Clinical Proof Before Launch Gate
Require human evidence before turning promising science into a product
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 96%
The gate separates three milestones that founders often collapse: being able to manufacture something, seeing encouraging pre-clinical efficacy, and proving a meaningful benefit in people. Before launch, the team defines the human outcome that must be observed and funds a clinical trial capable of testing it. Revenue does not begin merely because the first two milestones look promising. If the trial clears the outcome gate, the evidence supports launch and the associated claims. If it does not, the money is treated as the cost of learning and the product returns to design rather than being pushed through marketing. Cutcliffe's story shows the mechanism as a reusable decision rule for high-stakes innovation: proof determines commercial readiness, while technical progress alone determines only whether further testing is justified.
Origin
Cutcliffe described refusing an investor's suggestion to launch after manufacturing and pre-clinical success. She required clinical proof first, and the investor backed the company because of that stance.
Core principles
- 01Manufacturing feasibility is not proof of customer benefit
- 02Pre-clinical efficacy is a reason to test, not permission to overclaim
- 03Define the required human outcome before spending on launch
- 04A failed trial should return the product to design rather than marketing
- 05Long-term trust matters more than early revenue
How to run it
- 1
Separate the milestones
List manufacturing feasibility, pre-clinical efficacy, and demonstrated human benefit as distinct gates. Do not let success at one stand in for another.
Pro tip Write down what each body of evidence can and cannot support.
Watch out Promising mechanisms can create false confidence about real-world outcomes.
- 2
Define the human outcome
Specify the measurable benefit required before a commercial launch. Set the criterion before seeing trial results.
Pro tip Choose an outcome that matters to users, not merely one that is convenient to measure.
- 3
Design the evidence test
Run an appropriate clinical trial that can distinguish the product's effect from noise or placebo. Preserve enough rigor that the result can guide a real decision.
Pro tip Use qualified trial and statistical expertise for the study design.
Watch out A weak test can produce a confident-looking but unusable answer.
- 4
Apply the gate
Compare the results with the predefined outcome. Launch only when the evidence supports the benefit and associated claims.
Watch out Do not move the threshold after seeing disappointing results.
- 5
Redesign when proof fails
If the outcome is not achieved, return to product design and decide whether another evidence cycle is justified. Treat the trial spend as learning rather than a reason to launch anyway.
Pro tip Record what the failed gate changed about the product hypothesis.
Watch out Sunk costs are not evidence of customer benefit.
In the wild
After the company learned to manufacture its strain and obtained encouraging pre-clinical evidence, an investor suggested launching immediately to build revenue. Cutcliffe refused to launch until a clinical trial showed that the product actually helped people. She said that if the desired outcome did not appear, the company would return to design instead.
→ The evidence requirement became a signal of founder conviction, and Sequoia led the funding round.
Common mistakes
Treating feasibility as efficacy
A product that can be manufactured reliably has passed an operational test, not a human-benefit test.
Launching to justify sunk cost
Money already spent on development or trials does not lower the evidence threshold. A missed outcome should trigger redesign, not aggressive marketing.
Defining success after results
Choosing the winning endpoint after seeing the data makes the gate movable. Set the meaningful outcome in advance.
Is it for you?
Best for
Founders making consequential health, science, or performance claims where weak evidence can harm users and trust.
Not ideal for
It is excessive for low-risk products whose value can be tested safely through ordinary customer use and rapid iteration.
From the transcript
“until we have clinical trial data and we have proof that these products are actually going to help people we're not going to launch a…”
“if we don't get the outcome that I want we're just that all is all going to go to waste so we're going to go…”
From the episode
Episode 279: Colleen Cutcliffe: How to Heal Your Gut, Control Your Glucose Levels, and Speed Your Metabolism
Colleen Cutcliffe