Thread Transfer Blog

Stop guessing which product to build next.

The Discovery Machine is a repeatable system that turns raw ideas into validated product candidates in 2–4 days. Templates, scoring rubrics, and cheap experiments—all in one kit.

This is for you if:

  • You're already using AI daily but can't get consistent validation outcomes
  • You've shipped before and want a repeatable system, not random prompts
  • You're tired of building things nobody wants
  • You value evidence over vibes
  • You want a 2–4 day validation cycle, not months

This is NOT for you if:

  • You've never shipped a product and want magic
  • You're looking for a 'make money fast' scheme
  • You don't want to do the work—templates don't replace effort

What your product validation looks like right now

Most builders skip validation entirely or do it wrong. Here's what that costs you:

Problem #1

Months spent building features nobody asked for

Problem #2

Random Reddit posts instead of structured research

Problem #3

No kill criteria—you just 'feel' when something is dead

Problem #4

Copying competitors without understanding their wedge

Problem #5

Expensive launches that validate nothing

Problem #6

The same vibes-driven mistakes on repeat

How the Discovery Machine works

A 4-step system that runs in 2–4 days per validation cycle

1

Mine 5 problem sources

Network, competitor reviews, forums, job posts, marketplaces. Collect raw pain points with evidence links.

2

Write candidates in 6-line format

Buyer role, trigger event, job-to-be-done, workaround, Week-1 win, wedge type. No fluff.

3

Score with the 8-category rubric

Pain intensity, buyer can pay, distribution clarity, wedge strength, proof speed, retention loop, build speed, giant risk. Pass threshold: 16/24.

4

Run the cheapest experiment

Smoke test, fake door, concierge MVP, or Wizard-of-Oz. Numeric kill thresholds tell you when to stop.

INPUT
PROCESS
OUTPUT
2-4 Day Cycle

What's inside the kit

Everything you need to run evidence-based product validation

Candidate Template

The 6-line format that forces clarity. Buyer, trigger, job, workaround, win, wedge.

Scoring Sheet

8-category rubric (0–3 each) with Person Gate and pass thresholds.

Source Playbooks

How to mine Network, Competitor Reviews, Forums, Job Posts, and Marketplaces.

Experiment Library

Smoke test, fake door, concierge, Wizard-of-Oz templates with kill thresholds.

Runbook

2–4 hour execution checklist per discovery cycle. Copy-paste ready.

AI Operator Prompt

Paste into ChatGPT/Claude to run the machine with an AI copilot.

Wedge Playbooks

Workflow, Vertical, Channel, Integration, and Compliance wedge strategies.

Winner Brief Template

What to produce when a candidate passes: MVP scope, distribution angle, pricing hypothesis.

What's NOT included:

Custom consultingHosted softwareMagic results without work

Built for builders who ship

This isn't theory. It's the same system used to validate real products.

2–4 days

Validation cycle

8

Scoring categories

16/24

Kill threshold

5

Wedge types

validation-output.log

[Discovery] Running validation cycle...

[Score] Candidate passed: 18/24

[Result] Moving to experiment phase

Ready to build.

Simple pricing

One kit. One price. Yours forever.

Recommended

Discovery Machine Kit

$10one-time

Everything you need to validate product ideas with evidence, not vibes.

  • All templates (candidate, scoring, evidence pack)
  • 5 source mining playbooks
  • 5 wedge strategy guides
  • Experiment library with kill thresholds
  • AI operator prompt
  • Runbook + execution checklist
  • Lifetime access + updates
Get the Kit

Questions

No. It's a complete validation system: templates, scoring rubrics, source playbooks, experiment designs, and kill thresholds. The AI prompt is one component—it helps you run the system faster, but the system works without AI too.

The kit is designed for builders who've shipped at least one thing before. If you've never built a product, start there first. This system assumes you know what 'shipping' means.

No. The operator prompt works with ChatGPT, Claude, or any capable LLM. You can also run the system manually with just the templates.

The templates are designed for B2B/B2C software products. If you're building hardware, physical goods, or regulated industries with unique constraints, you'll need to adapt the rubrics.

Yes. When the system improves, you get the updates. No subscription, no extra charge.

Stop validating with vibes.

Get the system. Run a discovery cycle. Know what to build.

Get the Kit — $10

One-time purchase. Instant download. No subscription.