Binary Search

An algorithm that finds a target in a sorted dataset by repeatedly halving the search space—checking the middle element and eliminating the irrelevant half. Binary search finds any item among a million possibilities in just 20 steps (log₂ of 1,000,000 ≈ 20). As a thinking tool: when diagnosing any problem with many possible causes, don't test them one by one—divide the possibility space in half with each test. 'Is the problem in the frontend or backend?' eliminates half the codebase instantly. Binary search is the most efficient diagnostic strategy when possibilities can be ordered.

When to use it

When diagnosing problems with many possible causes; when debugging any system and need to isolate the failure point efficiently; when the search space is large and testing every possibility is impractical; when structured problem-solving is needed instead of random hypothesis testing.

How it can help

When troubleshooting any problem, structure your diagnostic questions to eliminate half the remaining possibilities with each answer. Instead of testing hypotheses sequentially (which takes N steps), divide and conquer (which takes log₂N steps). In business diagnostics: 'Is the revenue problem on the acquisition or retention side?' halves the problem space. Then: 'Is the acquisition problem in lead generation or conversion?' halves it again. Three well-chosen binary questions can narrow 8 possible causes to 1. This applies to debugging code, diagnosing process failures, and isolating root causes.