A Tier 1 automotive supplier committed to deploying a new quality management system across three plants in nine months. The board approved the timeline. The quality director signed off. The project charter contained detailed Gantt charts, milestones mapped to the week, and resource allocations that appeared reasonable on paper. The team built in a three-week buffer for unexpected issues.
Twenty-three months later, the system was still not fully operational. Two plants had gone live with partial implementations. The third was running parallel systems. Cost overruns had consumed the contingency budget and doubled it. The quality director had been replaced. Everyone had agreed the project was on track just six months earlier.
This is not unusual. It is the norm. The planning fallacy, identified by Daniel Kahneman and Amos Tversky, describes the systematic tendency to underestimate the time, cost, and risks of future actions while overestimating benefits. In quality management, where timelines determine inspection schedules, supplier qualifications, process validations, and regulatory submissions, this fallacy creates the conditions under which defects thrive and compliance fails.
Inside the Fallacy: Two Systems, One Prediction
Kahneman's dual-process theory explains why the planning fallacy persists in organisations full of experienced professionals. System 1, the fast intuitive processor, generates the initial estimate by constructing a best-case scenario. Not the average case. Not the worst case. It imagines everything going right: no supplier delays, no key personnel resignations, no regulatory surprises, no equipment failures.
System 2, the slow analytical processor, is supposed to review and correct this estimate. But System 2 is lazy. It requires effort that the brain conserves by default. When the project manager presents a nine-month timeline that feels reasonable, System 2 nods along, makes minor adjustments at the margins, and approves. The result is not a prediction. It is a wish dressed in professional clothing.
In quality organisations, this plays out with remarkable consistency. Process validation timelines assume every run produces acceptable data on the first attempt. PFMEA completion schedules allocate two days for what requires two weeks of cross-functional analysis. Corrective action plans treat root cause investigation as a linear exercise rather than an iterative one. Each plan describes what happens when nothing goes wrong, in a domain where something always goes wrong.
Reference Class Neglect and Historical Amnesia
The most damaging feature of the planning fallacy is what Kahneman calls reference class neglect: the failure to look at similar past projects and use their actual outcomes as a basis for prediction. When a quality team plans a PPAP submission timeline, they almost never start by asking how long the last twelve submissions actually took, from kick-off to final approval.
They start by imagining the steps, estimating each one in isolation, and adding them up. This inside-view approach systematically excludes the reality of delays, revisions, failed tests, customer rejections, and the organisational friction that turns three-week tasks into three-month ordeals. The data is usually available. PPAP logs exist. Project management tools track actual versus planned dates. But the data is never consulted, because the planning mind prefers its own narrative construction over the messy evidence of history.
I have audited plants that discovered this gap the hard way. One organisation tracked their average CAPA closure time at 127 days over three years. Their target for new CAPAs was 45 days. When I asked how they arrived at 45, the quality manager cited the standard recommendation. The standard described a best practice, not a realistic target for an organisation that had never closed a complex CAPA in under 90 days. The gap between aspiration and capability drove people to close CAPAs superficially, checking boxes without addressing root causes.
The Social Mechanics of Unrealistic Plans
The planning fallacy persists not only because of cognitive architecture but because of organisational dynamics that reward optimism and punish realism. Consider the typical planning meeting. A quality director presents a timeline. The VP of Operations asks if it can be compressed. The director, who knows the timeline is already optimistic, faces a choice: defend the estimate and be seen as obstructionist, or agree to compression and be seen as collaborative.
The political incentives favour compression. The facts favour defence. Politics wins. This dynamic creates aspirational scheduling: timelines that everyone knows are unrealistic but nobody challenges because the social cost is too high. The project manager who points out that similar projects take twice as long is labelled negative. The engineer who asks for more testing time is called risk-averse. The quality lead who requests additional resources is told to work smarter.
I witnessed this at a manufacturer where the leadership team approved a 12-month timeline for IATF 16949 certification. The consultant told them 18 months was realistic. The internal quality team privately estimated 24 months. Leadership chose 12 because the customer required it by Q4. Certification was achieved in 22 months, after two failed audits, one consultant replacement, and near-complete turnover of the quality department staff who burned out chasing an impossible target.
A buffer that matches historical variance is a realistic plan. A plan without such a buffer is a fiction.
Where the Fallacy Strikes Hardest
Process validation protocols assume processes behave predictably from run one. They do not. First runs reveal problems. Second runs reveal different problems. By the third run, you understand the process well enough to design a proper validation. But the plan allocates three consecutive successful runs in two weeks, and when run one fails, the entire schedule collapses.
Root cause investigation is planned as a discrete event: a two-day workshop that produces a definitive answer. In practice, genuine investigation is iterative. You form a hypothesis, test it, discover it is wrong, form a new hypothesis, expand the scope, involve new people, and gradually converge on the truth. This takes weeks or months. But the 8D or CAPA form has a field for root cause identified date that is expected to be filled in within 72 hours of the initial complaint.
Supplier development plans typically include the supplier's commitment to implement changes within an agreed timeline. What they do not include is the supplier's competing priorities, internal resistance to change, resource constraints, financial limitations, and the fact that they promised the same improvements to three other customers. The plan assumes the supplier is a reliable execution partner. Experience shows they rarely are.
Inside-View vs. Outside-View Planning
Inside-view planning
- Estimates each step in isolation, assuming nothing goes wrong
- Ignores historical data from previous PPAP, CAPA, or validation cycles
- Produces a single point estimate that leadership can compress
- Collapses when the first unexpected result invalidates the schedule
Outside-view planning
- Starts with the median actual duration of the last 10 to 20 comparable projects
- Adjusts the base rate for project-specific complexity and resourcing
- Produces a range estimate tied to documented historical variance
- Absorbs early failures without cascading into a full schedule collapse
Reference Class Forecasting: The Correction
Kahneman's proposed remedy is reference class forecasting: basing predictions on the actual outcomes of similar past projects rather than on the specific details of the current one. The technique is straightforward but requires discipline. Identify the reference class, gather historical data on actual durations, compute the base rate, adjust for specific factors, and build the plan from the adjusted base rate.
A Tier 1 automotive supplier I advised implemented this for their APQP timelines. Their historical data showed that new product launches had a median duration of 14 months from kick-off to PPAP approval, with a range of 11 to 23 months. Their standard planning template had assumed 9 months. When they recalibrated to the historical base rate and communicated the realistic timeline to their customer, the customer pushed back hard.
But when the supplier presented 47 completed launches with actual timelines, the negotiation shifted from why can you not do it faster to what would it take to accelerate, and what are the risks. The actual launch came in at 13 months, one month under the base rate. It was the first time in the supplier's history that a major launch beat a realistic schedule, because the realistic schedule gave them space to plan properly, execute carefully, and address problems when they were small.
Gary Klein's pre-mortem technique provides a complementary defence. Before the project begins, gather the team and state: it is twelve months from now, this project has failed, what went wrong? This question activates a different cognitive pathway. Instead of constructing a best-case narrative, the team constructs a failure narrative. People who sat quietly during the optimistic planning session suddenly produce a list of risks they were reluctant to mention. In quality projects, pre-mortems consistently surface the same categories: key personnel pulled to other priorities, customer requirement changes after work begins, IT systems not ready, and management attention shifting to the next crisis.
Building a Realistic Planning Culture
Overcoming the planning fallacy requires organisational systems that reward accuracy over optimism. Track and publish actual versus planned data. When every project has a retrospective comparing the original plan to the actual outcome, and those comparisons are stored in a searchable database, the reference class data builds itself. Make the gap visible. The gap is not a failure of execution. It is the measurement of a distortion that has always existed.
Separate aspiration from prediction. It is valid to aspire to close CAPAs in 45 days. It is dishonest to plan as though 45 days is the expected outcome when the historical average is 127. State the aspiration. State the prediction. Track both. Let the gap between them drive improvement, not deception. Reward the engineer who says this will take six months, not three. They are not being negative. They are being accurate.
Use range estimates instead of point estimates. Instead of saying the QMS implementation will take 12 months, say it will take 12 to 18 months depending on resource availability and legacy system complexity. Range estimates are more honest, more useful, and more defensible when leadership inevitably asks for compression. Build planning buffers that match historical variance. If process validations typically take 40 percent longer than planned, a 40 percent buffer is not padding. It is calibration.
The Cost of Aspirational Scheduling
The planning fallacy persists because it serves a psychological function. Optimistic plans feel good. They signal confidence, ambition, and capability. They tell stakeholders what they want to hear. But in quality management, the cost of unrealistic planning is compromised quality. When timelines compress, the first sacrifice is thoroughness. Inspections get rushed. Training gets abbreviated. Validation runs get fewer iterations. Root causes get superficial treatment.
The planning fallacy does not just create late projects. It creates defective products, failed audits, customer complaints, and the erosion of quality culture as people learn that the plans are performative. The real plan becomes do whatever it takes to hit the date, and hope nothing goes wrong. The antidote is not pessimism. It is realism. And realism begins with a question most organisations never ask: the last time we did this, how long did it actually take? The answer is the only honest starting point for the next plan.