most costly
failures I’ve witnessed were almost never caused by risks that were
technically difficult to identify. They were caused by risks that were
organizationally difficult to acknowledge.
Everyone knew the single-source supplier was a vulnerability.
Everyone knew the new process hadn’t been validated at full production
speed. Everyone knew the training was rushed. These weren’t secrets.
They were open secrets — things that everyone knew and nobody said, or
that someone said once in a meeting where it was inconvenient to hear,
so it was noted and moved past.
The premortem doesn’t give you better analytical tools. Your FMEA
training already gave you those. What it gives you is a better
social tool — a structured way to make the unsayable sayable,
to make the known-but-unacknowledged into something the team can
actually work with.
The quality director who asked “imagine this launch failed — what
went wrong?” didn’t have access to information that the rest of the team
lacked. She had access to a technique that made the information the team
already possessed safe to share.
That’s the value of the premortem. Not new data. New honesty.
Building
Premortems Into Your Quality System
If you’re convinced enough to try, here’s how to make premortems a
sustainable part of your quality practice rather than a one-time
experiment.
Start with one high-stakes project. Don’t mandate
premortems for everything. Pick a launch, a process change, or a
corrective action that has real visibility and real risk. Run the
premortem well. Let the results speak for themselves.
Document the outcomes. Track which risks were
identified through the premortem that weren’t captured through normal
risk assessment processes. Track which of those risks actually
materialized in similar past projects. This evidence builds the case for
wider adoption.
Integrate with existing tools. Don’t create a
separate premortem deliverable. Feed the results into your FMEA, your
risk register, your control plan. Make the premortem an input to the
tools you already use, not an additional tool you need to maintain.
Train facilitators. The premortem is simple, but
facilitation matters. Someone needs to create the right atmosphere,
manage the energy in the room, and ensure that the exercise doesn’t
devolve into blame or complaint. Quality engineers, with their training
in structured problem-solving and their neutral organizational position,
are natural facilitators.
Revisit past premortems. After a project completes —
whether it succeeded or failed — go back to the premortem document and
compare the predicted failures with what actually happened. This
calibration improves future premortems and builds organizational
intuition about where your real risks live.
The Deeper Lesson
The premortem technique works because it acknowledges something that
most quality tools don’t: that the biggest barriers to quality are often
psychological, not technical. We have the tools to identify risks,
analyze root causes, and design effective controls. What we lack is a
reliable mechanism for overcoming the human tendency to see what we want
to see and ignore what we’d rather not.
Every quality professional knows that prevention is more effective
than detection. Every quality professional knows that early
identification of risks saves more money than late reaction to failures.
And every quality professional has watched their organization fail to
act on risks that were identified early and then expressed shock at
failures that were entirely predictable.
The gap isn’t knowledge. The gap is courage — the organizational
courage to look at your own plans with honest eyes and admit that they
might fail. The premortem doesn’t create that courage. But it creates
the conditions where courage becomes unnecessary, because pessimism is
no longer an act of bravery. It’s an assignment everyone was given.
The launch I described at the beginning of this article? It went
smoothly. Not because the premortem made the risks disappear — they were
real risks, and the team had to work hard to address them. It went
smoothly because the team addressed those risks before they
became problems, rather than after they became crises.
The disaster they predicted became the disaster they prevented. And
that is the highest form of quality work there is.
Peter Stasko is a Quality Architect with 25+ years
of experience transforming organizations across automotive, aerospace,
and pharmaceutical industries. He specializes in building quality
systems that don’t just comply with standards but create genuine
competitive advantage — systems where prevention is a discipline, not a
slogan, and where the best quality outcomes are the ones nobody notices
because they never became problems in the first place.
