Key Takeaways
Join the databases departments already own before hiring more inspectors.
A fire exposed the gaps. In 2011, three people died in a Bronx apartment divided by illegal walls. New York had hundreds of inspectors facing thousands of complaints and millions of properties. Mike Flowers and a small data team looked beyond the Fire Department's records, combining information held by building, finance, and other agencies. They found that risky homes were more likely to be old, located in poor neighborhoods, tied to mortgage or tax defaults, and flagged for building violations.
Joined-up data improved targeting. Inspectors who previously found problems in about 13 percent of the buildings they checked found them in roughly 70 percent after using the combined risk model. The city also applied data matching to identify likely tax evasion, illegal drug sales, and restaurant grease dumping. The gain came not from collecting more information, but from connecting information already scattered across departments.
What's striking is that the breakthrough depended on pairing computation with fieldwork. The team did not merely search for statistical patterns; it rode along with inspectors and learned what a dangerous building looked like in practice. That combination resembles good applied science: models generate hypotheses, and observation tests whether the variables mean anything in the world. The approach also carries a modern caution. Data assembled for one purpose can expose people to new forms of surveillance when reused for another. Better prediction is not automatically fairer policy, so agencies need transparent criteria, review, and ways to challenge mistaken classifications.
Question the categories that make your organization's habits feel natural.
Classification helps people think. The book calls an isolated department, social group, or mental worldview a silo. People need categories to manage complexity, but categories can conceal alternatives and make inherited arrangements seem inevitable. Anthropologist Pierre Bourdieu saw this in a French village where older farming men stood apart from younger dancers. Economic change had made the farmers less attractive marriage prospects, yet the villagers treated the division as ordinary rather than as a social pattern they could question.
Habits shape perception. Bourdieu used the term habitus for the learned dispositions and surroundings that shape how people act and interpret the world. In a Kabyle home, the division of space between men, women, guests, and animals reflected wider social beliefs, then reinforced them through daily routine. An insider-outsider perspective, knowing a world while also gaining distance from it, can reveal these patterns.
This perspective connects organizational design to cognitive science: mental schemas reduce the effort of interpreting a crowded world, but they also direct attention toward familiar evidence. The village story makes that tradeoff vivid. Categories are not simply mistakes to be erased; they help coordinate expectations and make action possible. The useful question is whether a category still fits the conditions it was built to manage. Leaders can encourage that question by asking who benefits from a boundary, what it leaves out, and whether people would draw it the same way from another vantage point.
Keep specialist teams, but stop their scorecards from turning them into rivals.
Sony's structure changed its behavior. The company was once known for products such as the Walkman, created through debate across engineering and marketing. As it grew, leaders divided it into increasingly independent business units, each judged on its own financial results. That improved accountability and helped cut debt, but it also encouraged managers to defend their unit and withhold ideas.
The digital music opportunity split apart. In 1999, Sony presented multiple competing music players built by separate groups, using incompatible technology. Its music division worried that online distribution would hurt existing sales, while electronics and computer teams pursued their own devices. Apple coordinated hardware, software, and music sales into a single offer, and its iPod overtook Sony's products. Howard Stringer tried restructuring Sony and urging collaboration, but public commands could not overcome entrenched incentives and resistance.
Sony shows why structure and incentives matter more than slogans. A company can praise collaboration while rewarding managers for protecting local profit and control. That is a predictable form of goal displacement: teams optimize what their scorecards measure, even when the whole organization loses. Yet integration has costs too. Centralizing every decision can suppress the initiative that made Sony's early products distinctive. The design challenge is not to abolish specialist units, but to give them shared measures, routes for exchanging ideas, and authority to build products that cross their boundaries.
Audit the labels that make dangerous assets look safe.
UBS trusted its own categories. Before the financial crisis, the bank's New York team accumulated mortgage-linked collateralized debt obligations, or CDOs, complex bundles of loans repackaged as securities. Many received AAA ratings and were classified as safe, temporary client business. A separate London team held bets against the housing market. Because the bank did not assemble a complete view of both positions, senior managers and regulators heard reassuring reports that UBS was protected.
The risks were hidden in plain sight. The New York CDO warehouse grew to more than $20 billion, yet senior leaders and risk officers did not understand its scale. When housing markets turned, UBS suffered losses of roughly $19 billion on mortgage investments and required public support. The failure was not just an absence of information. Incentives discouraged teams from sharing it, while accounting categories and risk models made the exposures appear safer than they were.
The case illustrates a subtle danger in measurement: a classification can become an excuse to stop asking questions. A high credit rating or a client-facing label may describe one feature of an asset, but it does not prove that the asset is safe under every scenario. Modern risk governance therefore needs both aggregation and challenge. A single dashboard can reveal correlated exposures, but independent reviewers must still question its assumptions and definitions. The book also avoids a simplistic blame story: silos mattered, but ambition, weak oversight, and distorted pay helped create the conditions in which those silos became consequential.
Make experts map the system beyond the borders of their job.
Economists watched only part of finance. Before the 2008 crisis, many central bankers focused on inflation and the so-called real economy, while financial regulators watched individual banks. New credit vehicles outside traditional banks fell between those areas of responsibility. Paul Tucker at the Bank of England noticed that broad money was expanding rapidly, but lacked enough data and research to explain the growth of these unfamiliar institutions.
A name made the hidden system visible. At a 2007 conference, investor Paul McCulley popularized the term shadow banking system for financial intermediaries that performed bank-like work outside ordinary bank structures. Naming the system helped researchers and policymakers discuss its links to mortgage securities and short-term funding. Estimates of its size varied widely, but the crisis showed that this overlooked network was large enough to threaten the wider economy. Experts had pieces of the picture, but too few connected them.
The episode demonstrates that expertise can fail through division of labor, not simply through ignorance. Each group may perform its assigned task competently while the system-level danger remains nobody's job. This resembles a problem in safety engineering: local controls can work as designed while interactions among components create a hazard. Broader mandates and cross-disciplinary teams can help, but they can also blur accountability. Effective coordination therefore needs named responsibility for system-wide risks, shared information, and permission for specialists to raise concerns outside their formal remit.
Cross into unfamiliar work to notice what your own world hides.
Brett Goldstein crossed a professional border. After working at the restaurant-booking company OpenTable, he joined the Chicago Police Department, despite having little in common with its traditional recruits. His technology experience later helped him combine reports about gang movements, shootings, arrests, and locations. The resulting maps showed how conflicts moved between neighborhoods and helped police direct units toward places where violence might erupt.
The experiment changed how information moved. Police had often relied on reports traveling up and down separate chains of command. Goldstein's system brought data from different districts into a shared view, and the city's homicide count fell during the period described. The book does not establish that the maps alone caused the decline. The program later lost support amid political conflict, budget pressures, and concerns about racial patterns in the predictions.
Crossing into another profession can supply more than new technical skills. It exposes assumptions that insiders no longer notice and creates translators who can connect different forms of knowledge. But predictive policing also shows why a fresh perspective is not automatically a neutral one. Historical crime data reflects where enforcement occurred, as well as where harm occurred. If those data guide more patrols to the same neighborhoods, the system can reinforce the pattern it claims merely to forecast. Useful analytics need independent fairness checks, community scrutiny, and clear limits on how predictions affect people.
Build cross-team ties on purpose, not just open offices.
Facebook designed repeated encounters. As the company grew beyond the roughly 150-person group associated with Robin Dunbar's research on human social networks, its engineers worried that project teams might become isolated. New hires went through a shared six-week Bootcamp before choosing a team. Hackamonth rotations let experienced engineers work elsewhere for a period, while hackathons brought people from different groups together to build something outside their usual assignments.
Architecture and rituals reinforced the ties. Facebook used open spaces, visible meeting rooms, shared events, and internal social posts to make colleagues more familiar to one another. The goal was not to eliminate project teams, which were necessary for focused work, but to keep communication flowing between them. The company still stumbled, including by being slow to adapt its products for mobile devices. Its culture of experimentation was intended to help it recognize and correct such failures.
The strongest idea here is social redundancy: employees need more than one path to colleagues beyond their immediate team. Research on psychological safety, the shared belief that a group can take interpersonal risks, helps explain why cross-team contact matters. Familiarity can make it easier to ask for help or admit a mistake. Yet open offices alone rarely create trust, and forced interaction can become performative. Facebook's rotations and collaborative projects are more substantive because they give people shared work and recurring contact. Their cost in time and coordination is real, not evidence that the approach has failed.
Organize services around people's problems, not professionals' titles.
Cleveland Clinic redrew its medical map. CEO and surgeon Toby Cosgrove was challenged by a patient family who valued the hospital's technical skill but found it lacking in empathy. He began asking whether medicine should be organized around doctors' specialties or around what patients experienced: a painful head, a heart problem, or a possible cancer. In 2008, the clinic replaced many traditional department divisions with 27 multidisciplinary institutes organized around body systems and illnesses.
The new map encouraged collaboration. Surgeons and physicians who had often worked apart were brought together, while all staff received empathy training and were described as caregivers. The clinic also retained overlapping structures for professional training and insurance billing, so the old categories did not simply disappear. The book reports improved patient satisfaction and examples of more coordinated treatment, including faster hospital admissions through expanded authority for emergency doctors.
Reclassifying work around the person receiving a service is a powerful test of whether an organization's boundaries serve users or merely reflect professional history. The same question applies to public agencies and product companies. Still, a patient-centered chart cannot by itself resolve conflicts over expertise, status, or resources. Cleveland Clinic's salaried compensation model and existing collaborative traditions made its redesign more feasible than it might be elsewhere. The useful lesson is to pair a new taxonomy with changes in incentives, decision rights, and routine collaboration, while preserving the specialist knowledge that good care requires.
Study other people's silos to find the prices and products they miss.
A hedge fund treated boundaries as clues. BlueMountain Capital studied how banks divided securities, teams, and responsibilities. Those divisions could distort prices because one desk might be allowed to trade a whole financial bundle while another could trade only its pieces. In 2012, JPMorgan's London traders built enormous positions in credit derivatives. BlueMountain recognized that the bank's demand for an index was affecting its price, took positions in the opposite direction, and later helped JPMorgan unwind the trades. Reports estimated BlueMountain made about $300 million across its dealings related to the episode.
The same method crossed asset classes. BlueMountain combined credit and equity analysis to reassess HanesBrands, which many investors treated as a risky cyclical retailer. The team argued that its underwear business generated steadier cash flows, bought its shares, and saw the price roughly double by summer 2013. Silo-busting was not only defensive. It could reveal opportunities created by other organizations' rigid categories.
BlueMountain turns the book's diagnosis into a competitive method: when institutions divide a system into artificial compartments, their blind spots can show up as mispriced assets or neglected products. This resembles arbitrage, the practice of profiting when related prices diverge, while adding close attention to organizational incentives. The approach is not automatically socially beneficial. Traders who exploit a mispricing may help prices reflect information, but aggressive positions can also intensify instability, as the JPMorgan episode itself suggests. The transferable lesson is not to imitate hedge-fund trading; it is to ask what the prevailing categories make people unable or unwilling to compare.
Analysis
The book's central argument is that silos are not simply boxes on an organizational chart. They are arrangements of attention. Departments, professions, incentives, jargon, and familiar social habits jointly determine which facts become visible and which remain peripheral. The New York fire team could not improve inspections until it connected data across agencies. UBS could not see the scale of its mortgage exposure because local positions, risk labels, and reporting structures concealed their combined effect. Economists could study individual financial innovations while failing to recognize the system they formed. Across these stories, blindness is produced less by a shortage of intelligence than by the way intelligence is partitioned.
The account is strongest when it treats integration as a practical design problem rather than a moral slogan. Sony's incentives, Facebook's rotations, Cleveland Clinic's multidisciplinary institutes, and BlueMountain's cross-asset analysis show that information sharing depends on real mechanisms: common measures, repeated contact, shared authority, and translators who understand more than one professional language. Simply putting everyone in an open office or publishing more data cannot substitute for those mechanisms. Nor should every specialist boundary be removed. Specialization protects focus and expertise; the task is to make boundaries permeable where work or risk crosses them.
A useful extension is to distinguish three failure points. First, categories can be outdated, so systems need periodic reclassification. Second, incentives can make useful cooperation personally costly, so appeals to teamwork will have limited force without reward changes. Third, integration can produce new harms. Predictive policing may reproduce biased patterns, while pooled personal data can enable intrusive uses. Joined-up systems need accountability and contestability alongside better coordination.
The cases are vivid, but they are not controlled comparisons. Falling crime or improving patient satisfaction cannot be attributed to one intervention from these accounts alone. The book is best read as a diagnostic lens, not proof that one organizational remedy works everywhere. Its durable contribution is the question it teaches readers to ask: what important relationship disappears when a problem is divided into the categories that currently feel obvious?
Review Summary
The Silo Effect explores how organizational silos can lead to poor decision-making and missed opportunities. Tett uses case studies to illustrate the dangers of silos and strategies for breaking them down. Many readers found the book insightful and relevant, praising Tett's anthropological perspective. Some felt the anecdotes were too lengthy or that the book lacked concrete solutions. Overall, reviewers appreciated Tett's analysis of how silos form and their impact on organizations, though opinions varied on the book's depth and practicality.
Glossary
silo effect
Isolation across teams or mindsThe silo effect is the fragmentation of people, information, or ways of thinking into groups that operate with limited connection to one another. It can arise from formal departments, professional tribes, or mental categories. Silos help organize complexity, but when boundaries become rigid they can block cooperation, hide combined risks, and prevent people from noticing opportunities.
habitus
Learned patterns shaping perceptionPierre Bourdieu's term for the dispositions and expectations people acquire through their social surroundings. Habitus shapes how people behave, interpret situations, and regard certain arrangements as normal. It is reinforced by everyday routines and physical environments, but does not determine behavior absolutely. Becoming aware of these learned patterns can help people question inherited ways of classifying the world.
insider-outsider
Familiarity with analytical distanceAn insider-outsider perspective combines knowledge gained from participating in a social world with enough distance to see its assumptions. Gillian Tett draws on Bourdieu's movement between French village life, elite Paris, and Algeria to show how crossing boundaries can reveal patterns that insiders take for granted. The perspective helps people examine their own institutions as well as unfamiliar ones.
participant observation
Learning by living alongside peopleA method associated with cultural anthropology in which a researcher spends time within a community, observes daily life, and participates in ordinary activities. Rather than relying only on formal accounts or distant data, the researcher learns how people actually behave and what shared practices mean to them. The method offers context for interpreting patterns that might otherwise be misunderstood.
Dunbar's number
Estimated limit for close social groupsRobin Dunbar's proposal that humans can maintain a stable circle of roughly 150 meaningful social relationships, based on research connecting primate brain size with social group size. The book describes Facebook's leaders using the idea to think about how rapid growth could weaken personal ties inside the company. The estimate is debated, and should be treated as a theory rather than a universal fixed limit.
shadow banking system
Bank-like finance outside conventional banksThe shadow banking system refers to financial intermediaries and arrangements that move credit outside the traditional deposit-taking banking system. In the book, it includes entities such as structured investment vehicles and conduits that raised short-term funds and invested in longer-term securities. Their links to banks and mortgage markets made their risks consequential, although their structures were difficult for many policymakers to track.
Hackamonth
Temporary cross-team engineering rotationHackamonth was Facebook's practice of temporarily moving engineers to work on a different project or team, often one far from their usual assignment. The rotation was intended to renew employees' interest and expose them to unfamiliar work. It also created connections across formal team boundaries, making it easier for engineers to exchange knowledge after they returned to their regular roles.
cultural translators
People who connect specialist groupsCultural translators are employees who can understand and communicate across multiple professional specialties. They help one group interpret another group's terminology, priorities, and findings, reducing the chance that important information will be ignored because it arrives in an unfamiliar language. The book argues that large organizations need some such boundary-spanning people, even while most staff remain focused specialists.
FAQ
What's The Silo Effect about?
- Exploring organizational silos: The Silo Effect examines how isolated departments within organizations can lead to inefficiencies and failures. Gillian Tett argues that these silos create tunnel vision, blinding teams to external risks and opportunities.
- Anthropological perspective: Tett uses her background in anthropology to analyze how cultural norms shape organizational behavior. She provides stories from various sectors to illustrate the pervasive nature of silos.
- Solutions offered: The book not only identifies issues caused by silos but also offers insights into breaking down these barriers. Tett emphasizes communication and collaboration as key strategies.
Why should I read The Silo Effect?
- Understanding organizational behavior: The book is essential for those interested in how organizations function and the hidden barriers to success. It's particularly relevant for leaders aiming to foster innovation.
- Real-world examples: Tett uses compelling narratives from finance, healthcare, and government to illustrate her points, making the concepts relatable and practical.
- Cultural insights: By incorporating anthropological perspectives, Tett encourages critical thinking about cultural norms influencing organizational behavior, aiding in navigating complex social dynamics.
What are the key takeaways of The Silo Effect?
- Silos create blind spots: Silos can lead to a lack of awareness about risks and opportunities, as seen in UBS during the financial crisis, with dire consequences.
- Importance of communication: Breaking down silos requires open communication and collaboration across departments, fostering innovation and adaptability.
- Cultural change is necessary: Organizations must undergo cultural shifts to combat the silo effect, promoting flexibility and interconnectedness.
What are the best quotes from The Silo Effect and what do they mean?
- “We can be blind to the obvious, and we are also blind to our blindness.”: This quote by Daniel Kahneman highlights the theme of the book, emphasizing the need for vigilance and open-mindedness in decision-making.
- “Everything here is arranged in a fragmented way. It’s tough to join it all up.”: Tett uses this to illustrate the challenges of integrating organizational functions, stressing the need for leaders to counteract fragmentation.
- “Breaking down silos isn’t about a series of actions but an attitude of mind.”: This quote from Paul Tucker suggests that overcoming silos requires a cultural shift, fostering curiosity and collaboration for long-term success.
How does Gillian Tett define "silos" in The Silo Effect?
- Cultural phenomenon: Silos are both physical structures and mental states arising from how people classify and organize their environments, leading to ineffective communication.
- Impact on behavior: Silos breed tribalism and tunnel vision, causing individuals to focus solely on their expertise, resulting in missed opportunities and increased risks.
- Need for balance: While specialization is necessary, Tett argues for a balance between specialization and collaboration to foster innovation and adaptability.
What are some examples of organizations affected by the silo effect in The Silo Effect?
- UBS and the financial crisis: UBS's fragmented structure led to a lack of communication, resulting in significant losses during the 2008 financial crisis.
- Sony's decline: Sony's isolated product divisions missed opportunities for innovation, leading to competing products that confused consumers and diluted the brand.
- Cleveland Clinic's success: The clinic actively worked to break down silos, improving patient care by encouraging collaboration among specialists.
How can individuals and organizations break down silos according to The Silo Effect?
- Encourage cross-functional teams: Organizations should create opportunities for employees from different departments to collaborate, fostering innovation.
- Promote open communication: Leaders should encourage transparency and dialogue, helping to identify risks and opportunities that may go unnoticed.
- Adopt an anthropological perspective: Using anthropological methods to analyze organizational culture can help develop strategies for a more interconnected environment.
How does The Silo Effect relate to the Great Financial Crisis?
- Fragmentation in finance: The silo effect played a significant role in the 2008 financial crisis, as banks failed to communicate effectively across departments.
- Case studies of failure: UBS is a primary example of how silos led to catastrophic failures, with significant losses due to a lack of information sharing.
- Lessons for the future: Tett emphasizes the need for organizations to learn from past mistakes by fostering collaboration to navigate future challenges.
What strategies does Tett suggest for fostering innovation in organizations?
- Embrace diversity of thought: Creating environments where diverse perspectives are valued can lead to creative problem-solving and innovative solutions.
- Implement flexible structures: Avoiding rigid hierarchies and adopting fluid structures can promote collaboration and communication across departments.
- Encourage experimentation: Organizations should create a safe space for employees to experiment with new ideas, driving innovation and adaptability.
What role does cultural anthropology play in The Silo Effect?
- Analyzing classification systems: Tett uses cultural anthropology to explore how humans classify environments, illuminating how silos arise and impact behavior.
- Understanding social dynamics: Applying anthropological methods helps examine cultural norms shaping organizational behavior, leading to a deeper understanding of silos.
- Insights from case studies: The book features case studies illustrating cultural anthropology principles, providing lessons on breaking down silos and fostering collaboration.
How does The Silo Effect suggest organizations can use technology to combat silos?
- Centralizing data: Creating centralized databases allows easy access to information across departments, ensuring informed decision-making.
- Enhancing communication: Technology can improve communication channels, enabling employees to share information and collaborate more effectively.
- Data-driven decision-making: Analyzing data from multiple sources helps organizations identify trends and opportunities, transcending departmental boundaries.
What are some examples of successful silo-busting from The Silo Effect?
- Chicago Police Department: Brett Goldstein used data analytics to improve crime response, centralizing information and fostering collaboration.
- Cleveland Clinic: The clinic restructured to focus on patient-centered care, improving satisfaction and health outcomes.
- Facebook: Programs like Bootcamp and Hackamonth encourage cross-departmental collaboration, maintaining a culture of creativity and adaptability.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.