Uploaded May 2026 | Updated September 2026, 2 weeks ago
A confidence interval is a range around a sample estimate, within which the true population value is likely to fall.
When a business surveys 1,000 customers and finds 60% prefer a new product, that 60% is a sample estimate, not a certainty. The confidence interval expresses the uncertainty around it.
A typical 95% confidence interval might say the true preference is between 57% and 63%. That means: if the survey were repeated many times, 95% of the resulting intervals would contain the true value.
Two things drive the width of a confidence interval. Sample size, bigger samples produce narrower intervals, because the estimate is more reliable. And variability, the more spread out the responses, the wider the interval.
Confidence intervals matter for decisions. A wide interval means the result could swing either way, risky to act on. A narrow interval means the estimate is tight enough to back a decision.
Ignoring intervals leads to overconfidence, treating a single survey number as fact, when it's actually a fuzzy estimate. For any business, confidence intervals are a discipline: they translate a number into a range, and a range into honesty about what we really know.
A confidence interval is a range around a sample estimate, within which the true population value is likely to fall.
When a business surveys 1,000 customers and finds 60% prefer a new product, that 60% is a sample estimate, not a certainty. The confidence interval expresses the uncertainty around it.
A typical 95% confidence interval might say the true preference is between 57% and 63%. That means: if the survey were repeated many times, 95% of the resulting intervals would contain the true value.
Two things drive the width of a confidence interval. Sample size, bigger samples produce narrower intervals, because the estimate is more reliable. And variability, the more spread out the responses, the wider the interval.
Confidence intervals matter for decisions. A wide interval means the result could swing either way, risky to act on. A narrow interval means the estimate is tight enough to back a decision.
Ignoring intervals leads to overconfidence, treating a single survey number as fact, when it's actually a fuzzy estimate. For any business, confidence intervals are a discipline: they translate a number into a range, and a range into honesty about what we really know.
