Start here · Lesson 1.2
12 Brand Brain AI Examples to Adapt (Illustrative)
It is easier to write a brand brain AI file when you can see one. Here are twelve short examples across niches. They are illustrative and invented, not real customers, and none of them reports a result. Each shows a goal line, a voice note and why the entry is useful, so you can borrow the shape for your own brain.
The full setup is in Lesson 1.2, Build your brand's brain. These examples fill the file the lesson helps you build.
A good brain entry is specific, short and true. Each example below gives a goal, a voice cue and a never say item. Copy the shape, not the words.
How to read the examples
Every example has three parts: a goal line, a voice cue and a never say item. Those three lines do the most work in a brain, so they are the ones shown. Add the rest of your brain file from the free template.
The twelve examples
1. Neighbourhood bakery. Goal: more locals see us. Voice: warm, a bit funny. Never say: "artisanal". Why it works: a concrete audience and a banned word stop generic food copy.
2. Independent bookshop. Goal: followers from nearby readers. Voice: bookish but not snobbish. Never say: "must read". Why it works: the never say item removes the most overused phrase in the niche.
3. Freelance photographer. Goal: bookings from engaged couples. Voice: calm and reassuring. Never say: "capture memories". Why it works: it names who is buying and the emotion they feel.
4. Personal trainer. Goal: followers, then enquiries. Voice: encouraging, no shame. Never say: body shaming language or guaranteed results. Why it works: it protects the brand from risky claims. Check platform and local rules for fitness claims from official sources.
5. Plant shop. Goal: views on care tips. Voice: friendly, plain. Never say: "easy to keep alive" for fussy plants. Why it works: honest expectations build trust.
6. Local accountant. Goal: enquiries from small business owners. Voice: clear, jargon free. Never say: specific tax advice in posts. Why it works: it keeps content educational and avoids advice you are not licensed to give in public posts. Confirm rules with your professional body.
7. Craft coffee roaster. Goal: orders from repeat buyers. Voice: curious, not snobby. Never say: "best coffee". Why it works: it replaces unprovable superlatives with detail.
8. Yoga studio. Goal: class sign ups. Voice: gentle and practical. Never say: medical claims. Why it works: it avoids health claims you cannot support.
9. Pet groomer. Goal: views, then bookings. Voice: playful. Never say: before and after claims about pet health. Why it works: it keeps content to what you can show.
10. Small software tool. Goal: trials from small teams. Voice: plain, direct. Never say: "revolutionary". Why it works: it matches buyers who are tired of hype.
11. Wedding florist. Goal: enquiries from engaged couples. Voice: romantic but practical. Never say: prices unless confirmed. Why it works: it avoids stale prices in captions.
12. Online course creator. Goal: followers, then waitlist sign ups. Voice: teacher like, patient. Never say: income promises. Why it works: it avoids claims that mislead and may break platform rules.
| Pattern | Seen in examples | Why to copy it |
|---|---|---|
| Specific audience in the goal | 1, 3, 6, 10 | Stops writing for everyone |
| Banned cliché | 1, 2, 3, 7, 10 | Removes the most overused phrase |
| Claim guardrail | 4, 6, 8, 9, 12 | Avoids risky promises |
| Prices left out | 2, 11 | Prevents stale numbers |
Adapting an example
- Pick the closest nicheChoose the example nearest to your brand, even if it is not an exact match.
- Rewrite the goalPut in your real audience and stage, with a date.
- Replace the voicePaste one real sample of your writing.
- Make your own never say listAdd the clichés in your niche and any claims you cannot support.
- Test itAsk for a short post in a fresh chat and compare the tone.
What the examples cannot tell you
They do not show that a given line improves performance. They show structure. Whether it works for your brand is for you to test and track, and the metrics post explains how.
A note on claims
Several examples touch on health, finance or income. Those areas have platform and legal rules that vary by country. Verify them from official sources, not from an assistant or a blog. A never say line is a guardrail, not a compliance check.
How the brain uses these lines
Anthropic's documentation says Claude Code reads CLAUDE.md files at the start of every session. A brain file with these three lines near the top is therefore in front of the assistant before you ask for anything. Put the important lines first.
Writing your own thirteenth example
After reading twelve, write one for your brand in five minutes. Start with the goal line: who do you want to reach and what stage are you at. Then write a voice cue using a real sample from something you already published. Finally, list three things you never want to say, drawing on drafts that annoyed you. Paste all three at the top of your brain file. Then ask for a short post in a fresh chat and see whether it sounds like you. If not, tighten the weakest line instead of adding new ones. Over time, replace general lines with specific ones taken from real customer wording. A line like warm and friendly is a start, but a line quoting how you actually greet customers is stronger because the assistant can imitate it.
Next steps
Build the whole file in Lesson 1.2, Build your brand's brain, which walks through the scan, goal, folder and media sort. For tool setup, see Lesson 1.1. To keep the file fresh, use the weekly routine.
Questions people ask
Are these real brands?
No. All twelve are invented for illustration.
Do they prove the approach works?
No. They show the shape of an entry. You need to test results yourself.
Can I copy them?
Copy the structure and write your own words.