Sell Before You Build: 10,000 KRW Validates 100 People's Intent (Fake Door Test)

Sell before you build. 10,000 KRW confirms 100 people's intent.

3 vs 200, what made the difference

One developer released 5 apps over a year. The tech was flawless and the UI was clean. But in the first month, total user count: 3 users. Three.

Meanwhile, a team that put up a 30-minute landing page with a 10,000 KRW Facebook ad at the planning stage? 200 pre-registrations in a week. The first team spent 3 months building. The second spent 3 hours validating. What made the difference is the Fake Door Test.

Sell before you build, the Fake Door mindset

Fake Door Test sounds harsh, but the idea is simple. You try to "sell" the idea to people before it's real. Not selling a fictional product — measuring interest in a feature that doesn't yet exist, or a service that isn't built yet.

Buffer started like this. Before the company even existed, they threw up a landing page. The concept of "scheduled posting for social media" was still unfamiliar. They made a simple CTA button: "Want this feature?" Clicking it said "In development, leave your email." Result? Thousands left their email. Only then did Buffer confidently start building, and got their first customer 7 weeks in.

Dropbox went further. Even spending tens of thousands of won on Google Ads, customer acquisition cost was $399. Too expensive. They made a video instead, showing how the product works. Posted to YouTube. 70,000 signed up for the waitlist overnight. Free, but the validation was rigorous.

Many think "Fake Door = lying." It isn't. Fake Door must be transparent. You must state "still in development, are you interested?" and if you decide not to build, you owe them an honest message.

Spin up a landing page in 30 minutes

The goal isn't a pretty page — it's fast validation. So tool choice matters.

Build React-based with v0

v0 is Vercel's AI tool. Give it a prompt, it stacks React components. Like this.

My service to sell:
- Marketing automation tool
- Target: solo developers
- Price: 9,900 KRW/month

Build a landing page that pitches this.

v0 builds a high-quality page in 5 minutes. Deploy is a few seconds on Vercel.

Design-first with Framer

Framer is one of the hottest no-code tools. Difference from v0: visual editing. Without seeing code, you can build sophisticated pages just by dragging and dropping. AI features were recently added.

Which tool?

Conclusion: Framer if you're non-technical, v0 if you expect to edit code. A 2024 no-code community survey found 73% chose Framer or v0 for their first landing page.

Validating with ads, in the Korean environment

You've built the landing. Now you need to confirm how many people "actually want it." That tool is ads. 10,000 KRW is enough.

Start with Naver and Kakao

In Korea, Naver and Kakao retarget fastest. You can start with a 10,000 KRW budget on Naver Ads. Naver CPC averages 1,200–2,000 KRW. With 10,000 KRW you can expect at least 5 clicks.

The key is target settings. If your service is "for marketers," narrow it down to age 25–40, occupation Marketing/PM, interest Marketing Tools. Spray too wide and you'll only collect noise.

Expand with Meta/Facebook

If Korean ads aren't enough, throw it at Meta too. Meta's strength is retargeting and LLA (Lookalike Audience). If you have 10 existing customers, Meta finds 10,000 people similar to them.

With a tiny budget, the first few days of data are messy. Run at least 3 days for trustworthy patterns.

Reading the data: signal vs noise

Weak signal: CTR (click-through rate)

CTR is (clicks ÷ impressions) × 100. Normal landing pages see 2–5%. CTR above 5% is signal. People aren't clicking by mistake — they clicked because they wanted to know.

Strong signal: Pre-registration

Leaving an email, entering a phone number, hitting "notify me." That's "real interest." If 10+ of 100 click-throughs leave their info, that's a good signal.

Strongest signal: Pre-payment

The strongest signal is money. "I'll pay 10,000 KRW/month when this ships." 3% pre-pay rate is confidence.

Korean crowdfunding platforms like Wadiz and Tumblbug see that 30%+ of pre-registrants actually pay in many projects.

Buffer's signups-to-customer conversion was ~15%. Dropbox's 70k waitlist became the early base — the strongest market validation possible.

10,000 KRW validation checklist

Landing page prep

• One-line core message: Can a 10-second visitor understand "who this is for, what problem it solves"?
• CTA button: Is one of "Pre-register," "Notify me," "Join waitlist" the most visible thing on the page?
• Email form: Max 3 fields (name, email, one extra question).
• Device test: Is the CTA easily visible on mobile?

Ad setup

1. Naver Ads (recommended): 10,000 KRW, narrow targeting, 3 days–1 week
2. Kakao Ads (secondary): slightly cheaper CPC than Naver
3. Meta Ads (expansion): start with Korean target, then LLA

Interpretation thresholds

Wrapping up

"10,000 KRW validation" gives you conviction. If 100 people showed interest, probably 1,000 will want it. Now you're really ready to build. But a question remains. "So what should I build it with?"

Next post: the 2026 AI build tools map. v0, Framer, Bolt, Lovable, Cursor — a decision tree for picking what fits your situation.

References

1. Fake Door MVP: How We Launched Buffer — Buffer case
2. How We Used Fake Door Testing to Validate Demand — Dropbox case (Chameleon)
3. Framer 2024 Year in Review
4. Facebook ad costs — Shopify Korea (2026)
5. 2024 Facebook ad benchmarks — 10Million

Previous: One Deep Interview Beats 100 Surveys (Mom Test)
Next: The 2026 AI Build Tools Map (Which of 160 Tools to Pick)

Minchul Kim, CEO of Freeive, Fail School