The 48-Hour Smoke Test: How to Sell Your Product Before It Exists
Type: media · article
Stage: Stage 3: Pricing Proof
Difficulty: beginner
A step-by-step guide to building a pre-order page that measures real purchase intent — without spending money on ads, writing production code, or misrepresenting what exists.
Overview
A smoke test is the oldest validation technique in the startup playbook: describe the product as if it exists, put a price on it, and see if anyone tries to buy. The modern version takes 48 hours to set up, requires no code, and produces the most direct signal available at Stage 3 — actual purchase intent from actual humans who have read your actual price.
The mechanics of intent
The smoke test setup:
1. Build a one-page site describing your specific solution and price point. Carrd (from Stage 2) takes under two hours. The page needs: a specific headline that names the outcome, a price, a brief description of what the buyer gets, and a button.
2. The button copy matters: 'Pre-order' or 'Join the waitlist for $X' is more honest than 'Buy Now' for a product that doesn't fully exist. 'Reserve your spot' works well for service-adjacent products.
3. Connect the button to a Stripe payment link (not a form — a payment link). The goal is to measure whether people will enter credit card details, not whether they'll submit an email address. Email submission is free. Credit card entry has friction — and that friction is the filter.
4. When users click and reach the payment page, redirect them after entry (or before, with a transparent message) to a 'You're in — we'll reach out before launch' confirmation. Don't charge them unless you're genuinely ready to deliver.
Success benchmarks
For a cold audience (people who don't know you) arriving from organic or targeted posts:
• Under 0.5% of visitors attempting to pre-order: the positioning or price isn't resonating. Revise the headline and retest.
• 1–3% of visitors attempting to pre-order: strong signal for a B2B tool. This is above average conversion for a product that doesn't exist yet.
• Above 3%: exceptional. Either the audience is very warm, the positioning is unusually clear, or both.
For a warm audience (your community, your network, people who know you):
• Multiply the cold benchmarks by 2–3x. A 5–10% attempt rate from a warm audience is good but not exceptional — it reflects relationship trust as much as product-market fit.
• The meaningful signal from a warm audience is the payment completion rate, not the attempt rate. How many people who started the payment flow actually completed it?
Track the full funnel: page visits → button clicks → payment page views → payment completions. Each drop-off point tells you something different about what's failing.
Ethical guardrails
The smoke test works only if it's honest. The ethical version:
• The page makes clear that the product is in development or pre-launch — 'launching in [month]' or 'currently in early access' is accurate framing
• Users who submit payment information are immediately told what they've signed up for — priority access, a founding member discount, an early beta seat — and when they'll hear from you
• You don't actually charge the card unless you're prepared to deliver something meaningful within the timeline you've stated
• You respond personally to everyone who attempts to purchase, even if the test is unsuccessful
The ethical guardrails matter for a practical reason: the people who try to buy are your best early customers. Burning that relationship with a misleading experience is a Stage 3 failure that's harder to recover from than a failed test.
Running the test in the right place
The channel where you run the smoke test determines who you're testing with — and who you're not.
High-signal channels for Stage 3:
• Niche communities where your target customer is active (Discord servers, Reddit threads, Slack groups, LinkedIn groups)
• Direct outreach to people who expressed interest in Stage 1 or Stage 2 interviews
• A post on X or LinkedIn with specific framing ('if you're a [specific customer] who deals with [specific problem], I built something for you')
Low-signal channels:
• Broad social posts without audience targeting
• Paid ads before you have a validated price and positioning
• Your personal network, unless they match your ICP precisely
The channel is part of the test. A 2% conversion rate from a niche community of your target customer is much stronger signal than a 2% conversion rate from a general audience post.