The Failure-Quota Numbers Game
Budget repeated losses as the route to a probable win
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 98%
Smith learned in door-to-door sales to view rejection against a larger run of attempts: if roughly one sale followed one hundred doors, a refusal was one of the expected non-sales rather than a personal verdict. He later applied the same perspective to ten months of emails before a purchase and years of regular social posting before earning money. The method defines a repeatable attempt, sets a provisional trial horizon, and records results so a single loss has the right statistical weight. Confidence comes from being willing to lose within the process while continuing to learn. The quota is not permission to harass people or repeat an ineffective script indefinitely; consent, cost, feedback, and periodic adjustment still determine whether continued attempts are justified.
Origin
Smith says knocking on hundreds of doors for an energy company taught him to treat sales as a numbers game and to expect many refusals before one sale.
Core principles
- 01A failed attempt can be one expected trial rather than a verdict
- 02Repeated attempts reveal a more useful base rate than one result
- 03Losing once is different from abandoning the endeavour
- 04The process must be reviewed rather than repeated blindly
How to run it
- 1
Define one trial
Specify the smallest complete attempt, such as one appropriate prospecting email, application, pitch, or published post. Keep trials comparable enough to learn from them.
Pro tip Track the channel and message version with each attempt.
Watch out Respect consent, platform rules, and requests for no further contact.
- 2
Set a provisional quota
Choose enough attempts to prevent one result from deciding the whole endeavour. Treat the number as a test horizon, not a promise of success.
Pro tip Use available base rates when they are credible; otherwise label the quota as an experiment.
Watch out Do not invent certainty from Smith's personal one-in-one-hundred example.
- 3
Log losses without verdicts
Record every attempt and result. Classify a rejection as one observation unless it reveals a specific flaw or stop signal.
Pro tip Separate no response, explicit rejection, and qualified interest.
- 4
Review the pattern
At scheduled intervals, inspect response quality, feedback, cost, and fit. Change the audience, offer, or method when evidence points to a problem.
Pro tip Review batches rather than emotionally redesigning after every attempt.
Watch out Persistence without learning is repetition, not a strategy.
- 5
Distinguish loss from defeat
After a failed trial, decide deliberately whether to continue, adapt, or stop. Do not let one loss make that decision automatically.
Pro tip State the evidence that would justify ending the experiment.
In the wild
Smith says his door-to-door experience suggested that about one hundred knocks might be needed for a sale. That expectation allowed him to interpret a refusal as one of many required trials rather than a reason to stop immediately.
→ Rejection became part of the measured process instead of a global judgment about his ability.
Smith says he sent emails every day for ten months before someone bought, and posted on social media for roughly four and a half years before it generated money. These are personal timelines, not promised benchmarks.
→ A long period without revenue did not automatically end either effort.
Common mistakes
Treating the estimate as a guarantee
A provisional failure quota structures a test; it cannot promise that the next attempt will succeed.
Repeating without reviewing
Accumulated losses should inform changes to the audience, offer, skill, or decision to stop.
Ignoring consent and cost
No quota justifies unwanted contact, unsafe behaviour, or losses beyond an agreed limit.
Is it for you?
Best for
People doing ethical outreach, publishing, sales, applications, or practice where outcomes vary across many attempts.
Not ideal for
Repeated actions that are harmful, unwanted, prohibitively costly, or unchanged despite strong evidence that the method is failing.
From the transcript
“i knew i'd have to talk to a finite amount of people to get a sale”
“losing is one thing being defeated is something completely different”
From the episode
James Smith: Become Confident In 100 Minutes