How to Validate a Business Idea Before You Build

Rahul Vohra spent months manually onboarding every single customer to Superhuman, an email app that costs $30/month when Gmail is free.

During those 90-minute sessions, he'd give a demo, then look users in the eye and explain the pricing. If they hesitated, he'd probe why. If they agreed, he'd ask at what price it would feel expensive but they'd still pay.

The median answer to that last question—expensive, but I'd still buy it—was exactly $30.

That's not luck. That's validation.

Today Superhuman is valued at $825 million. Vohra didn't guess at his pricing or copy competitors. He asked the market directly, one conversation at a time.

Most founders skip this. They assume demand exists, build for months, then discover nobody wants to pay. That's not entrepreneurship. That's expensive guessing.

---

What Validation Actually Is

Validation answers one question: Will people change their behavior or spend money to solve this problem?

Not "do they say they would." Not "does the idea make logical sense." Not "do my friends think it's cool."

Will they actually do something different?

Nick Swinmurn wanted to sell shoes online in 1999. The obvious question: would people buy shoes without trying them on?

He didn't survey people. He didn't build a warehouse. He walked into local shoe stores, photographed their inventory, and listed the photos on a basic website.

When someone ordered, he'd buy the shoes at retail, ship them, and eat the loss. Terrible unit economics. But he proved the only thing that mattered: people would buy shoes online.

That insight became Zappos, which Amazon acquired for $1.2 billion.

The key: Swinmurn didn't validate his solution (an e-commerce platform). He validated the behavior (buying shoes without trying them on). The solution came later.

---

The Sean Ellis Test

Sean Ellis ran early growth at Dropbox, LogMeIn, and Eventbrite. He noticed a pattern: the startups that eventually achieved product-market fit all passed a specific threshold.

He asked users: "How would you feel if you could no longer use this product?"

The magic number: 40% must answer "very disappointed."

Not "somewhat disappointed." Very disappointed.

When Superhuman ran this survey, they found something interesting: users who called the product "very fast" correlated strongly with "very disappointed" responses. Users who didn't mention speed were lukewarm.

This told Vohra exactly what to double down on. Not more features. More speed.

You can start getting directionally correct results with about 40 survey responses—much fewer than most people assume. But you need to filter: only survey users who've actually used the product meaningfully (Superhuman's threshold was using it at least twice in the last two weeks).

---

Five Signals That Predict Success

Not all ideas are equally validatable. Some show clear signals early. Others stay ambiguous no matter how much you test.

Here's what to look for:

1. People are already paying for bad solutions

Ivan Zhao noticed that professionals were cobbling together 5-6 tools—Evernote for notes, Google Docs for collaboration, Trello for tasks. The market was "crowded," but the experience was fragmented.

That fragmentation was the signal. People were willing to pay for multiple bad tools. They'd pay more for one good one.

Notion is now worth over $10 billion.

When nobody is paying for any solution to your problem, that's a red flag. It usually means the pain isn't painful enough.

2. The problem recurs frequently

Daily problems beat weekly problems. Weekly beats monthly.

Scheduling meetings is annoying every single day. That's why Calendly hit $70M ARR with minimal sales effort—the pain is constant and universal among business professionals.

One-time problems (getting a tattoo removed, planning a wedding) can work, but you need massive volume or high margins to compensate for non-recurring revenue.

3. Your target customers can actually pay

College students have lots of problems and no money. Enterprise customers have money and 18-month sales cycles. SMBs can pay and decide quickly.

Who experiences your problem? Can they write a check? How many people need to approve it?

Superhuman's early users were executives, investors, and power users who billed their time at hundreds of dollars per hour. $30/month to save 4 hours a week was a no-brainer.

4. You can reach these customers affordably

Superhuman started with a waitlist. To get access, users had to answer detailed questions about their email habits. This accomplished two things:

1. Filtered for high-intent users who genuinely cared about email
2. Created demand through scarcity

When you can't reach customers, you can't validate. When you can't validate, you can't build with confidence.

B2B often wins here because customers are findable: they have LinkedIn profiles, attend conferences, search Google for solutions. Consumer apps often die because CAC (customer acquisition cost) exceeds LTV (lifetime value).

5. The timing is right

Airbnb pitched the same idea in 2007 and 2008. In 2007, investors laughed. In 2008, during the financial crisis, people suddenly needed extra income. Same founders, same idea, completely different outcome.

What changed recently that makes your solution possible or necessary now?

• New technology? (LLMs making AI products viable)
• Regulatory shift? (GDPR created a market for privacy tools)
• Behavioral change? (Remote work changed how teams communicate)
• Economic conditions? (Recession creates demand for cost-cutting tools)

"The market has always needed this" is a weak answer. If it always needed it, someone would have built it already.

---

Validation Techniques That Actually Work

The Van Westendorp Method (For Pricing)

This is how Superhuman found $30/month. Four questions:

1. At what price would this be so expensive you wouldn't consider buying it?
2. At what price would it be so cheap you'd question the quality?
3. At what price would it feel expensive, but you'd still buy it anyway?
4. At what price would it be a bargain?

For products competing with free alternatives or big incumbents, orient around question 3. That's your sustainable price point where customers feel good about paying.

The Fake Door Test

Build a landing page for a product that doesn't exist. Run ads. Measure conversion.

Joel Gascoigne validated Buffer this way. Page one described the concept. Page two showed pricing. When users clicked a pricing tier, they saw: "We're not quite ready yet. Leave your email."

He collected 120 signups in the first week. More importantly, people went through a full signup flow including seeing the price before discovering it wasn't available.

That's real intent. Not "I might use this." They demonstrated willingness to pay.

The Concierge MVP

Before building automation, do it manually.

Food on the Table matched families with recipes based on weekly grocery store sales. The founders started by personally calling families every week to recommend recipes.

Terrible efficiency. But they learned exactly what customers wanted—and what they'd pay for—before writing any code.

You can always automate later. You can't un-build software nobody wants.

The Pre-Sale

This is the gold standard: getting paid before the product exists.

Pebble raised $10.2 million on Kickstarter before manufacturing a single watch. Tesla takes deposits years before delivering vehicles.

If someone puts money down without being able to use the product immediately, you've validated demand as strongly as possible without building.

---

When to Kill Your Idea

Killing ideas is emotionally brutal. It's also the most valuable skill a founder can develop.

Kill signals:

• 20+ customer interviews with no consistent pain pattern. If everyone describes the problem differently, there's no product that solves "it."

• Nobody is paying for existing solutions. This usually means the problem isn't painful enough. Educating the market is expensive and usually fails.

• Every conversation requires convincing people the problem exists. If you're selling the problem before you can sell the solution, you're in trouble.

• The only excited people are other founders. Founders are not customers. They're optimists who love ideas. Find the people who have the problem daily.

The best founders kill 3-5 ideas before finding one worth building. That's not failure—it's efficient filtering.

---

The Validation Trap

Here's the counterpoint: sometimes validation lies.

Airbnb's early data was terrible. Nobody wanted to rent air mattresses to strangers. The founders ate Obama-themed cereal to survive.

By every reasonable validation metric, they should have quit. They didn't. Today it's worth over $80 billion.

So when should you ignore the data?

The answer is founder-market fit. Brian Chesky understood the rental experience because he'd lived it, deeply and repeatedly. He knew, at a gut level, that people would pay for authentic local experiences.

The data was wrong about timing and scale. It wasn't wrong about the fundamental insight.

If you have years of personal experience with the problem—not book learning, actual lived experience—you've earned the right to push through some negative signals. If you're a tourist in the market, trust the data instead.

The founders who fail are often the ones who can't tell which category they're in.

---

What to Do Next

If you're considering an idea:

1. Write down your core assumptions. What must be true for this to work? Prioritize the assumptions most likely to kill the business if wrong.

2. Design cheap tests. Can you validate the riskiest assumption in 2 weeks with $500 or less? If not, simplify the test.

3. Talk to strangers. Not friends, not family. People who match your target profile and have no reason to be nice to you.

4. Look for behavioral evidence. What people do matters more than what they say. Are they already paying for solutions? Have they tried and failed to solve this?

5. Set kill criteria in advance. Before testing, decide what results would make you walk away. It's easier to be honest before you're emotionally invested in specific data.

The founders who succeed aren't the ones with the best ideas. They're the ones who learn fastest. And learning starts with validation you can trust.

---

Further Reading

• The Startup Validation Checklist — 15 questions to assess any idea before building
• Why ChatGPT Can't Validate Your Startup Idea — The dangers of AI-generated market research
• B2B vs B2C Validation — Different markets require different playbooks
• Founder-Market Fit — Why your background matters more than your idea

---

Ready to compress your market research from weeks to minutes? Try Bedrock Reports free and get evidence-backed validation data for your startup idea.