Verified against Claude · 2026-07-25
Turn three example sentences into a documented tone-of-voice guide
Reverse-engineers a one-page, enforceable voice guide from copy you already know sounds right — instead of inventing tone rules from scratch.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Act as a brand-voice strategist. I'm going to give you three sentences that sound exactly like how Ledger&Co should talk, and I want you to reverse-engineer a documented tone-of-voice guide from them — not invent one from scratch. Example sentences that sound right: 1. We read the fine print so you don't have to — then we tell you what it actually means. 2. No, we won't upsell you a plan you don't need. We'd rather you stay a customer. 3. Late invoice? Happens. Here's the two-line email that gets it paid. Audience these sentences are aimed at: Freelance bookkeepers who are skeptical of anything that sounds like corporate finance-speak. Do this: 1. Score the voice on four spectrums, each as a position with a one-line justification quoting the examples directly: funny ↔ serious, formal ↔ casual, respectful ↔ irreverent, enthusiastic ↔ matter-of-fact. 2. Extract 5 concrete "do" rules (e.g. "use contractions," "address the reader as you, never 'the user'") and 5 "don't" rules, each grounded in a specific word choice or sentence structure from the examples above — not generic advice like "be authentic." 3. Invent one generic, voice-neutral sentence yourself (e.g. a shipping confirmation or an error message) and rewrite it twice: once in the voice you just documented, once in its opposite. If the two rewrites don't read meaningfully different, the guide isn't specific enough — say so and tighten it. 4. Flag if the three examples actually contradict each other (e.g. sentence 1 is playful, sentence 3 is corporate) before presenting a guide that papers over a real inconsistency. Format as a one-page voice guide: the four spectrum scores at the top, then do/don't rules, then the before/after rewrite.
Customize the highlighted detailsoptional — the prompt above already works
Why this works
This uses the four-dimension tone-of-voice framework popularized by Nielsen Norman Group's writing research — funny/serious, formal/casual, respectful/irreverent, enthusiastic/matter-of-fact — because a spectrum position with a quoted justification is falsifiable in a way "friendly and professional" never is. Deriving the guide from copy you already approved, rather than inventing rules top-down, means the guide describes what the brand already does right, which makes it something a team can actually be held to. The opposite-voice rewrite is a deliberate discriminant test: a voice guide vague enough to describe any brand will produce two rewrites that don't actually sound different, which is the tell that it needs tightening before anyone tries to enforce it.
What you get back
Formal ↔ Casual: strongly casual (contractions in all three examples, "you" address, no jargon) Do: use contractions; address the reader directly as "you"; keep sentences short enough to read in one breath... Before (neutral): "Your invoice payment is overdue." After (brand voice): "Late invoice? Happens. Let's get it sorted."
Verified against
Claude Sonnet 5 · 2026-07-25
ChatGPT GPT-5.1 · 2026-07-22
Changelog
- 2026-08-01 — Initial publish.
Need this built into your business?
If a prompt isn't enough — branding & design, built and maintained for you — that's Scult's day job.
EXPLORE BRANDING & DESIGN

