---
title: What is Statistical power?
url: https://markin.ai/glossary/statistical-power
category: Experimentation and causality
---

# Statistical power

> Statistical power is the probability that a test detects an effect of a given size when that effect is real. It is fixed before launch by sample size, baseline variance and the minimum detectable effect, and an underpowered test mainly produces expensive ambiguity.

## How it is calculated

```
Power = 1 - P(false negative). Convention: 80% at a 5% significance level.
```

Decide the minimum detectable effect first. Choosing it after seeing the data invalidates the test.

## Why it matters for ARPU

Most enterprise growth tests are underpowered, so real effects get shelved as inconclusive and the organisation slowly stops believing in evidence.

## Related terms

- [A/B test](https://markin.ai/glossary/ab-test), An A/B test randomly assigns customers to two or more variants and compares a pre-declared metric between them.
- [Sequential testing](https://markin.ai/glossary/sequential-testing), Sequential testing allows results to be monitored continuously and stopped early without inflating false positives, using always-valid confidence intervals instead of a single fixed-horizon check.
- [Experiment](https://markin.ai/glossary/experiment), An experiment is a controlled release of a candidate action against a randomised holdout, sized in advance so the result can distinguish a real effect from noise.
- [Guardrail metric](https://markin.ai/glossary/guardrail-metric), A guardrail metric is a measure an experiment must not damage even if the primary metric improves: unsubscribe rate, complaint volume, margin, support contacts, app uninstalls.

## Go deeper

- [Experimentation vs continuous decisioning](https://markin.ai/compare/experimentation-vs-continuous-decisioning), Why throughput needs power discipline.
- [Markin product](https://markin.ai/product), Sizing before launch.

Source: https://markin.ai/glossary/statistical-power