Bloom is now Rainbrand

By Tomas Ward

Published

Updated

3 min read

Brand voice examples, and how to write rules AI can follow

Learn from seven published voice guides, then turn their useful habits into rules, examples and tone instructions for your own agents.

TL;DR

Learn from how published voice guides turn a brand's voice into instructions: clear limits, preferred wording, paired examples and tone that fits the situation. Choose rules for your own audience, and make them specific enough that an agent can apply them and a reviewer can check the result.

Seven published voice guides

Study the instructions behind the copy. Each of these public guides turns a trait into a writing decision an agent can follow.

Guide and audienceRule and application
Mailchimp, its contentClarity takes priority over humour. Freddie, its mascot, never speaks. Name the priority when traits conflict, and give explicit limits.
GOV.UK, government contentUse familiar words and explain specialist terms. Its replacement list preserves technical exceptions. Give replacements and exceptions together. “Deploy”, for example, remains useful in software and military contexts.
Microsoft, products and contentReplace “Invalid ID” with “You need an ID that looks like this: someone@example.com”. Show an actual edit and explain the improvement: here, the required input becomes clear.
Google, developer documentationPlace a recommended sentence between versions that are too informal and too formal. Show the acceptable range for traits such as “conversational”.
Atlassian, interfacesVary tone by audience and message type; a flourish can become annoying when repeated. Specify different treatment for repeated errors and one-time completion messages.
Shopify, admin appsConfirm a routine action with “Product saved”. Explain the impact and next steps for serious problems. Choose tone from the observable situation, without assuming the reader's mood.
Monzo, bank communicationsIts channel table sets warm wit to “None” for operational communications and customer service. State the level by channel, then show examples of what each level means.

Keep the guide's audience in mind. A rule for an error message does not show how a whole article should develop. If you use named traits, explain them too: Monzo's “everyday magic” means a welcome touch in the experience, rather than magical vocabulary.

Resolve conflicting rules for your audience

These guides disagree on useful details:

Choose the rule that serves your readers and record the reason. Copying all the rules would give an agent conflicting instructions.

Turn an example into your own instruction

For one rule from your own guide, fill this record. Anthropic's prompting guidance recommends relevant, varied examples and the reason behind an instruction.

Source rule and link:
Audience and message type:
Writing decision this rule settles:
Reason for the choice:
Approved copy showing the rule:
Earlier draft and the edit it needed:
Facts the next task must supply:

Use approved copy and explain the edit. If the source gives only a trait, test a proposed instruction before adopting it. Keep changing facts separate so an old example does not supply a new message's diagnosis, price or date.

Ask an agent to identify the gaps:

Ask your agent
Read our brand voice guide. For each trait, identify its scope, practical rules, reasons, examples and any priority when rules conflict. Quote the source passage for each finding. Flag adjectives that still leave the writing decision open. Draft proposed additions separately, using our own approved copy where available. Do not treat a proposed example as an existing brand fact or change the maintained guide yet.

Keep the source and the rule together

Keep each rule linked to its source and your own example. Rainbrand maintains brand context that connected agents can use:

Once the brand is ready, or after applying an update, request a new Brief, the guidance and files for a task. Name its audience, channel and situation so the relevant tone rules can be selected.

Start with one recurring disagreement and record its rule, reason and near-miss. The voice-guide tutorial shows how to build and test the rest.

  • 4 min read

    How to keep AI writing in your brand voice across tools

    ChatGPT, Claude, Gemini, Jasper and your own agents each bring their own defaults, settings and copy of your guide, so one voice comes out several ways. Write the voice in the parts that carry across models, keep one version every tool reads, and check the same tasks in each tool.

  • 6 min read

    How to write brand voice guidelines AI tools can follow

    Build a voice guide from real edits, with rules, examples, tone by situation and a worksheet you can test on your agents' writing tasks.

One brand, every agent.

Connect your agents