# Growth Experiment

> A small test of one growth idea, with a guess written down first.

**Field:** Marketing & SEO  
**Category:** Growth & Acquisition  
**Also known as:** Growth Test, Growth Experiments

## Think of it like

It is like trying a new way to walk to school. You time it for a week, then keep the faster road.

## Definition

A growth experiment is a small, planned test of one idea for getting more users or sales. It starts with a written guess: if we change this, that number will go up by this much. The team runs the test for a set time, often as an A/B test. Then it keeps, drops or changes the idea. Teams keep a list of ideas that are waiting to be tested.

## Example

A meal box firm guesses that showing the price per meal will lift sign-ups by 5%. It tests it for two weeks on half its visitors, and sign-ups rise 3%.

## Why it matters

Most growth ideas do not work. Testing them small and fast finds the few that do, before money is spent on them.

## Common mistake

Running tests with no guess written down. Then any result looks like a win, and nothing is learned.

## Three principles

- **Write the guess.** Say what will change and by how much.
- **One change.** Test one idea at a time.
- **Keep a log.** Write down every result, wins and losses.

## Related terms

- [ICE Scoring](https://learn.benibix.com/term/marketing/ice-scoring)
- [Growth Hacking](https://learn.benibix.com/term/marketing/growth-hacking)
- [A/B Test](https://learn.benibix.com/term/marketing/ab-test)
- [Pirate Metrics (AARRR)](https://learn.benibix.com/term/marketing/pirate-metrics)
- [Minimum Detectable Effect (MDE)](https://learn.benibix.com/term/general/minimum-detectable-effect)

![A loop of four steps. Ideas go into a list. Each idea gets an ICE score and the top one is picked. It runs as a test with a written guess. The result goes into a log, and what was learned feeds new ideas back into the list.](https://learn.benibix.com/diagrams/mkt-grow-experiment.svg)

---

Source: https://learn.benibix.com/term/marketing/growth-experiment
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-20
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
