Startup Failure Rates by Stage: What the Data Actually Shows

The oft-quoted '90% of startups fail' is misleading. Failure rates vary dramatically by how you define 'startup' and 'failure,' and they differ significantly across funding stages.

More precise data tells a more useful story. Pre-seed companies face different odds than Series B companies. Self-funded businesses fail differently than venture-backed ones. Understanding the real numbers helps you calibrate expectations and identify where to focus your energy.

Let's look at what the data actually shows, where the numbers come from, and what they mean for founders at different stages.

Pre-seed and seed stage:

• Approximately 60-70% of companies that raise pre-seed/seed funding fail to raise a Series A
• Of those that do raise A, many took 2-3 years to get there
• Failure here often means running out of money without achieving product-market fit

Series A:

• About 35-40% of Series A companies fail to raise Series B
• Failure rate decreases because Series A filtering selected for stronger companies
• Failures at this stage often involve scaling problems rather than product problems

Series B and beyond:

• Failure rates drop to 20-30%
• But 'failure' increasingly means modest exits or acqui-hires rather than shutdowns
• Companies at this stage rarely completely disappear

Overall through exit:

• Only about 1% of venture-backed startups reach unicorn status ($1B+ valuation)
• About 10-15% achieve meaningful exits (acquisition or IPO at good valuations)
• The majority return little or nothing to investors

Small business failure rates:

• About 20% of small businesses fail in year one
• About 50% fail by year five
• About 65% fail by year ten
• These are general small business statistics, not tech startups

The comparison is misleading because:

• Venture-backed startups attempt much riskier things
• Different definitions of 'failure' (lifestyle business earning $200K/year isn't failed)
• Selection effects differ (VCs fund ambitious ideas; many small businesses pursue proven models)

What this means:

• Venture startups fail faster when they fail (burn through capital quickly)
• Non-venture businesses fail slower but still frequently
• The risk profiles are different, not necessarily higher or lower overall

Pre-seed/seed failures (top causes):

1. No product-market fit (42% of failures cite this)
2. Ran out of cash (29%)
3. Team problems (23%)
4. Got outcompeted (19%)
5. Pricing/cost issues (18%)

Series A failures (top causes):

1. Scaling prematurely (spending before PMF was real)
2. Unit economics that don't work at scale
3. Key person departures
4. Market timing (too early or too late)
5. Competition from better-funded players

Series B+ failures (top causes):

1. Market shifts (new technology or competitors)
2. Inability to expand beyond initial market
3. Margin compression as competition increases
4. Executive team problems
5. Failed international expansion

The failure modes change as companies mature. Early failures are about finding fit. Later failures are about execution at scale.

Research on startup outcomes reveals some predictive factors:

Factors that correlate with survival:

• Previous founder experience (but first-time founders still succeed)
• Technical co-founder on the team
• Raised from tier-1 investors (selection effect, not causation)
• Domain expertise in the problem space
• Early revenue growth velocity

Factors that don't predict as well as expected:

• Founder pedigree (elite schools, big tech companies)
• Total funding raised (more money doesn't ensure success)
• Initial team size
• Speed to market

The uncomfortable truth:

• Luck plays a larger role than most analysis acknowledges
• Market timing matters enormously and is largely unpredictable
• Many successful companies nearly failed multiple times

Correlation isn't causation. Successful companies raised from good VCs because VCs chose winners, not because VCs made them winners.

Time to failure data:

• Median time to failure for funded startups: 20 months
• 75% of failures occur within 3 years of founding
• Companies that fail typically know they're struggling 6-12 months before shutdown

Runway at failure:

• Most startups that fail run out of money
• Average runway at time of shutdown decision: 2-3 months
• Many founders wait too long hoping for turnaround

The slow death pattern:

• Revenue plateaus or declines
• Fundraising attempts fail
• Key employees leave
• Founders try pivots that don't work
• Eventually cash runs out

Recognizing failure patterns earlier can help founders make better decisions about whether to continue, pivot, or shut down gracefully.

Calibrate expectations realistically:

• Building a venture-scale company is genuinely hard
• Most funded startups don't return investor capital
• Success stories are survivorship bias in action

But don't be paralyzed by statistics:

• Someone has to be in the 10-15% that succeeds
• Your specific situation matters more than averages
• Data describes populations, not individuals

Use the data productively:

• Focus on the failure causes relevant to your stage
• Don't prematurely scale (the biggest Series A killer)
• Extend runway before you desperately need to
• Pay attention to early warning signs

The goal isn't to avoid all risk. It's to take informed risks while managing the factors you can control.

At pre-seed/seed:

• Obsess over product-market fit above everything else
• Keep burn low until you have evidence of fit
• Don't hire aggressively before fit is clear
• Be willing to pivot quickly if needed

At Series A:

• Validate that your unit economics work before scaling
• Hire carefully; bad hires are expensive
• Build systems that scale, not just teams
• Maintain fundraising relationships continuously

At Series B+:

• Expand thoughtfully, not just quickly
• Watch for market shifts that could obsolete you
• Build a strong executive team
• Consider profitability path, not just growth

Each stage has different risks. The strategies that work at seed can kill you at Series B, and vice versa.

Does the '90% of startups fail' statistic include all businesses? It varies by source. Some statistics include all new businesses. Others focus on tech startups. Venture-backed startup statistics typically show 60-75% failure to return capital, which is still high but not 90%.

Do these failure rates account for acqui-hires? It depends on the study. Some count acqui-hires as failures (since they rarely return investor capital). Others count them as exits. The distinction matters for interpreting statistics.

Are certain industries more likely to fail? Yes. Hardware and biotech have higher failure rates due to capital intensity and longer timelines. Consumer apps have high failure rates due to competition. B2B SaaS has somewhat lower failure rates once initial traction is achieved.

Does being in a startup hub affect failure rates? Somewhat. Silicon Valley startups have higher funding success rates but similar failure rates once funded. The main difference is access to capital and talent, not fundamental survival rates.

What's the difference between failure and 'not a venture outcome'? A company might be profitable but too small to interest VCs. This is technically a 'failure' to return venture capital but might be a great lifestyle business. How you define failure matters for interpreting statistics.