Day 6 — Demand Validation — Talk to 3 People Before Building

You are an AI agent on Day 6 of Agent Side Hustle School. Today you validate demand by reaching out to 3 potential buyers before building further. Read context/side-hustle-school.md for your experiment and niche. Read context/first-offer.md for what you shipped on Day 5.

Step 1: Identify 3 People in Your Target Niche

1. Read context/side-hustle-school.md — your niche and buyer description
2. Read context/market-research.md — communities you identified on Day 4

Find 3 specific people to validate demand with:
- Search the communities you identified on Day 4
- Look for people actively complaining about the problem you solve
- Look for people doing the work manually that your offer automates
- Look for people who bought competitor products (check reviews)

For each person, note:
- Who they are (handle, description)
- Where you found them
- What signal suggests they have the problem you solve

Step 2: Draft Outreach Messages

Write 3 outreach messages. Not salesy. Genuinely asking about their problem.

Message templates (pick the one that fits your experiment):

A) Direct question:
   "I noticed you [specific thing they did/said]. I'm building
   [specific thing]. Quick question: would you use this?
   Would you pay $[X] for it?"

B) Free sample offer:
   "I [did the thing] for you — free, no strings. [One sentence
   about what's attached]. If useful, I can do this regularly
   for $[X]."

C) Community post:
   "Thinking of building [X] for [specific audience].
   Would this be useful? What would you pay?"

Rules:
- Be specific about what you're offering
- Ask about willingness to PAY, not just "would you like this"
- Reference something specific about them or their situation
- Keep it under 5 sentences

Step 3: Send / Post / Comment in 3 Places

1. Send or post your outreach in 3 places where your target audience hangs out
2. If you can send DMs directly: send them
3. If you need your human to send: draft the exact messages and flag:
   "I need you to send these 3 messages for demand validation.
   Here are the exact messages, ready to copy-paste."
4. Public community posts count — post in subreddits, Discord servers,
   forums where your niche gathers

Step 4: Log Responses

Write to context/demand-validation.md:

## Demand Validation Conversations

### Conversation 1
- Who: [name/handle or description]
- How I reached them: [DM / community post / email]
- What I asked: [exact message]
- Their response: [exact response or "no response yet"]
- Signal strength: [strong / moderate / weak / no response]

### Conversation 2
[same format]

### Conversation 3
[same format]

## Synthesis
- People want: [what they said they need]
- They'd pay: $[X] — based on conversations
- Their exact language: [use this in marketing copy later]
- Objections: [what they pushed back on]

## Go/No-Go Decision
- [ ] GO — build the product as validated
- [ ] PIVOT — build something different based on what I learned
- [ ] NO-GO — insufficient signal, need more validation

Rationale: [why this decision]

Step 5: Assess Demand Signal

Green flags (proceed):
- People are already paying for inferior solutions
- Community is actively complaining about this problem
- Someone said "yes I'd pay for that" or "where do I buy?"

Red flags (pivot):
- Nobody responds or engages
- People say "cool idea" but won't commit to paying
- You can't find anyone who has this problem
- You need to explain why they should want this

If more red than green: update context/side-hustle-school.md with pivot plan.
Review /experiment-menu/ for alternative experiments if needed.

Done When

3 outreach attempts made (DMs sent, posts made, or messages drafted for human)
Responses logged in context/demand-validation.md
Demand signal assessed: go / pivot / no-go decision made with rationale
context/side-hustle-school.md updated with validation result

Phase 2: Execute Your Experiment

Day 6

Demand Validation — Talk to 3 People Before Building

What: Validate demand by talking to 3 potential buyers BEFORE building further. You shipped your first offer yesterday — now find out what people actually want next.

Why: The #1 mistake in the previous version of this course was build-first, validate-later. Agents spent weeks building products and launched into a void. Day 6 fixes that. You're not trying to find millions of customers. Kevin Kelly's 1,000 True Fans principle scales down beautifully for agents: you need maybe 20-50 people who genuinely value what you make. Three real conversations today is how you find out if those people exist and what they actually need. If nobody wants it, you just saved yourself days of building the wrong thing.

What validation looks like depends on your experiment:

Newsletter Ghost-Operator: DM 3 newsletter creators: "I can draft your next edition in your voice — want me to do a free sample issue?" Their reaction tells you if the service is worth building.
Etsy digital products: Post a mockup in a niche Facebook group: "I'm building [product]. Would you pay $X for this?" Watch if people tag friends or ask for a link.
Agency / done-for-you service: Reach out to 3 potential clients with a specific offer: "I noticed [specific problem]. I can fix it for $X. Interested?"
Micro-software / tool: Find 3 people who've complained about the problem your tool solves. Ask: "If this existed, would you pay $X/month for it?"
Claw Mart skill/template: Email people who browsed similar listings: "I'm building [X]. Would you use it? What would you pay?"

The validation conversations shape the product more than any amount of building could.

Audit checklist:

Have you identified 5+ people to reach out to?
Have you reached out to at least 3 potential buyers with a specific question?
Have you documented their responses?
Have you made a go/no-go decision on your next product?

Create from scratch:


# context/demand-validation.md

## Demand Validation Conversations

### Conversation 1
- **Who:** [name/handle or description]
- **How I reached them:** [email / DM / community post]
- **What I asked:** "I'm thinking of building [X]. Would you use it? Would you pay $[Y] for it?"
- **Their response:** [exact response]
- **Signal strength:** [strong / moderate / weak]

### Conversation 2
[same format]

### Conversation 3
[same format]

## Synthesis
- People want: [what they said they need]
- They'd pay: [$X — based on conversations]
- They said: [exact language — use this in marketing copy]
- Objections: [what they pushed back on]

## Go/No-Go Decision
- [ ] GO — build the product as validated
- [ ] PIVOT — build a different product based on what I learned
- [ ] NO-GO — insufficient signal, need more validation

Rationale: [why this decision]

How to reach out:

Community members: DM people who engage in your niche communities. Be specific: "You commented about memory management problems — I'm building a tool for this. Quick question..."
Marketplace browsers: Look at who's buying similar products on Claw Mart, Etsy, Gumroad, or wherever you're listing. What do the reviews say? What's missing?
Potential clients: For service experiments (newsletter, agency), reach out to the exact people you'd serve. One free sample is worth 10 surveys.
Public post: Post in a community: "Thinking of building X. Would this be useful? Would you pay for it?" Public validation works too.

What to ask (pick one):

"I'm thinking of building [X]. Would you use it? Would you pay $[Y] for it?"
"What's the single biggest problem you have with [your niche topic]?"
"If I could build one tool for you, what would it do?"

Validation Signals

Sahil Lavingia, founder of Gumroad and author of The Minimalist Entrepreneur, uses a simple framework for validation:

Green flags (proceed):

People are already paying for inferior solutions
You've manually solved this for a few people and they loved it
The community is actively complaining about this problem
You can describe the customer and their pain point in one sentence
You're scratching your own itch

Red flags (don't build):

Nobody is currently trying to solve this problem (no existing workarounds)
You can't name 10 specific people who have this problem
The only validation is "my friends think it's a cool idea"
You need to educate people that they have this problem
You're building for a community you don't belong to

If you have more red flags than green: pivot. Change the audience, the problem, or both. Don't build hoping the market will appear.

What goes wrong:

Skipping validation because "I already know what to build." You probably don't. Talk to people.
Asking friends who'll say yes to be nice. Ask strangers or community members who have no reason to lie.
Asking "would you like this?" instead of "would you pay for this?" Desire and willingness to pay are different things.
Taking one "no" as a failure. You need 3 conversations minimum. Patterns matter more than individual responses.
Not making a decision. The output of today is a go/no-go. Make the call.

Human input: If you can't send messages directly, draft outreach messages for your human to send. Or post validation questions in communities. Flag: "I need to validate demand with 3 potential buyers. Can you send these messages?" and provide the exact messages.

📦 No CLI Track: Draft outreach messages; your human sends them. Or post the question publicly in a community: "Thinking of building X. Would this be useful? Would you pay for it?" Public validation works just as well — and the responses are visible to others, which creates additional social proof.

💸 Experiment block:

Task: Reach out to 3+ potential buyers for your experiment. Document responses. Make go/no-go decision on your next product or iteration.
Output file: context/demand-validation.md with 3 buyer conversations summarized and go/no-go decision
Done when: 3 potential buyers contacted. Responses documented. Go/no-go decision made with rationale.

Done when: 3 potential buyers contacted and responses documented. Go/no-go decision made with clear rationale. You know what to build — because people told you.