Business
Growth Experiment Generator
A growth experiment generator produces a structured card so growth ideas get tested rigorously rather than shipped on a hunch. Pick the growth lever — Acquisition, Activation, Retention, Referral, or Revenue — describe your idea, and it returns a card with a falsifiable hypothesis, the one primary metric with a baseline and target, a test design, an ICE scoring field, and slots for the result and learning. Growth teams and founders use it to run a disciplined experiment pipeline, prioritise the highest-leverage tests, and capture learnings whether an experiment wins or loses. Most growth ideas fail — the teams that win test cheaply, learn fast, and double down only on what proves out. Score with ICE before running, and record the learning even when the result disappoints.
How to use
- Choose your options above
- Click Generate
- Copy your result
Detailed instructions
- Pick the growth lever and describe the idea.
- Click Generate to produce the experiment card.
- Set the metric, baseline, target, and ICE score.
- Run it, record the result, and capture the learning.
Use Cases
- •Documenting a growth experiment before running it
- •Writing a falsifiable hypothesis tied to one metric
- •Prioritising experiments with an ICE score
- •Building a repeatable growth experiment pipeline
- •Capturing learnings from wins and losses alike
Tips
- →Tie every experiment to a single primary metric.
- →Score with ICE before running to prioritise.
- →Write the hypothesis so it can be proven false.
- →Record the learning even when the test fails.
FAQ
What is an ICE score and how do I use it?
ICE rates an experiment on Impact, Confidence, and Ease, each scored 1–10. Combining them gives a priority number so you run high-impact, high-confidence, easy tests first instead of whatever feels exciting. The generated card includes an ICE field so every experiment is scored consistently.
Why tie a growth experiment to a single metric?
A single primary metric makes the result unambiguous. If a test could move several numbers, you risk cherry-picking the one that looks good afterward. One metric with a baseline and a target keeps the experiment honest and the learning clean.
What if the experiment fails?
A failed experiment that produces a clear learning still wins. Most growth ideas do not work, so recording why something failed sharpens your model of users and feeds better hypotheses for the next round. Capture the learning every time.
Which growth lever should I pick?
Pick the lever where you have the most pressing constraint. If users never complete signup, focus on Activation. If they sign up but leave quickly, focus on Retention. The five AARRR levers each have different metrics, and mixing them muddies the experiment.
How long should a growth experiment run?
Long enough to collect a statistically meaningful sample and cover at least one full business cycle. Stopping early when the numbers look promising inflates false positives. Set your sample size and duration before you start and commit to running the full test.
You might also like
Popular tools from other categories that share themes with this one.
Try these next
More free tools from other corners of the catalog, picked by shared themes.