# The Skill Extraction Prompt **From Chapter 2.2 of *The 4-Hour Side Hustle*.** The prompt that finds the product hiding in your day job — the one you can't see because you've been doing it for years. --- ## Before you run it You need a transcript of yourself describing your work. Not notes. Not bullet points. **Talk, and record it.** Fifteen to twenty minutes, out loud, into your phone's voice recorder or any transcription tool. Describe: - What you actually did last Tuesday, hour by hour - The part of your job people ask you about - The thing you fix that nobody else on your team can - What you complain about - What you assume everyone knows **The last two matter most, and they're the ones people leave out.** Expertise you complain about is expertise you have. Knowledge you assume is universal is almost never universal — that assumption is precisely what makes it invisible to you and valuable to someone else. Don't edit the transcript. Don't tidy the digressions. The passing remarks are what the prompt is looking for. --- ## Pass 1 — extract > You are a business analyst who specializes in productizing professional expertise. Below is a transcript of me describing my day job. I have been doing this work for years and I can no longer see which parts of it are valuable to other people. > > Your task: > > 1. Identify every piece of specialized knowledge in this transcript — including things I mention in passing, complain about, or clearly assume everyone knows. Pay particular attention to anything I describe as obvious. > 2. For each one, name who else would pay for it, and roughly what they'd pay. > 3. Propose 10 specific digital products. Each must be deliverable as a file, a template, or a short course — nothing requiring my ongoing time per customer. > 4. For each of the 10, note the single biggest reason it would fail. > > Rules: Do not invent expertise I did not describe. Do not propose generic products (courses on productivity, general consulting, "coaching"). Prefer boring and specific over exciting and broad. If something in the transcript suggests a product I'd be uniquely credible selling, say so explicitly and explain why. > > Transcript follows. --- ## Pass 2 — argue against it Send this as a **second message**, in the same conversation: > Now argue against your own top three. For each, tell me why a busy professional with fifteen hours a week should NOT build it. **Do not skip this.** Pass 1 is a model being agreeable — it will find ten products in almost any transcript, because that's what you asked for. Pass 2 is where the useful answer lives. What survives an argument with itself is worth taking to the scorecard. What collapses under one paragraph of scrutiny was never a product; it was a plausible sentence. --- ## Reading the output Ignore the exciting ones. This is genuinely counterintuitive and it's the whole trick. The product that makes you think *"that's a real business"* is usually broad, competitive, and something you'd be one of thousands selling. The product that makes you think *"surely everyone already knows this"* is the one where you're one of a few hundred people on earth who could write it credibly. **Three questions for each survivor:** 1. **Could I write the first version in a weekend?** If not, it's a book or a course, and it's a year, not a quarter. 2. **Would I be embarrassed if a peer bought it?** Good sign, oddly. It means it's specific enough to be recognizably yours. 3. **Does it need me after it ships?** If every sale costs you an hour, you've found a service. Services are fine businesses. They are not this book's business. Anything still standing goes to the Chapter 2.1 scorecard. Anything scoring above 18 goes to the 72-hour demand test. --- ## Why AI is the wrong tool for the first half of this Worth naming, because it explains the shape of the prompt. AI cannot tell you what you know. It has no access to your Tuesday. What it's genuinely good at is the second job — taking a pile of things you *said* and spotting the patterns you've stopped noticing, because you're too close and it isn't close at all. That's why the input is a transcript of you talking rather than a question you answer. You supply the expertise. The model supplies the outside view. Reverse those and you get ten generic product ideas that would suit anybody, which is the same as suiting nobody. --- ## Checklist - [ ] 15–20 minutes recorded out loud, not written - [ ] Includes what you complain about and what you assume everyone knows - [ ] Transcript un-edited, digressions left in - [ ] Pass 1 run with the transcript pasted at the end - [ ] Pass 2 run as a separate message in the same conversation - [ ] Exciting-sounding products discounted; boring specific ones examined - [ ] Each survivor tested against the three questions - [ ] Survivors scored on the Chapter 2.1 scorecard - [ ] Anything above 18 booked in for the 72-hour demand test