The core idea

A bug found later costs more because:

• more users are affected
• more systems depend on the faulty behavior
• fixes require more coordination and retesting

Typical cost curve

Bugs are cheapest to fix when caught:

• during requirements/design (clarify before coding)
• during development (unit tests)

They’re most expensive in production:

• hotfixes
• customer support
• refunds
• reputation damage

Diagram: cost increases over time

  graph LR
A[Requirements] --> B[Design] --> C[Implementation] --> D[Testing] --> E[Production]

A -. low cost .-> A
B -. low cost .-> B
C -. medium cost .-> C
D -. higher cost .-> D
E -. highest cost .-> E

Real impact examples

• Checkout bug causes lost revenue
• Incorrect tax calculation causes legal risk
• Data leak causes security incidents

Testing as risk management

Testing doesn’t prove absence of bugs.

Testing gives you:

• evidence the system behaves as expected
• confidence to ship and refactor
• faster debugging