Failure-to-Learning Innovation Loop
Turn bounded failures into learning that improves the next experiment
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 92%
The loop treats failure as a potential information-producing stage of innovation rather than as the opposite of progress. Select an uncertain but important assumption, constrain the possible downside, and run a real test. When the attempt fails, separate criticism of the idea or execution from condemnation of the person. Identify what the result revealed, including hidden dependencies and capabilities that need strengthening. Change the next attempt based on that evidence rather than simply repeating the failure or retreating to familiar work. Over time, uncomfortable experiences build adaptability; the speaking example in the episode shows how a presentation malfunction improved the host's ability to perform without slides. The mechanism converts bounded losses into better judgment and stronger execution.
Origin
Extracted from Marketing Against The Grain, where the hosts connect public mistakes, marketing experimentation, and a failed conference presentation to the development of better judgment and adaptability.
Core principles
- 01Progress requires attempts that may fail
- 02A loss can produce information even when it produces no immediate win
- 03Safe repetition gradually erodes competitive advantage
- 04Critique ideas without vilifying the people testing them
- 05Capability grows when failures are examined and reused
How to run it
- 1
Select an uncertain assumption
Identify a belief that matters to progress but has not been validated. Frame it so a practical experiment can produce useful evidence.
Pro tip Test one central uncertainty at a time.
Watch out Do not label routine execution as experimentation when nothing meaningful is being tested.
- 2
Bound the downside
Set limits on budget, duration, audience exposure, and operational risk before acting. Preserve room for genuine failure without risking irreversible damage.
Pro tip Write explicit stop conditions before launching.
Watch out Celebrating failure does not justify careless or unethical behavior.
- 3
Run the real test
Execute the experiment in conditions capable of disproving the assumption. Accept that visible attempts may create discomfort or criticism.
Pro tip Decide in advance what observations will count as evidence.
Watch out An experiment designed only to confirm existing beliefs produces little learning.
- 4
Separate person from result
Examine the failed idea, preparation, process, or execution without treating the person as a failure. Encourage constructive debate about what was wrong.
Pro tip Describe specific decisions and evidence rather than motives or character.
Watch out Personal attacks suppress future experimentation and honest reporting.
- 5
Extract the learning
Record what the outcome revealed and what should change. Include unexpected skills, dependencies, or contingency plans exposed by the failure.
Pro tip Ask what the team can now do or understand that it could not before.
Watch out Failure without reflection is only loss.
- 6
Modify the next attempt
Apply the lesson to a new test, operating procedure, or capability. Continue the cycle until evidence supports scaling, changing direction, or stopping.
Pro tip Make the procedural change before attention moves elsewhere.
Watch out Do not repeatedly celebrate the same preventable mistake.
In the wild
A speaker's slides repeatedly race to the end during a large conference presentation. Rather than abandoning public speaking, the speaker learns to deliver material more spontaneously and becomes less disrupted when presentation technology fails.
→ An embarrassing failure develops stronger ad hoc speaking ability and resilience.
A marketing team gives a new channel a capped two-week budget and defines the customer-response signals it needs to see. The campaign misses its acquisition target, but interviews reveal that the offer language—not the channel—caused confusion. The team revises the message before its next test.
→ The failed campaign produces a specific correction instead of prompting a retreat to old tactics.
Common mistakes
Avoiding all visible failure
Refusing experiments because they may look unsuccessful preserves comfort but prevents meaningful progress and learning.
Treating failure as automatically valuable
A loss creates value only when its evidence is examined and used to change future behavior.
Vilifying the experimenter
Personal condemnation discourages people from testing uncertain ideas or admitting what went wrong.
Is it for you?
Best for
It is best for teams exploring new marketing channels, products, messages, or operating methods under uncertainty.
Not ideal for
It is not ideal for reckless experiments with unbounded legal, financial, safety, or reputational consequences.
From the transcript
“But the opportunity in losing is where you find the most learnings and the most value.”
“Pushing the envelope means you'll fall on your face at some point.”
“But I got way better as a speaker after that experience because I could actually do things more ad hoc.”
From the episode
How Web 3 Will Impact Your Marketing