Experiment Card

What it is

One belief, one test, one bar. Your mentee writes a belief, a test and a bar for success, then the result.

When to send it

Send it when a belief needs proof. It fits once you both know which belief to test first.

• After an assumption mapwhen the riskiest one is known
• Before a big spenda hire, a build, an ad budget
• When opinions clashlet a small test decide

Filling it in

Two parts, before and after. Fill in the first part now and the second after the test.

1. Write the belief and the testSay who will do what, and why.
2. Set the bar and a dateDecide what counts as success before you see the result.
3. Come back with the resultAdd the number, choose a decision, then send.

Reading the answers

Compare the result with the bar. The bar, the number and the decision sit side by side.

• Success barset before the test
• Resultthe number they got
• Decisionkeep going, change course or drop it

Questions to ask

Ask about the belief, then the bar. People who ask follow-up questions are liked more (Huang and others, 2017).

• Is this the belief that would hurt most if it were wrong?then: what makes it the riskiest?
• Could a cheaper or faster test answer it?then: what would that test be?
• What will you do if the result sits right on the bar?then: can you decide that now?

Why it works

In randomised trials with Italian startups, founders taught to test hypotheses this way made better decisions about which ideas to drop.

Sources

• Camuffo, Cordova, Gambardella & Spina (2020). A scientific approach to entrepreneurial decision making: Evidence from a randomized control trial. Management Science 66(2):564–586.
• Camuffo, Gambardella, Messinese, Novelli, Paolucci & Spina (2024). Large-scale replication and extension. Strategic Management Journal 45(6):1209–1237.

How a tool works on Mentornity

1. 1The mentor picks a tool from the library and sends it with a short message.
2. 2The mentee fills it in step by step; the answers are saved as they go.
3. 3Both see the answers, comment on any part and agree the next steps.
4. 4Only the two of them see the answers, not the program admins.

Guides for both sides

• Experiment Card for MentorsFor mentors and coaches: when to send the Experiment Card, how to read the result against the bar, which questions to ask and how to agree the next test.
• Experiment Card for MenteesFor mentees and founders: what the Experiment Card is, how to fill it in before and after the test, where to find it and what your mentor does with it.

More tools: Venture

• Assumption Map20 min

  List what must be true for your idea to work. Rate each assumption, then plan a test for the riskiest one.

• Pre-mortem15 min

  Imagine your plan has failed a year from now and list why. Rate each reason, then plan for the worst three.

• Session Brief10 min

  Before the session, write where you are and the one question you want answered. The time goes to that question.

• Customer Interviews20 min

  Plan your questions, then log each interview. The summary shows how many people really have the problem.

• Pitch Rehearsal20 min

  Plan your slides with a time for each, then rehearse against the clock. Each run is kept so you can see the change.

• Unit Economics15 min

  Enter four numbers and see what one customer is worth, and how long it takes to earn back what you spent to win them.

• Cash Runway10 min

  Enter your cash, costs and growth. See how many months the money lasts and whether revenue catches up in time.

• Monthly Update10 min

  Once a month: last month’s goals, one key number, a win, a difficulty, and what you need. Sent every month, it shows the trend.

All tools →