Key Takeaways
When an airplane reveals a flaw, fix the design before blaming the pilot.
Early Boeing treated trouble as evidence. After a 707 demonstration crash, test pilot Tex Johnston argued that training pilots harder would not solve a design problem. Engineer Ed Wells agreed to redesign the tail and rudder, despite the cost. Boeing accepted responsibility for the repair, helping reassure its shaken customer, BOAC. The episode captures a former norm: when the machine behaved dangerously, engineers were expected to improve the machine.
That instinct weakened. The book contrasts this approach with Boeing’s later tendency to point first to pilot error, airline maintenance, or the limits of procedure. The MAX crashes exposed the danger of assuming that a checklist can compensate for an aircraft whose controls behave unexpectedly, especially when the pilots were never told how the software worked.
What's striking is that safety here begins as a question of institutional character: does an organization treat bad outcomes as evidence about its design, or as an inconvenience to explain away? In complex engineering, individual error is real, but blaming the operator too quickly can hide predictable weaknesses in the system. The early Boeing response offers a useful standard: investigate what made a mistake possible, then ask whether a reasonable person could reliably prevent it. That standard does not excuse pilots or airlines. It keeps accountability from stopping at the cockpit door.
Boeing's shareholder-first turn made aircraft quality compete with financial targets.
The company changed its center of gravity. After Boeing acquired McDonnell Douglas in 1997, executives influenced by Jack Welch’s General Electric pushed financial measures such as return on net assets, asset sales, and stock performance. The book describes a shift from treating engineering capability as a strategic asset to treating costs and facilities as numbers to reduce. Boeing spent more than $30 billion on stock buybacks during development of the MAX, while the company’s leaders benefited from compensation tied to financial results.
The trade-offs were consequential. Boeing’s 737 was already an older design, and a new aircraft would have required a much larger investment than the MAX derivative. A low-cost update offered near-term savings and protected the production schedule. But the shareholder logic made the costs of training, testing, and redesign appear immediate, while the risks of underinvestment were deferred and borne by pilots, passengers, and families.
The case is not that financial discipline is inherently unsafe. Large engineering projects need budgets, and waste can itself threaten a company’s ability to innovate. The deeper problem is what a metric leaves out. Buybacks and return ratios are legible to investors; accumulated engineering knowledge, spare testing capacity, and trust are harder to price. When bonuses reward the visible measure, managers can improve the score while quietly consuming the capabilities that make the business resilient. A strong governance system therefore asks not only whether capital is being used efficiently, but what critical capacity is being depleted.
Outsource production without losing the expertise that makes the whole aircraft work.
The Dreamliner exposed the integration cost. Boeing outsourced unusually large portions of the 787’s design and production, aiming to reduce development time and expense. For the 777, a supplier’s electronics specification ran to about 2,500 pages. The comparable Dreamliner document was only about twenty pages. Some suppliers lacked the engineering capacity Boeing expected, and the company had to send drawings repeatedly before contractors understood how components should connect.
Control of information also fragmented. Customer service employees received sections of aircraft manuals marked proprietary, leaving them without the details needed to explain or support the systems. Boeing eventually had to bring work back in-house, acquire suppliers, and spend far more than planned to complete the aircraft. The book’s point is not that every supplier relationship fails. It is that outsourcing a task does not outsource the manufacturer’s responsibility for integration.
A finished aircraft is a system, not a bundle of independently successful contracts. That distinction matters in any industry where components interact and failure can travel across organizational boundaries. Outsourcing may reduce the apparent size of a company while increasing the work of coordination, verification, and knowledge transfer. Those costs often appear later, when schedules are already tight. The strongest case for outsourcing is specialization; its limit is that someone must still understand the interfaces. A manufacturer that loses that understanding can remain legally responsible while becoming practically less able to control the product.
Keep safety regulators independent from the companies whose products they certify.
Delegation blurred accountability. The FAA had long assigned some certification work to engineers employed by aircraft manufacturers. The 2005 Organization Designation Authorization system expanded that arrangement: Boeing selected its own representatives, who reported within Boeing, while FAA supervision became more distant. A Seattle FAA office created to oversee Boeing had about forty agency employees confronting roughly 1,500 company delegates.
The conflict was built into the structure. FAA specialists described pressure to approve work on schedule and difficulty getting their technical concerns heard. In the MAX process, the FAA received a safety assessment based on an earlier, less powerful version of MCAS. The agency did not know that a late design change had substantially increased the system’s authority. When the regulator depends on the producer for expertise, information, and timely completion, formal oversight can remain on paper while practical independence drains away.
Delegation can be sensible when regulators lack enough specialized staff to inspect every detail. The risk arises when delegated experts answer primarily to the organization being regulated, or when agency managers are rewarded for speed. This is a classic principal-agent problem: the public delegates authority to a regulator, which then depends on the producer to carry out part of the regulator’s task. Independence requires more than a rulebook. It requires access to current technical information, authority to challenge decisions, and protection for employees who slow a project over a credible safety concern.
A cheap software fix turned one faulty sensor into a command to dive.
MCAS addressed a design compromise. The MAX’s larger engines changed the aircraft’s behavior in certain turns. Rather than redesigning the tail, which would have cost more and threatened the schedule, Boeing used software called the Maneuvering Characteristics Augmentation System, or MCAS, to push the nose down in conditions it interpreted as a potential stall.
A late change removed safeguards. During testing, MCAS was expanded to operate at lower speeds and to move the stabilizer farther. At those speeds, a second input that had helped provide redundancy no longer applied, so the system could act on one angle-of-attack sensor. If that sensor supplied bad data, MCAS could repeatedly command the nose down. The FAA’s safety paperwork described an earlier version of the software, not the final one. Pilots were not told about MCAS in their operating manual.
Automation is often introduced to manage complexity, but software can also hide a design problem behind an apparently simple control action. The important question is not whether the system is automated; it is whether its inputs are trustworthy, its authority is bounded, and its behavior is understandable when something goes wrong. Redundancy is meaningful only when the system actually compares independent signals under the conditions in which it operates. A safety analysis that describes yesterday’s software cannot establish the safety of today’s aircraft.
Do not ask pilots to defeat hidden automation with memory and luck.
The cockpit made diagnosis harder. The 737 lacked the electronic checklist system Boeing had introduced decades earlier on the 757 and 767. Pilots facing a fault had to interpret alerts, consult a paper handbook, and identify the correct procedure. Yet Boeing sought approval for MAX pilots to transition with brief tablet training rather than simulator practice. Southwest had been promised $1 million per aircraft if simulator training became necessary.
The emergency came at the worst time. On takeoff, pilots had little altitude and many simultaneous warnings. Cutting power to the stabilizer also meant losing the easy electric trim control and using a manual wheel. On the Ethiopian flight, the pilots recognized the suspected sensor problem and followed Boeing’s instructions, but the wheel was too difficult to move at the aircraft’s speed. A procedure is not adequate just because it can be completed in a calm simulator.
Human-factors engineering treats a cockpit as a partnership between people, controls, displays, procedures, and time pressure. A system that depends on a pilot noticing one subtle cue amid alarms may be technically controllable yet operationally unforgiving. This is why realistic simulation matters: it tests not only whether a maneuver is possible, but whether a trained person can perform it in context. Still, more training cannot redeem every design. If the emergency is confusing, physically difficult, and unannounced, responsibility belongs to the organization that designed the conditions, not only to the person facing them.
Protect engineers who raise inconvenient safety concerns.
Warnings existed inside Boeing. Curtis Ewbank and other flight-controls engineers proposed a safeguard that would compare sensor readings and reject implausible data. Managers rejected it twice, citing cost and possible training consequences. A manager reportedly told Ewbank that people had to die before Boeing would change things. Other experienced engineers, including human-factors specialists, left or were laid off as the MAX was developed.
The working climate discouraged resistance. Employees described repeated layoffs, threats tied to cost targets, and pressure to keep the program moving. A countdown clock in the MAX project room made schedule milestones visible, while the chief engineer had no staff reporting directly to him. In such a structure, a technically sound objection could be easy to ignore and personally costly to repeat. The result was not one missing warning, but an organization poorly equipped to make warnings matter.
Dissent is valuable only when it can change a decision. Organizations often praise candor in principle, then penalize it through promotion systems, layoffs, or reputational damage. Safety-critical teams need channels where concerns are documented, assessed by people with relevant expertise, and escalated beyond managers responsible for schedule or cost. This does not mean every objection should halt production. It means the burden of explaining why a warning is wrong should not fall on the employee with the least authority. A culture that can hear bad news is a technical control, not a morale perk.
Treat the first crash as evidence demanding action, not reassurance.
The first Lion Air flight was a warning. On October 29, 2018, Flight 610 crashed into the Java Sea, killing 189 people. The previous flight in the same aircraft had suffered repeated nose-down commands, but an off-duty pilot in the cockpit recognized the moving trim wheel and helped the crew shut off the system. That successful landing showed the fault could be survived, but not that the procedure was reliable for every crew.
The response prioritized continued operation. Boeing and the FAA issued guidance directing pilots to an existing runaway-stabilizer checklist, while the MAX continued flying. Boeing emphasized airline maintenance and pilot performance, and the software itself was not clearly explained to crews. Five months later, Ethiopian Airlines Flight 302 crashed, killing 157. After the second crash, analysis of flight data showed the same basic pattern, and regulators grounded the aircraft.
A first accident rarely arrives with complete causal certainty. Waiting for certainty can seem prudent, especially when grounding a fleet is costly and disruptive. But risk decisions should also consider severity, exposure, and the quality of available safeguards. Here, the warning involved a new system capable of overriding pilot input, and the prescribed response depended on quick recognition under stressful conditions. Precaution is not the same as panic. It is a disciplined choice to limit exposure while investigating when the possible harm is catastrophic and the proposed mitigation has not been convincingly demonstrated.
A new CEO cannot repair a safety culture without changing its incentives.
The leadership response repeated the problem. After the Ethiopian crash, Boeing’s chief executive initially declared the MAX safe and defended its existing process. The company later apologized, grounded production, and faced intense scrutiny from families, pilots, Congress, and regulators. The book describes relatives who had to fight for information and, after the first crash, were pressured to sign broad legal releases in exchange for modest payments.
Reform must reach the system. A 2020 law restored some FAA involvement in delegated certification and required more attention to human factors, but implementation remained uncertain. Boeing also changed reporting lines and discussed new safety mechanisms. Yet the book ends with a warning: the people and incentives that shaped the MAX were not all removed, and the next CEO had served on the board through the crisis. A change in leadership matters less than a change in what decisions are rewarded and independently checked.
Corporate crises are often narrated as failures of a few individuals because names are easier to grasp than incentive systems. Individual accountability still matters, but replacing a chief executive cannot by itself repair reporting lines, board oversight, regulator independence, or rewards tied to short-term results. The families’ experience also shows that harm continues after the accident through opaque investigations and unequal access to justice. A credible recovery therefore needs observable protections: transparent reporting, independent review, meaningful consequences for interference, and a way for affected people to see whether promises changed practice.
Analysis
Flying Blind works as investigative history, engineering explanation, and institutional critique. Its central argument is not simply that cost-cutting caused two crashes. It traces a sequence in which financial incentives narrowed the design choices, weakened the workforce’s ability to challenge them, and made government oversight increasingly dependent on the manufacturer. The MAX’s technical flaw mattered enormously, but the book shows how many separate barriers that should have caught or contained it were compromised: sensor redundancy, test time, pilot training, documentation, internal reporting, and FAA scrutiny.
The strongest contrast is between early Boeing’s willingness to spend money correcting the 707 and the later preference for inexpensive adaptations. That contrast is vivid, but the history should not be mistaken for a golden age without defects. Boeing had earlier failures, design compromises, and institutional blind spots. The useful distinction is not old versus new, or engineers versus managers. It is whether decisions preserve the expertise and independence needed to detect bad assumptions before they become operational hazards.
The book also complicates the usual story of a single villain. Executives set targets, boards approved priorities, engineers made design choices, pilots and regulators handled incomplete information, and airlines sought lower costs. Responsibility was distributed, but not equally. People with authority to fund testing, disclose software, or ground the fleet had more power than those facing the emergency in the cockpit. Distributed causation should not become diluted accountability.
One broader lesson is that safety is an organizational property. It depends on technical safeguards and on conditions that allow people to notice, report, and act on weak signals. Redundancy in hardware is useless if communication is not redundant too. A company can display safety as a value while its incentives quietly reward schedule, cost, and investor returns. The decisive test is what happens when safety competes with those objectives, especially before catastrophe makes the answer obvious.
Review Summary
Flying Blind by Peter Robison explores Boeing's transformation from a safety-focused engineering company to one driven by profits and cost-cutting, culminating in the 737 MAX crashes. Readers praise Robison's thorough research and engaging writing style, finding the book both informative and infuriating. Many commend its exposure of corporate greed and regulatory failures. Some criticize the book's political bias and excessive detail on peripheral characters. Overall, reviewers appreciate the comprehensive look at Boeing's history and the events leading to the tragic crashes, recommending it as an important read on corporate culture and aviation safety.
People Also Read
Glossary
Maneuvering Characteristics Augmentation System (MCAS)
Nose-down flight-control softwareA Boeing 737 MAX flight-control function intended to push the aircraft’s nose down in conditions interpreted as a potential stall. In the MAX design discussed in the book, it could act on faulty angle-of-attack data, repeatedly moving the horizontal stabilizer. Pilots had not been fully informed about the system or trained to recognize its behavior before the crashes.
Organization Designation Authorization
Delegated aircraft certificationAn FAA program that allows aircraft manufacturers to designate employees to perform certain certification tasks on the agency’s behalf. The book describes how the system expanded Boeing’s role in selecting and supervising its own certification representatives, while reducing the direct involvement of FAA specialists. This created concerns about whether the regulator could independently scrutinize the manufacturer’s work.
Boeing Aviation Safety Oversight Office
FAA office overseeing BoeingThe FAA office established in Seattle to oversee Boeing’s delegated safety and certification work. The book describes it as a relatively small group of FAA staff responsible for monitoring a much larger number of Boeing employees performing delegated tasks. Its structure and limited resources raised questions about whether federal officials had enough authority and technical access to challenge Boeing.
angle-of-attack sensor
Measures wing angle to airflowAn exterior sensor that measures the angle between a wing and the air moving past it. Aircraft systems can use this measurement to assess stall risk. On the 737 MAX, faulty data from a single sensor could trigger MCAS. The book notes that a sensor on the Lion Air aircraft was misaligned, and that the MAX’s relevant software did not adequately guard against erroneous readings.
Engine Indicating and Crew Alerting System (EICAS)
Prioritized cockpit alertsA cockpit display system used on Boeing aircraft such as the 757 and 767. It presents system information and fault messages in a more detailed, prioritized form than the warning lights and paper procedures used in older 737 models. The book uses EICAS to illustrate how cockpit design can help pilots understand a problem and choose a response.
program accounting
Spreads costs across deliveriesAn accounting method used by Boeing to estimate revenue and costs across a long production program rather than treating early aircraft as the full measure of a project’s economics. Because aircraft development costs occur upfront while sales continue for years, the method can spread those expenses across expected future deliveries. The book notes that Boeing used it in accounting for the costly 787 program.
FAQ
What's Flying Blind: The 737 MAX Tragedy and the Fall of Boeing about?
- Focus on Crashes: The book examines the tragic crashes of the Boeing 737 MAX, exploring the design flaws and corporate culture that contributed to these disasters.
- Corporate Culture Shift: It critiques Boeing's transition from an engineering-driven company to one focused on shareholder value, impacting safety and accountability.
- Human Element: Personal stories of victims and their families are woven into the narrative, adding depth to the technical and corporate discussions.
Why should I read Flying Blind by Peter Robison?
- Aviation Safety Insight: Offers a detailed look at the balance between safety and profit in the aviation industry, essential for those interested in aviation or corporate ethics.
- Corporate Governance Lessons: Highlights the consequences of prioritizing profits over safety, providing relevant lessons for all industries.
- Engaging Storytelling: Combines investigative journalism with storytelling, making complex topics accessible and engaging.
What are the key takeaways of Flying Blind?
- Safety Culture Importance: Emphasizes the necessity of a strong safety culture in industries where lives are at stake, warning against neglecting this aspect.
- Outsourcing Oversight Risks: Discusses how delegating safety oversight to manufacturers can create conflicts of interest and reduce accountability.
- Corporate Decision Impact: Illustrates how decisions driven by financial metrics can have dire consequences for employees and customers.
What are the best quotes from Flying Blind and what do they mean?
- “The ghost won.”: Highlights the tension between human operators and automated systems, as seen when the 737 MAX's systems overpowered pilot control.
- “This airplane is designed by clowns, who in turn are supervised by monkeys.”: Reflects employee frustration with management decisions that compromised safety.
- “Everybody thinks they’re different.”: Critiques the complacency within Boeing and the FAA, serving as a cautionary reminder of overconfidence.
What led to the crashes of the 737 MAX according to Flying Blind?
- MCAS Software Issues: The Maneuvering Characteristics Augmentation System malfunctioned due to faulty sensor data, leading to the crashes.
- Pilot Training Gaps: Inadequate training on the new software contributed to pilots' inability to regain control during emergencies.
- Regulatory Failures: The FAA's reliance on Boeing for safety oversight allowed critical flaws to go unchecked, compromising aircraft safety.
How did Boeing's corporate culture change over time as described in Flying Blind?
- Shift to Shareholder Value: Transitioned from an engineering-focused company to one prioritizing shareholder returns, leading to cost-cutting measures.
- Management Practices: Rewarded financial performance over engineering excellence, favoring profits at the expense of safety.
- Impact on Employees: Created an environment where employees felt pressured to conform to financial goals, compromising professional integrity.
What role did the FAA play in the 737 MAX crisis according to Flying Blind?
- Regulatory Capture: The FAA's close relationship with Boeing led to a lack of rigorous oversight, allowing the company to self-certify aspects of the MAX.
- Delegation of Authority: Allowed Boeing to appoint its own safety representatives, diluting the agency's authority and oversight capabilities.
- Failure to Act: Despite warnings, the FAA was slow to respond to safety concerns, contributing to the tragic outcomes.
How did the crashes affect Boeing's reputation as discussed in Flying Blind?
- Loss of Trust: Severely damaged public trust in Boeing, shifting its perception from a symbol of American ingenuity to one of corporate negligence.
- Financial Consequences: Grounding of the 737 MAX and investigations cost Boeing billions, impacting its stock price and market position.
- Cultural Reflection: Prompted discussions about corporate ethics and accountability, leading to calls for regulatory reforms.
What is the MCAS system discussed in Flying Blind?
- Purpose of MCAS: Designed to enhance handling characteristics of the 737 MAX, particularly during high-speed maneuvers, by preventing stalls.
- Design Flaw: Relied on data from a single sensor, creating a critical vulnerability that could lead to loss of control.
- Consequences: The flawed design contributed to two fatal crashes, highlighting the importance of redundancy in safety systems.
How did Boeing's management respond to the crises according to Flying Blind?
- Leadership Changes: Underwent significant leadership changes, including the ousting of CEO Dennis Muilenburg, to restore confidence.
- Crisis Management Strategies: Implemented public relations campaigns and safety reviews, though often criticized as insufficient.
- Focus on Safety Culture: Renewed emphasis on rebuilding safety culture, with promises to prioritize engineering excellence and accountability.
What changes were made to the 737 MAX after the crashes as per Flying Blind?
- Software Updates: Implemented changes to the MCAS software, including using data from two sensors to prevent erroneous activations.
- Pilot Training Requirements: FAA mandated simulator training for the MAX, reflecting a commitment to improving pilot preparedness.
- Regulatory Reforms: Legislative efforts aimed to reform FAA's oversight practices, enhancing safety and preventing similar incidents.
What does Flying Blind say about the future of Boeing?
- Challenges Ahead: Faces significant challenges in restoring its reputation and ensuring aircraft safety, with lasting scars from the MAX crisis.
- Cultural Shift Needed: Emphasizes the need for a cultural shift prioritizing safety and ethical decision-making over profits.
- Long-Term Impact: Indicates that the MAX crisis will have long-term implications for Boeing's business model and operational practices.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.