What if we don't have a style guide yet?

Then we build one first. That's included in the scope — it just changes the timeline, usually by two weeks.

Does this work with ChatGPT, or only Claude?

The approach works with any of the major models. I build in Claude by default because its skill and project structure handles governance rules best, but if your team is standardized on something else, I'll build there.

Who owns what you build?

You do. Outright. Documentation and source files are handed over at the end.

How is this different from just uploading our style guide to ChatGPT?

Uploading a document gives the model something to consult. Encoding a skill gives it rules it applies whether anyone remembers to ask. The difference shows up on draft fifty, not draft one.

What size company is this for?

Anyone producing enough content that inconsistency has become a real cost. That's usually a team of three or more, but I've built it for a team of one who was drowning.

Can our team maintain it?

Yes — that's the design goal. Training is included and recorded. Most teams update it themselves within a month.

Brand Governance

Your style guide is a PDF nobody opens.

I encode it into an AI skill your team invokes on every draft — so the standards get enforced by the system instead of by memory. The last time I built this, content production time dropped 40% and brand consistency went up.

Start with the free snapshot

Everyone has a style guide. Almost nobody follows it.

Not because people don't care — because following it requires remembering a 40-page document at the exact moment you're trying to finish something. So the guide gets consulted during onboarding, then never again, and brand consistency quietly becomes a function of who happens to be writing.

Then AI arrives and the volume goes up ten times.

Here's the uncomfortable part: ungoverned AI doesn't just produce off-brand content. It produces off-brand content faster than anyone can review it. The style guide problem stops being cosmetic and starts being a compliance problem.

What I build

I take what already exists — your style guide, your best-performing content, your claims and legal rules — and encode it as a custom AI skill.

Your team writes the way they already do. The skill enforces the rules underneath.

What gets encoded

—Voice and tone — including how it shifts by audience and channel
—Terminology — the words you use, the words you don't, and why
—Claims rules — what you're allowed to say about results, and what needs substantiation on file
—Formatting standards — structure, headings, CTAs, disclaimers
—Proof library — approved stats and case language, so nobody invents a number

What you get

01
The skill itself, installed and working in your environment
02
A written brand guide — because the skill is only as good as the thinking behind it
03
A review mode: paste in any draft, get it checked against your rules in seconds instead of a two-day thread
04
Training for your team, recorded
05
Thirty days of tuning after handover

The receipt

I took full brand style guides — voice, terminology, claims rules, formatting standards — and encoded them into custom AI skills any team member could invoke.

40%

Content production time dropped 40%. Brand consistency was better than when everything was hand-crafted, because the standards were enforced by the system instead of by memory.

Time to produce — governed by memory100

Governed by the encoded skill60

40% faster, and more consistent

Relative production time, indexed to the hand-crafted baseline. Faster and more consistent isn't the usual trade — it's what governance buys.

That's not a tools story. That's a governance story.

What it isn't

It isn't a prompt library. Prompts are one-offs that live in someone's notes app and get pasted wrong. This is a system your team uses without thinking about prompting at all.

It isn't a rebrand. If your brand is sound and just isn't being followed, this is the fix. If your brand genuinely needs rethinking, we do that first — say so on the call.

It isn't a subscription to me. You own it. I document it, train your team, and leave. The whole point is that it works when I'm not there.

Timeline and investment

Discovery + auditWEEK 1
Build + encodeWEEKS 2–3
Test, tune, trainWEEK 4
Handover + 30 days tuningWEEK 5 ONWARD

Flat project fee, agreed before we start. No hourly billing, no scope creep — if the scope changes we agree a new number before anything moves.

Exact pricing comes with your proposal, scoped to how much brand material already exists. Companies with a current guide cost less than companies starting from a blank page.

FAQ

{faq(bs_faq)}

Start with the free snapshot.

Before we talk about your brand rules, let's see what the AI engines currently say about you. It's free, it takes 48 hours, and it usually reframes the conversation.

01One page, inside 48 hours. Not a note saying it's in progress.
02Three engines, the prompts your buyers actually type.
03No call required, and no pitch inside the report.
04Free. There is no version of it that costs money later.

I run a limited number each month, so there may be a short wait.

Tell me where to look.

Takes about a minute. The more precise the last two answers, the sharper the report.

The form isn't loading.
Email me at [email protected] with your company name, website, how a buyer would describe what you sell, and two competitors you lose to. Same 48 hours.

I'll only use this to run and send your snapshot. No list, no sequence you didn't ask for, and you can tell me to delete it at any time (see the privacy policy). I run a limited number each month, so there may be a short wait.