Contents
Natural Sciences
Section 1
The Core Idea
A chain reaction is a sequence in which each step triggers the next. One event causes another, which causes another; the reaction propagates. In nuclear physics, a neutron hits a nucleus and releases more neutrons, which hit more nuclei — the reaction sustains and can grow. In chemistry, a free radical can initiate a cascade of reactions. In strategy and organisations, the same structure appears: one default triggers margin calls that trigger more defaults; one piece of content goes viral and triggers more sharing; one adoption triggers a wave of imitators. The key is that the output of step N is an input to step N+1. The reaction is self-propagating.
Chain reactions can be controlled or uncontrolled. A controlled chain reaction (e.g. a nuclear reactor) is sustained at a steady rate. An uncontrolled one (e.g. a nuclear bomb, a bank run) grows until it runs out of fuel or hits a limit. The strategic question is whether you are inside a chain reaction (and want to sustain or dampen it) or trying to start one (and want to initiate and fuel it). Viral growth, financial contagion, and technology adoption can all be modelled as chain reactions: each adoption or event increases the probability or rate of the next.
The difference between a chain reaction and a simple cascade is that in a true chain reaction, the products of the reaction feed back into the reaction — they are both output and input for the next step. That creates the possibility of exponential growth or collapse. Use the model when you see propagation that depends on each step producing the trigger for the next. Mitigation often means breaking the chain — inserting a barrier or absorbing the propagating element so the next link never fires.
Section 2
How to See It
Look for sequences where each event causes the next in a way that could sustain or accelerate. The tell is propagation: the reaction moves through the system, and the rate of propagation can depend on how much "fuel" (e.g. susceptible nodes, uncommitted capital) remains.
Business
You're seeing a Chain Reaction when one high-profile customer adoption triggers a wave of similar customers — "if they use it, we should too." Each adoption increases visibility and legitimacy, which triggers more adoptions. The reaction propagates through the market until saturation or a competing reaction.
Technology
You're seeing a Chain Reaction when a failure in one service causes dependent services to fail, which cause more failures. The outage propagates. Each failure is both effect and cause of the next. The chain continues until redundancy or isolation stops it.
Investing
You're seeing a Chain Reaction when a price drop triggers margin calls or redemptions, which force selling, which triggers more price drops. The reaction is self-reinforcing. The same structure can run in reverse in a bubble: each buyer attracts more buyers.
Markets
You're seeing a Chain Reaction when a standard or technology gains adoption because each new adopter increases the value for the next (network effects). The "reaction" is adoption; each adoption is the product that triggers more adoption. The chain runs until the market saturates or a rival chain wins.
Section 3
How to Use It
Decision filter
"When you see propagation — one thing leading to the next in sequence — ask whether it is a chain reaction: does each step produce the trigger for the next? If yes, you can try to start it, sustain it, or break it. If you are inside one, identify the links and where to intervene."
As a founder
Viral and referral loops are chain reactions: each user brings more users. Your job is to initiate the chain (first users, first content, first proof) and to design the mechanism so each step produces enough "output" to trigger the next (e.g. shareable experience, incentive to refer). The mistake: assuming the chain will run without the right design. Each link must reliably pass the trigger. The second mistake: ignoring the reverse chain. Churn can be a chain reaction (one departure reduces value for others); design to break or slow that chain.
As an investor
Financial and market chain reactions (contagion, bubbles) are critical to understand. When you see propagation — selling begetting selling, or FOMO begetting FOMO — ask how long the chain can run and what would break it. Position for the chain when it is early and sustainable; avoid being the last link when it reverses.
As a decision-maker
In crises, chain reactions are common: one failure or rumor triggers the next. The response is to break the chain — contain the first link, communicate to stop propagation, or isolate so the reaction cannot reach the next node. In positive contexts (e.g. change initiatives), design so that each success triggers the next — early wins that create momentum.
Common misapplication: Calling any sequence a "chain reaction" when the steps do not feed back. A simple cause-effect chain (A causes B causes C) is a cascade; a chain reaction requires that the output of the reaction be an input to its continuation. Second misapplication: Assuming chain reactions run forever. They run until fuel is exhausted, a barrier is hit, or a competing reaction dominates. Model the fuel and the limits.
Section 4
The Mechanism
Section 5
Founders & Leaders in Action
Reed HastingsCo-founder & CEO, Netflix
Netflix's shift to streaming and then to original content was designed to create a chain reaction: each hit attracts subscribers, which fund more content, which attracts more subscribers. The recommendation system adds another chain — each view improves the model, which improves recommendations, which drives more viewing. Hastings understood that growth could be a self-propagating reaction if each step fed the next.
Elon MuskCEO, Tesla & SpaceX
Musk uses chain reactions in narrative and product: each successful launch or delivery reduces perceived risk and attracts the next customer or partner. Starlink's growth depends on each new user improving coverage and economics for the next. He designs for propagation — each step must make the next step more likely.
Section 6
Visual Explanation
Chain Reaction — Each step produces the trigger for the next. The reaction propagates; rate depends on how many triggers each step produces (e.g. >1 = growth).
Section 7
Connected Models
Reinforces
Compounding
Compounding is a chain reaction in time: each period's output (interest, growth) becomes input for the next. The structure is the same — output feeds back as input — and the result can be exponential growth.
Reinforces
Critical Mass
Critical mass is the point at which a chain reaction becomes self-sustaining. Below critical mass, the reaction fizzles; above it, it propagates. Reaching critical mass is the goal when you are trying to start a chain reaction.
Reinforces
Second-Order Effects
Second-order effects are the consequences of consequences. A chain reaction is a sequence of second-order (and higher) effects: each effect is the cause of the next. Modelling the chain is modelling second-order (and nth-order) effects.
Leads-to
Tight Coupling
Tight coupling between components allows a chain reaction to propagate quickly — one failure or change immediately affects the next. Loose coupling can slow or break the chain. Architecture choices determine how easily reactions propagate.
Leads-to
Information Cascade
An information cascade is a sequence where each agent observes and follows prior actions. A chain reaction is a cascade where the output of each step is the input to the next in a way that sustains or amplifies the sequence. Cascades and chain reactions share the structure of dependent propagation.
Tension
[Lead Domino](/mental-models/lead-domino)
Lead domino is the first piece that starts the cascade. It is the initiator of a chain reaction; the tension is that identifying the lead domino does not guarantee the chain will run — fuel and linkage matter.
Section 8
One Key Quote
— Enrico Fermi"In the chain reaction the neutrons produced by the fission of one uranium nucleus can hit other uranium nuclei and cause them to fission, and so on. The reaction can be self-sustaining."
Self-sustaining is the defining property: the products of the reaction are the triggers for more reaction. Strategy is the same — design so that each step produces the trigger for the next.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Start the chain; then fuel it. The hardest part is the first link — the first users, the first proof, the first adoption. Once the chain is running, each step can feed the next. Design the mechanism so that the output of step N is a reliable input to step N+1.
Know when you are inside a chain reaction. In markets and organisations, propagation can be positive (viral growth, cultural change) or negative (contagion, panic). If you are in a negative chain, the priority is to break a link — contain, communicate, or isolate — before the reaction runs to completion.
Model the fuel and the limits. Chain reactions run until they run out of fuel (susceptible nodes, capital, attention) or hit a barrier. Estimate how far the chain can go and what would stop it. That tells you whether to ride it or exit.
Summary: A chain reaction is a sequence where each step triggers the next. Use it to design self-propagating growth or to break negative propagation. The key is whether the output of each step is the input to the next.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A referral program pays $10 for each new user. Each new user refers 0.3 users on average. Referrals grow for a while then plateau.
Scenario 2
A data center has a cooling failure. One rack overheats and fails. Adjacent racks then fail. The outage spreads to the whole hall.
Section 11
Top Resources
01
Historical
First controlled nuclear chain reaction; the physics of self-sustaining propagation.
02
Book
How products design chain reactions in user growth (viral loops).
03
Concept
How financial distress propagates through a system — a negative chain reaction.
04
Book
Popular treatment of how ideas and behaviours propagate — chain reactions in social systems.
05
Reference
Mathematical models of propagation (epidemics) as chain reactions through networks.
Why this matters next
mental modelsNetwork Effects
Chain Reaction applied the Network Effects mental model
mental modelsEpidemic Models
Chain Reaction applied the Epidemic Models mental model
mental modelsCompounding
Chain Reaction applied the Compounding mental model
mental modelsMomentum
Chain Reaction applied the Momentum mental model
mental modelsNarrative
Chain Reaction applied the Narrative mental model
mental modelsChain Reaction
Chain Reaction applied the Chain Reaction mental model
Frequently asked questions
What is Chain Reaction?
Chain Reaction is a mental model used for better thinking and decision-making.
How do you apply Chain Reaction?
To apply Chain Reaction, identify situations where this framework is relevant, then use it as a lens to evaluate your options and decisions. The model is most useful when combined with other complementary mental models.
What category does Chain Reaction fall under?
Chain Reaction falls under the Natural Sciences category of mental models. Other models in this category can be found on the Natural Sciences hub page.
Why is Chain Reaction important?
Chain Reaction is important because it provides a structured way to think about problems that would otherwise be approached with intuition alone. Understanding this model helps you avoid common reasoning errors and make better decisions.
Continue exploring
AECSCMENFLIN
Mental model
Adaptation & Red Queen Effect
In competitive environments, continuous adaptation is required just to maintain
Mental model
Complex Adaptive Systems
Systems composed of many interacting agents that self-organise, adapt, and produ
Mental model
Critical Mass
The minimum threshold of resources, users, or energy needed for a process to bec
Mental model
Entropy
All ordered systems tend toward disorder over time unless energy is continuously
Mental model
Flywheel
A self-reinforcing cycle where each push builds momentum — no single action crea
Mental model
Incentives
People respond to what they are rewarded or punished for — not to what they are
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.