Experiment Catalog
The library of validation recipes you drop into a sprint — proven tests, so you don't design every experiment from scratch.
What it is
The Experiment Catalog is the library of validation recipes you can drop into a sprint. Each recipe describes a specific test — what to do, what evidence to look for, how to interpret the result — so you don’t have to design every experiment from scratch. It’s the bridge between “I know what I need to learn” and “I have a concrete thing to run this week.”
Where it lives
- Top-level at
/experiment_catalog; also chat-scoped at/chats/:chat_id/experiment_catalog. - Reachable from a sprint via the “Add experiment” button on the Sprint Board.
- Each recipe has a preview page before you commit to running it.
- A recommended next step is highlighted at the top — the single recipe LEANSpark thinks fits your current state best, chosen from your intent, stage, canvas, and progress. It’s a suggestion, not a gate; browse the full catalog whenever you disagree.
How to use it
Know the assumption you’re testing before you browse — otherwise you’ll pick what sounds interesting instead of what’s informative. Reach for the catalog when you’ve identified the assumption and need a proven way to test it, when you’re starting a sprint and want experiments that fit your weakest canvas block, or when an experiment failed and you want a different angle on the same assumption.
The recipe is a starting shape, not gospel: you still define the specific hypothesis, audience, and threshold that make it real for your business. And resist loading a sprint with five recipes — most founders run one or two well, not five poorly. A recipe you commit to flows onto the Sprint Board with a hypothesis, success criteria, and a runbook, ready to execute.