VIII · 40 models
Fixing problems
Why does this keep going wrong, and how do I fix it for good?
Models for fixing problems and systems.
Finding the real cause
Look past the symptom to what is really going on.
Black Box
When you cannot see how a system reaches its answers, you cannot easily spot its mistakes.
First Principles Thinking
Strip a problem back to what you know is true, then build up again from there.
Inversion
To find the way to success, first work out what would guarantee failure, then avoid it.
Lateral Thinking
When digging harder gets you nowhere, the answer is often to dig somewhere else.
Maslow’s Hammer
If the only tool you know is a hammer, every problem starts to look like a nail.
Post Mortem
After something goes wrong, look honestly at how it happened, so it never happens again.
Proximate vs Root Cause
The trigger that sets off a problem is rarely the condition that made it possible.
Reframe the Problem
Changing the question can turn an impossible problem into a solvable one.
The 5 Whys
The first explanation for a problem is rarely the deepest, so keep asking why.
How systems push back
Loops, balances and bottlenecks that decide how a system responds.
Bottlenecks
A system can only move as fast as its narrowest point, so find that point first.
Ecosystems
Nothing acts alone: a change in one place travels along links you cannot see.
Feedback Loops
When a system’s results feed back into it, they either steady it or spin it out of control.
Homeostasis
Stability is not stillness: healthy systems stay steady by correcting themselves all the time.
Hysteresis
Remove the cause of a problem, and its effects may still linger.
Le Chatelier’s Principle
Push on a system in balance and it pushes back, settling into a new balance.
How problems spread
Chains, tipping points and the ways to contain them.
Broken Windows Theory
Small signs of neglect tell people that nobody cares, and invite bigger trouble.
Chain Reaction
In a tightly linked system, one small trigger can set off everything after it.
Containment
When you cannot solve a problem at once, stop it spreading first.
Critical Mass
Many efforts only take off once they pass a threshold; below it, they fizzle out.
Domino Effect
One small fall can set off a chain that topples far bigger things.
Herd Immunity
Some protections only work when enough of us take part, and then they shield everyone.
Quarantine
Isolate what might be harmful long enough to learn whether it really is.
Tipping Point
Pressure builds quietly for a long time, then change arrives all at once.
When fixes backfire
Solutions that create new problems.
Blowback
Actions meant to control others can return, years later, as the very harm they were meant to prevent.
Cobra Effect
A reward for fixing a problem can end up paying people to create more of it.
Collateral Damage
Every action aimed at one target also lands on people who never chose it.
Hydra Effect
Attack a problem without touching its cause, and it can come back in several new forms.
Jevons Paradox
Making a resource cheaper to use often means we end up using more of it, not less.
Law of Unintended Consequences
Actions in complex systems produce results that nobody planned.
Spillover Effects
The effects of a decision travel far beyond the people who made it.
Zero Tolerance Policy
Rules that allow no exceptions bring clarity, but can punish honest mistakes as harshly as real harm.
Rewards and rules
Incentives, measures and regulations, and how they go wrong.
Cap and Trade
Fix the total you can afford, then let people trade to meet it as cheaply as possible.
Coase Theorem
When bargaining is cheap, people can settle a shared harm with a deal.
Goodhart’s Law
When a measure becomes a target, people learn to hit the number and miss the point.
Government Failure
Government action meant to fix a problem can leave things worse than doing nothing.
Incentives
People do what they are rewarded for, not what they are told or asked.
Market Failure
Free markets go wrong when prices leave out costs that fall on someone else.
Perverse Incentives
A badly designed reward can pay people to produce the opposite of what you wanted.
Regulatory Capture
Watchdogs set up to guard the public can end up serving the industry they watch.
Shirky Principle
Institutions tend to preserve the problem to which they are the solution.