Random the Book

Questions for you:

• When you make routine decisions (which conference to attend, where to shop, whether to travel whilst feeling slightly unwell), do you consider the potential exponential consequences through networks you’re connected to?
• In your organisation’s risk planning, do you account for how trivial individual choices by employees, customers, or partners could cascade into systemic crises?
• Looking at past crises you’ve faced, how many originated from random individual decisions or chance encounters, rather than predictable systemic failures executed by obvious culprits?
• If preventing global crises depends on countless unpredictable individual choices, what does that suggest about the limits of surveillance, control, and traditional risk management – and, therefore, how will you manage and consider that risk, if at all?

Questions for your organisation:

Network vulnerability and cascade analysis: Organisations are networks where individual actions can have cascading exponential effects. One employee clicking a phishing link can compromise entire systems. One supplier failure can shut down production. One customer complaint going viral can damage a reputation globally. Map network structures, identifying points where individual random actions could trigger cascades. Unlike pandemic spreads, which are genuinely unpredictable, organisational networks can be redesigned with “circuit breakers” preventing localised failures from becoming systemic crises.

Fragility versus robustness in connected systems: Highly connected, efficient systems are maximally vulnerable to random shocks. Just-in-time supply chains fail when one random link breaks. Open-plan offices amplify disease transmission. Centralised IT systems create single points of failure. Build redundancy and slack specifically to absorb random shocks. Less efficient systems are more robust – multiple suppliers, distributed infrastructure, and departmental autonomy all reduce vulnerability to arbitrary individual events triggering crises.

Early warning systems acknowledging randomness: Traditional risk management identifies known threats and plans their responses. But crises often emerge from random initial conditions impossible to predict. Instead of trying to predict which specific random event will trigger problems, build monitoring systems capable of detecting early cascade patterns regardless of origin. Focus on detecting exponential growth (complaints, errors, infections, failures) early rather than predicting specific triggering events and implementing specific solutions.

Individual responsibility and systemic fragility: When organisational crises can originate from random individual actions, two approaches exist: (1) Extensive monitoring and control of individual behaviour (surveillance, restrictions, procedures), or (2) System redesign making organisations robust to random individual variations. The first approach is expensive, demoralising, and ultimately fails because you cannot control all random actions. The second accepts randomness as inevitable and builds systems that don’t catastrophically fail when individuals inevitably make arbitrary choices.

Further reading

Pandemics, networks, and exponential spread

The Rules of Contagion by Adam Kucharski – epidemiologist’s explanation of how disease spreads through networks, demonstrating how random individual movements and encounters determine whether outbreaks remain localised or become pandemics.

Spillover by David Quamman – traces origins of zoonotic diseases showing arbitrary nature of pandemic triggers, where random encounters between humans and infected animals in specific locations at specific times alter global history.

The Pandemic Century by Mark Honigsbaum – historical analysis of pandemic origins revealing how trivial decisions and chance encounters (sick soldiers, market visits, travel choices) triggered worldwide health crises affecting billions.

Network effects and cascade failures

• Linked by Albert-László Barabási – explains network science showing how highly connected networks amplify random local events into global cascades, relevant to understanding why individual actions can trigger organisational or societal crises.
• The Black Swan by Nassim Nicholas Taleb – argues that rare, unpredictable events with massive consequences dominate history, demonstrating how systems vulnerable to random shocks (like pandemic spread) require robustness rather than prediction.
• Normal Accidents by Charles Perrow – demonstrates that in tightly coupled complex systems, random small failures inevitably cascade into major disasters, showing why preventing specific triggering events is less effective than building resilient systems.

Fragility, robustness, and systemic risk

• Antifragile by Nassim Nicholas Taleb – distinguishes systems that break under random shocks from those that benefit from them, arguing for building redundancy and optionality specifically to handle unpredictable individual events triggering cascades.
• The Resilience Dividend by Judith Rodin – examines how cities and organisations build resilience to unpredictable shocks, showing that robust systems accept randomness as inevitable rather than attempting comprehensive control.
• Drift into Failure by Sidney Dekker – demonstrates how complex systems gradually become vulnerable to random triggering events through accumulated small decisions, arguing that resilience requires understanding systemic fragility rather than controlling individual actions.

Interactive exhibit

Watch how infections spread amongst populations with our Exponential Cascade Simulator: https://experiments.randomthebook.com/infection/i

About the image

This thermometer from our medicine cupboard saw a lot of use during the pandemic.

Photo montage and photo by Matt Ballantine, 2026