# AI Side-Project Validation Checklist

Use this before spending serious time or money on an AI side project.

## 1. Reader/customer problem

- Who has the problem?
- What are they doing manually today?
- What would be useful even without AI?
- What would be risky or misleading to automate?

## 2. Smallest useful version

- One clear workflow
- One target user
- One output format
- One measurable success signal
- No promises about income, traffic, rankings, or approval

## 3. Source and policy check

Useful references:

- https://developers.google.com/search/docs/fundamentals/creating-helpful-content
- https://support.google.com/adsense/answer/10502938

- List the official docs, pricing pages, or policy pages the project depends on.
- Add a last-checked date.
- Note what could change.
- Avoid using private or sensitive data until you have reviewed privacy and terms.

## 4. Validation plan

- Show the idea to 3–5 target users.
- Ask what they already pay for or spend time on.
- Test a manual version before automating everything.
- Track whether people would bookmark, share, pay, or request a template.

## 5. Monetization path

Pick one realistic path:

- free tool with ads later
- paid worksheet/template
- affiliate recommendation with clear disclosure
- service package
- paid version only after useful free value exists

## 6. Risks and limitations

- What could the AI get wrong?
- What should the user verify manually?
- What should this project never promise?
- What would make the project not worth building?