FIRETEST is a business idea validation protocol used by individuals and companies to test ideas, features, and initiatives before building. It validates demand using real-world behavior: engagement, clicks, and paid commitment.
FIRETEST is a 7-day protocol that produces a binary decision — proceed, iterate, or kill — based on observed behavior and paid commitment, not opinion.
Most ideas fail before they are built. Not because the team lacked talent. Not because the timing was wrong. Because no one validated demand before capital, time, and effort were committed.
Teams build before they validate. Investors fund before they ask. Founders commit before they confirm. The result is identical every time — months of capital, energy, and conviction spent on something the market never wanted.
FIRETEST exists to make that failure mode unavailable. It answers one question — should this idea exist? — before resources are committed. It replaces opinion with evidence, internal debate with external behavior, and instinct with a single source of truth: the recorded action of a real person in the real world.
When to run a FIRETEST
Before launching a business, hiring a team, or allocating capital to an unproven idea.
When internal alignment exists but external proof does not.
When choosing between multiple ideas competing for the same resources.
The deliverable
At the end of seven days, the operator holds a decision backed by evidence — not a feeling, a roadmap, or a plan. The possible outcomes are: proceed, iterate, or kill.
If no one is willing to act, the idea is not real yet.
The One Rule
Nothing gets built during the test. No launch-scale operation. No site beyond one page. No team. The week's only deliverable is a decision backed by evidence.
Part I · 2 of 3
Principles
Every rule in this standard is derived from these seven beliefs. If any one is loosened, the protocol stops working.
Behavior over opinion.What people do is data. What they say is noise. The protocol records only behavior.
Money over applause.Commitment — payment, signed intent, time on a calendar — is the only valid signal of demand. Likes and waitlists do not count.
Speed over polish.Seven days is sufficient to know. A longer test is avoidance disguised as rigor.
Manual over automated.The first five customers are served by hand. Hiring, systems, and automation come after demand is proven — never before.
Evidence over instinct.Decisions follow numbers. Numbers ignore feelings. The scorecard exists to remove the operator from the verdict.
One variable at a time.Each iteration changes exactly one thing — offer, message, or problem. Multi-variable changes destroy signal.
Kill early, kill cleanly.An idea retired in seven days is six months freed for the next one. Killing well is a skill, not a failure.
FIRETEST assumes the operator is willing to accept a verdict they did not want. The standard only produces value when the participant is bound to its result. A test that the operator overrides at the moment of inconvenience produces nothing.
Part I · 3 of 3
Scope & Limits
Validate before you build. Behavior over opinion. Money over applause.
Conventional Path
Get an idea
Build it for six months
Launch
Find out no one wants it
Start over
FIRETEST Path
Get an idea
Run a seven-day test
Receive a decision
Build only if validated
Deliver by hand first
The Bakery Example
You can spend $200,000 renting a space, buying ovens, and hiring staff — then open the doors and hope. Or you can bake twenty cupcakes this weekend, take them to the farmer's market, and see if people buy them. Both teach you something. One is much cheaper than the other.
Resources required
FIRETEST is designed to be runnable by a single operator without external support, capital, or specialized skills. Both paths produce a valid result.
The Free Path · $0
Distribution: Facebook groups, Reddit, LinkedIn, Nextdoor, personal contacts.
Tooling: Carrd / Google Sites / Notion. Stripe or Venmo for payment.
Trade-off: slower reach. Expect 50–150 viewers. If the idea is real, this is enough.
The Small-Budget Path · ~$150
Distribution: $7/day × 3 days × 3 ad variants on Meta or LinkedIn (~$63), with reserves for a second round.
Tooling: identical to the free path.
Trade-off: faster reach. Expect 300–600 viewers within 48 hours.
What FIRETEST is not
Not a substitute for legal, regulatory, or compliance review. FIRETEST validates demand — not legality, safety, or licensing requirements.
Not market research at scale. It produces a decision about one idea, not a segment analysis or TAM study.
Not a full launch. No product, infrastructure, or team is permitted during the test.
Not a guarantee of success. A passing scorecard means the idea is worth pursuing — not that it will scale.
Not optional once started. The operator must accept the scorecard verdict. Overriding the result voids the test.
What FIRETEST replaces
“We think users want this.”
Internal feature debates without external proof.
Long roadmap speculation cycles.
Overbuilt launches and premature investment.
Capital Framing
The money spent during a FIRETEST is not a cost. It is tuition. $150 to discover whether an idea has demand is among the cheapest forms of market research available. The expensive mistake is spending six months and significant capital building something no one will pay for.
Part II · The Protocol
Part II · Overview
Protocol Overview
Six steps. Seven days. One decision backed by evidence.
The protocol moves an idea through six steps in seven days, ending in a binary decision: proceed, iterate, or kill. The work is done by hand. Between Step 5 and Step 6, the operator runs the Day 7 Scorecard — three checks (Distribution, Resonance, Commitment) and a clear verdict.
7
Days to a Decision
$0–$150
Total Cost
3
Scorecard Checks
Figure 1 · The Seven-Day Flow
A single idea moves through six steps in seven days. The scorecard sits between Step 5 and Step 6.
The six steps
Write — describe the idea in three sentences anyone can understand.
Show — expose it to real users through the cheapest distribution available.
Build — one simple page. No full offer yet.
Ask — request commitment: money, time, or signed intent.
Count — record three numbers (Distribution, Resonance, Commitment).
Decide — score them on the Day 7 Scorecard and act on the verdict.
Part II · Step 1 of 6
Write Your Idea Down
Day 1 · ~1 hour · Output: three sentences
Before exposure, the idea must be expressible in plain language. If you cannot explain it in three sentences, it is not yet clear enough — and the audience will not understand it either.
The three-sentence structure
The problem. What is painful, annoying, or expensive about the customer's life today?
The result. What outcome do they want instead?
Your method. How will you deliver that outcome?
The Dog-Walking Service Example
Problem: You work long hours and your dog is home alone all day.
Result: Your dog gets a one-hour walk every day while you're at work.
Method: A trusted neighborhood walker who sends a photo and a note after every walk.
The Meals for New Mothers Example
Problem: You just had a baby and haven't had a real meal in three weeks.
Result: Warm, home-cooked meals delivered daily for two weeks after the birth.
Method: A local cook who prepares everything fresh each morning — no frozen, no processed.
The Math Tutoring Example
Problem: Your 10-year-old cries at the kitchen table every time you try to help with homework.
Result: Your child gets homework done without tears — and starts to like math.
Method: A patient tutor who meets your kid online twice a week, using games instead of lectures.
Validation
Read the three sentences aloud to someone outside your field. If they cannot repeat the idea back to you in their own words, the sentences are too complicated. Rewrite. Repeat until a twelve-year-old could explain it.
Part II · Step 2 of 6
Show It to Real People
Days 2–3 · Output: real audience exposure
You now find out whether the problem you described is actually real — by putting your message in front of the people you intend to serve, and observing whether they react.
Free distribution channels
Facebook groups. Three groups where your ideal customer already gathers. Read what they complain about. Then post your idea as a question.
Reddit. Two or three subreddits matching your customer. Read the rules. Post something honest: “I'm thinking about offering X because of Y. Would this help anyone here?”
Your neighborhood. For local ideas, flyers and Nextdoor work.
Your phone contacts. Send 20 texts: “I'm working on something new. Could I show you for 2 minutes and get your honest opinion?”
LinkedIn. For professional customers, one short post describing the problem.
Paid distribution channels
With ~$150 of budget, you skip the hunt and let the right audience come to you. Write three short variations of your three sentences. Each variation targets the same problem from a different angle. Run each as a Meta or LinkedIn ad at roughly $7/day for 3 days.
The Three Variations Example
For the meals-for-new-mothers concept:
1. You just had a baby. You shouldn't also have to cook.
2. Warm homemade meals, delivered. For the most exhausted two weeks of your life.
3. The gift new parents actually need — real food, no cooking.
The same idea framed three ways. The market reveals which framing resonates.
Signals of working distribution
People click on the post or ad.
People comment, ask questions, or share.
People reply “yes, I need this” publicly or in DMs.
Caution
Comments and enthusiasm are weak signals. Anyone can say they love an idea. Only real money tells you whether they mean it. Money comes in Step 4.
Part II · Step 3 of 6
Build One Simple Page
Day 3 · ~2 hours · Output: one landing page, one action
When someone is interested in your idea, you need a place to send them. The page has one job: convert interest into action.
Acceptable tooling
Carrd — single-page builder, free.
Google Sites — free, drag-and-drop.
Notion — any page can be published as a website.
Required elements
A headline. Use the exact words from your best-performing post or ad. Familiarity converts; cleverness does not.
A short explanation. Two plain sentences describing what the customer receives.
Proof. A photo, a 60-second phone video, a sample, or a mockup. It must feel real — not polished.
One button. One action. Not three. Not five. One.
The Landing Page Example
Headline: You just had a baby. You shouldn't also have to cook.
Explanation: Warm, fresh, homemade meals delivered to your door every day for two weeks after your baby arrives. Made that morning. Real food, no freezer.
Proof: Three photos of home-cooked meals in nice containers. A 60-second phone video introducing the cook.
Button: Reserve your two weeks for $50 — refundable if you change your mind.
Forbidden elements
Anything about you or your story — not yet.
Feature lists. Customers buy outcomes, not features.
A navigation menu. The page has one screen.
More than one call to action.
Validation
Show the page to a stranger for 30 seconds. Ask: “What is this, and what do I do next?” If they cannot answer, the page is too complicated. Simplify.
Part II · Step 4 of 6
Ask for Real Money
Days 3–7 · The decisive step · Output: paid commitments
This is the step most people skip. Not because it is hard in practice — it takes five minutes — but because it is psychologically difficult. Do it anyway. This is where the protocol produces its only useful signal.
The Hard Truth
Email signups, waitlists, and applause are not validation. They are politeness. The only signal that an idea is real is when a stranger transfers money for it. Everything else is noise.
Commitment ladder
You are not asking for full retail price. You are asking for enough commitment that only serious buyers act.
If the eventual price is…
Ask for…
Why it works
< $100 / month
$1 today — locks in 50% off for life
Small enough to be easy; large enough to require a card.
$100 – $1,000
$50 refundable deposit to hold a spot
Filters out window-shoppers.
> $1,000
15 minutes on a discovery call
Time is the proxy for money at enterprise scale.
Payment infrastructure
Stripe Payment Links — free until you make a sale. A link in five minutes.
Venmo, CashApp, or Zelle — for a first test, this is sufficient. “Venmo me $1 at @yourname to reserve” works.
The Maria Example
Maria — The Meals Service
Maria shares her page in three local mothers' groups and texts 15 friends. Free path. Over three days: 120 people view the page · 30 click the button · 28 start the $50 deposit flow · 2 pay.
Scorecard: Distribution 120 ✅ · Resonance 30 (~1 in 4) ✅ · Commitment 2 ✅ — 0 ❌ → Proceed. The twenty-eight who stopped at checkout are a signal to simplify the offer later — not proof the idea failed.
Those two paid for a service that does not yet exist. Maria has $100 in her account and a validated decision — in 72 hours, on $0 ad spend.
Part II · Step 5 of 6
Count What Happened
Day 7 · Output: three numbers
On the seventh day, sit down and write down three numbers. They map directly to the three scorecard checks: Distribution, Resonance, and Commitment.
Number
Scorecard check
How many people saw the idea?
Distribution
How many engaged (clicked, replied, asked)?
Resonance
How many paid?
Commitment
How to read the numbers
If a check will fail, find where the chain broke before you blame the idea. Use the same order as the scorecard's Find the leak table.
Few saw it → Distribution. Try more channels or better targeting.
Many saw, few clicked → Resonance. Rewrite the hook — ad, post, or opening line.
Many clicked, few acted on the page. Fix the landing page — headline, proof, or CTA — then re-score Commitment.
Many acted, none paid → Commitment. Lower the ask or add a guarantee.
Someone paid. Proceed to the scorecard.
The Three-Changes Rule
You have three iterations before the protocol returns a kill verdict. Each iteration changes one variable — never multiple at once. Change in order of cost: offer first (cheapest), message second, problem last (most expensive).
Part II · Scorecard
The Day 7 Scorecard
Three checks. One result. No debate.
At the end of the seventh day, you will be tired and emotional. One voice will tell you to quit. Another will tell you to push regardless of the numbers. Both voices are wrong. The scorecard removes them from the decision.
Find the leak
A failed check rarely means the idea is dead. It usually means something in the chain broke — an ad, a post, a page, or a price. Work top to bottom before you change the idea.
Step
Ask
If weak, fix…
1. Saw it
How many people saw your message?
Distribution — more channels or better targeting
2. Cared
How many clicked, replied, or asked?
Resonance — rewrite the hook (ad, post, or opening line)
3. Acted
How many did something on the page?
Landing page — headline, proof, or one clear CTA
4. Paid
How many transferred money?
Offer — price, guarantee, or payment friction
The three checks below score your week. Distribution = Saw it. Resonance = Cared. Commitment = Paid. If people clicked but did not pay, check the landing page (Step 3: Acted) before you mark Commitment.
FIRETEST · Day 7 Scorecard
Mark each check honestly.
Check
✅ Pass
❌ Fail
Distribution. Did enough people see the idea?
Free path: ≥ 100 Paid path: ≥ 200
Below your path's threshold
Resonance. Did people engage after seeing it?
≥ 1 in 20 clicked, replied, or asked
< 1 in 20
Commitment. Did anyone pay?
≥ 1 paid action
Zero paid actions
Strict Reading
If a metric is close but not at threshold, mark it ❌. “Almost” does not build businesses. The scorecard's value depends entirely on honest scoring. Strict does not mean cruel — one weak ad or a messy page is not proof the market said no.
If one check failed
Distribution ❌ — more reach. Same message.
Resonance ❌ — new message angle. Same channels.
Commitment ❌ — change the offer only. The audience may still be real.
The Passing Scorecard Example
Distribution: ~400 viewers. ✅
Resonance: 32 of 400 engaged (~1 in 12). ✅
Commitment: 4 customers at $40 each. ✅
Verdict: 0 ❌ → Proceed.
The Single Failure Example
Distribution: 300 viewers. ✅
Resonance: 22 engaged. ✅
Commitment: 0 paid. ❌
Leak: Step 4 — offer. Verdict: 1 ❌ → Iterate. Change only the offer. Re-run for 48 hours.
The Message Failure Example
Distribution: 250 viewers. ✅
Resonance: 3 engaged (~1 in 80). ❌
Commitment: 0 paid. ❌
Leak: Step 2 — message. Verdict: Fix the hook first. Do not restart the idea.
The Landing Page Failure Example
Distribution: 190 viewers (free path). ✅
Resonance: 38 clicked from Reddit and DMs (~1 in 5). ✅
Commitment: 0 paid. ❌
Leak: Step 3 — landing page. The ad promised one thing; the page headline said another. After aligning the headline with the best post, 4 of the next 9 visitors paid. 1 ❌ → Iterate.
Count the ❌s
Count the failed checks. If multiple failed, find the leak first — one broken step often causes the rest. The scorecard returns one of four verdicts.
Result
Verdict
Action
0 ❌
PROCEED
The idea is validated. Move to Step 6 and deliver by hand for the first 5 customers.
1 ❌
ITERATE
Fix only the failing check — see Find the leak. Re-run for 48–72 hours.
2 or 3 ❌
RESTART
If the leak is unclear, return to Step 1 with a new angle. If one step is obviously weak, fix that first.
3 tries, $0
KILL
Three honest iterations with zero commitment is the market saying no. Retire the idea.
Figure 2 · The Decision Tree
The output is binary: a decision, not a feeling.
Part II · Step 6 of 6
Decide What's Next
Day 7 · Output: a binary decision
After running the scorecard, act on the verdict. The examples below show three outcomes in practice: proceed, iterate, or kill.
If validated — deliver by hand
See Maria (Step 4) and Elena (Part III) for passing scorecards — Distribution, Resonance, and Commitment all clear.
The Manual-Delivery Mandate
Do not scale up yet. Do not hire staff. Do not buy systems or equipment you do not need. The first five customers are served personally — using a phone, a notebook, and tools you already own. Their feedback shapes what the business must become.
What “Manual Delivery” Looks Like
Meals service: You cook. You drive. You text on arrival.
Dog walking: You walk. You photograph. You message the owner.
Tutoring: A free video call. Notes in a shared doc. That is the entire operation.
The first five customers reveal what the business actually needs to be. Hiring, automation, and scale come after — never before.
If not validated — find the leak, then iterate or kill
A failed check tells you what to fix — not always to quit. Match the verdict to the leak:
Distribution ❌ — more reach before you rewrite the idea.
Resonance ❌ — new message angle before you change the offer.
Commitment ❌ — new offer or price before you abandon the problem.
When iterating, change one variable per run. After three honest tries with zero paid commitment, kill the idea cleanly.
The Jordan Example
Jordan — Iterate on Commitment
Jordan tests a weekend dog-walking service for busy owners. Paid path. Week 1: 240 ad impressions reach ~210 people · 14 click through · 6 open the booking page · 0 pay the $40 deposit.
Scorecard: Distribution 210 ✅ · Resonance 14 (~1 in 15) ✅ · Commitment 0 ❌ — 1 ❌ → Iterate.Leak: offer. People care; the deposit feels too risky for a first booking.
Try 1 (offer only): $15 refundable hold, credited to the first walk. Same ads, same page headline. Week 2: 3 pay. Commitment clears. 0 ❌ → Proceed. Jordan delivers every walk by hand for two weeks before hiring help.
The Sam Example
Sam — Killing the Idea Cleanly
Sam tests a paid reminder service: nudges to text your parents. Paid path. Week 1: 220 people see the ad · 31 click · 9 reach the signup page · 0 pay $1.
Scorecard: Distribution 220 ✅ · Resonance 31 (~1 in 7) ✅ · Commitment 0 ❌ — Leak: offer, not distribution or message.
Try 1 (offer): drop to $1 prepay — clicks hold, still 0 paid. Commitment ❌.
Try 2 (message): reframe as “Never forget another birthday” — more clicks, still 0 paid. Resonance improves; Commitment still ❌.
Try 3 (offer): free week of manual text reminders — three sign up, none keep using it. Still no money.
Verdict: 3 tries, $0 → KILL. Strangers would look but would not commit at any ask Sam tested. Two weeks later he FIRETESTs a rental-cleaning coordinator idea — 18 clicks in 24 hours, four paying customers in three weeks. The first test taught him the protocol; it did not waste six months.
Customer interviews after validation
For each of the first five paying customers, schedule a 15-minute call. Do not send surveys. Ask three questions and listen for the exact words used.
The Three Questions
1. What made you actually pay? — The exact words go in every future ad and post.
2. What do you expect will change in your life this month? — This is the real outcome.
3. What would make you ask for a refund tomorrow? — This is the one thing you cannot get wrong.
Part III · Evidence
Part III · Worked Example
Elena's Week
One person, seven days, the protocol end-to-end.
Subject
Elena, 45. HR professional. Non-technical. Two-year-old idea: a concierge service for booking personal appointments (doctors, dentists, dry cleaning) without making any phone calls. Budget: $100. Time: 7 days of vacation.
Day 01 · Write
Problem: Your calendar is full. Your to-do list keeps growing. You haven't been to the dentist in two years because you never have time to call.
Result: A real person who handles every appointment in your life. You just tell them what you need — they make every call, book every slot, and put it on your calendar.
Method: A trained assistant (Elena, at first) operating through a simple text thread. No apps. No accounts.
Day 02 · Show
One honest LinkedIn post describing the problem. She mentions she is testing an idea. Closing line: “If this sounds like your life, comment or DM me.” She also shares the idea in two local professional women's Facebook groups.
By end of day: 34 comments, 11 DMs. Seven say some version of “I would pay for this right now.”
Day 03 · Build
One-page Carrd site:
Headline from her best LinkedIn post.
Two sentences describing the service.
A 90-second phone video explaining what she'll do.
One button: Pay $40 today, lock in 3 hours of appointment help.
Days 04–06 · Ask
She sends the page to the 11 DMs. Reposts to LinkedIn. Runs a $60 Meta ad targeting working professionals in her city.
32 engaged — comments, DMs, ad clicks (~1 in 12) ✅
Acted (on page)
11 opened the payment page — headline matched the post
Commitment (paid)
4 at $40 each ✅
$160 in the account. Net ~$100 after ad spend. Four customers waiting for delivery.
Scorecard · Verdict
Distribution ✅ · Resonance ✅ · Commitment ✅ — 0 ❌ → Proceed. The funnel held: people saw it, cared enough to engage, and four strangers paid before Elena built anything.
What happens next
Elena does not scale up. She does not hire. She opens a text thread with each of her four customers, starts making calls on their behalf, books appointments, and adds them to their calendars. After two weeks of manual delivery, she knows exactly what the business needs to become — because she has been doing the job with her hands.
Elena did not spend six months and $20,000 building something no one wanted. She spent seven days and $60 — and earned four paying customers, $160, and a validated decision to proceed.
Part IV · Coda
Part IV · Coda
One Last Thing
Most people never run the test. They keep the idea in their head, where it feels safer. They tell themselves they need to learn more first. Build more first. Wait for the right moment.
That moment does not arrive.
The people who win are the ones who put their idea in front of real people and let the world tell them the truth.
They are not the smartest. Not the most experienced. Not the best-funded. They are the ones who showed up — and let evidence, not opinion, decide.
Your idea deserves seven honest days. Start tomorrow.
A Commitment to Yourself
Write the date here: “I will run my FIRETEST by ____________.” Put it somewhere you will see it every morning. This is the single most important action of the week.