95% of new products fail  ·  88% of transformations fail  ·  92% of startups fail  ·  NYT Games killed 97% of their ideas  ·  70–90% of innovations fail  ·  only 3% of companies are "innovation ready"  ·  fail faster. learn faster.  ·  95% of new products fail  ·  88% of transformations fail  ·  92% of startups fail  ·  NYT Games killed 97% of their ideas  ·  70–90% of innovations fail  ·  only 3% of companies are "innovation ready"  ·  fail faster. learn faster.   

Chapter 01 of 06

FC Resource

FAIL
beautifully.

Most innovation fails. Most startups fail. Most ideas die quietly. And that's not a bug — it's the whole point. This is what happens when you stop pretending failure is optional.

The numbers

95%

of new product launches fail

Christensen, HBS

88%

of business transformations miss their goals

Bain, 2024

92%

of startups eventually fail

Startup Genome

97%

of game concepts NYT killed before finding Wordle

Jonathan Knight, NYT Games

6%

of executives are satisfied with their innovation performance

Viima research

3%

of companies are truly "innovation ready" — down from 20% in 2021

BCG, 2024

The rate hasn't changed. The question is what you do with it.

Why failure is a feature

Four things failure actually does.

01

Failure is the fastest data you'll ever collect.

Market research, focus groups, and strategy decks tell you what people say. Failure tells you what they actually do. A product nobody wanted is information you couldn't have bought any other way. CB Insights analyzed 431 startup post-mortems: "ran out of cash" is almost never the real cause — it's the last symptom. The root? No one wanted what you built.

02

Orgs that embrace failure outperform by 3.6×.

Companies using separate teams for exploration vs. exploitation — and giving teams permission to fail fast — succeed at breakthrough innovation 90% of the time. Companies without that structure? 25%. The differentiator is structural, not cultural: separate teams, pre-set failure criteria, and explicit permission to stop.

03

Avoiding failure is the slowest path to success.

NYT Games tests 120 concepts to ship 3. They kill 97% — not because they're bad at games, but because they have a single filter: 30-day retention. If it fails the filter, it dies. No committee, no second chances. This ruthlessness is how you find Wordle, Connections, and Spelling Bee. Playing it safe means you ship nothing worth playing.

04

The prophets who "failed" changed everything.

In 1970, David Barrett predicted 350 million Christians in Africa by the year 2000. Social scientists thought he was catastrophically wrong. Missionaries feared it was wishful thinking. He was right — and his "reckless" projection rewired how the whole world understood global Christianity. Demography as prophecy. The boldest claims feel like failure right up until they don't.

Hall of useful failures

Things that had to fail first.

The unreached peoples movement

Ralph Winter's 1974 Lausanne talk was dismissed by many as too data-driven, too analytical. It redirected global missions strategy for 50 years. Good frameworks survive rejection.

Missionary radio, then internet, then smartphones

Every wave of Christian media adoption looked like a distraction or a threat — until it didn't. The "wrong" channel is always right three years later.

Wordle

One of ~120 concepts NYT Games tested. 117 others died so Wordle could live. Every win is built on a graveyard of good ideas that were wrong.

Penicillin

A contaminated petri dish Fleming was about to throw away. He looked closer instead of cleaning up. Curiosity about failure saves lives.

WD-40

Named after 39 failed formulas. The 40th worked. The name celebrates the failures required to get there.

Post-it Notes

A failed attempt at super-strong adhesive created a "useless" weak one. It sat in a lab for 5 years before someone realized it was perfect for bookmarks.

The failure failure

The worst kind of failure is innovation theater.

Steve Blank calls it the real crisis: hackathons, design sprints, innovation labs — activities that look like innovation but produce nothing deployable. Organizations perform failure tolerance without actually tolerating failure. The data backs this up: 25 years of consulting reports show project failure rates haven't budged. Maybe because the firms writing the reports are also charging for the fixes.

Hackathons with no path to production

Innovation labs isolated from business units

Pilots that never graduate to scale

Metrics that measure activity, not learning

Real failure: ship it, watch it die, understand why

The FC take

"Faithful" and "fruitful" are not opposites of "failed."

The missions world has a complicated relationship with failure. We celebrate breakthroughs and memorial­ize martyrs, but we rarely name our wrong bets, sunsetted programs, or organizations that died before their time.

A failed program tells you something about the field that a successful one can't — it tells you where people weren't ready, where the model didn't fit, where the Spirit was doing something different than you planned. Naming failure honestly is an act of stewardship.

By 2075, 83% of Christians will be in the Global South. The orgs reaching them are building, testing, and failing right now. Not the ones managing their reputations.

Common questions

But doesn't failure waste resources?
Depends entirely on whether you run a post-mortem within 48 hours. A $10K pilot that dies in 90 days and teaches you your core assumption was wrong is one of the best investments you can make. A $10K pilot that you keep funding because nobody wants to call it — that's waste. The cost of failure is almost always less than the cost of zombie programs.
What about accountability to donors and partners?
Counterintuitively, naming failure builds more trust than hiding it. Kingdom Impact research shows 89% of faith-driven investors say impact measurement is "very important" — but donors who feel like they're getting a curated story eventually stop trusting you entirely. The orgs that thrive long-term are the ones that tell donors: "this didn't work, here's what we learned, here's what we're doing differently."
How do you fail without demoralizing your team?
Separate the project failure from the person's worth. NYT Games doesn't fire game designers when concepts die — they celebrate the learning and move to the next one. The signal a leader sends when a project fails matters more than the failure itself. "You failed" kills teams. "We learned something the market couldn't have told us any other way" builds them.
What's the difference between productive failure and just being bad?
Pre-committed criteria. Before you launch, write down exactly what success looks like — and exactly what failure looks like. NYT Games uses 30-day retention. You need your version of that. If you can't name what would make you stop, you're not running an experiment — you're running a program you're already married to. That's not a test. That's confirmation bias with a launch date.
What does it look like to actually build a culture that fails well?
It starts before launch, not after. NYT Games celebrates the death of bad ideas — killing a game concept is treated as a win, not a loss. The leader signals everything: if killing a project looks like career suicide, nobody will call anything dead. The practical moves: name kill criteria before funding starts, run a post-mortem within 48 hours of every stop decision, and separate the project's failure from the person's judgment. The full framework is in Chapter 03.
"When doing innovation, the first question is not 'Is this going to work?' but 'If it works, would it matter?'"

— Eric Toone, former ARPA-E Director, on what separates innovation funding from execution funding