What is Brand Voice?
Brand voice is a brand’s verbal identity: the character that stays constant across all texts, regardless of who happens to be writing. It is defined as a set of three to five attributes (e.g. "precise, direct, slightly cheeky") and shows up in concrete decisions: formal or informal address, active or passive, a number with a source or a claim without one.
It is distinct from tone of voice: the voice is the personality, the tone its situational expression. The same brand sounds more relaxed in a social post than in an invoice email - but the character behind it stays recognisable. And unlike brand identity, voice is purely verbal: you should be able to recognise it without logo or colour scheme.
Since language models started co-writing, voice has moved from feeling to specification: attributes with write-this/not-that example pairs, a banned-words list, firm formatting rules. Documented that way, two different authors hit the same sound - and an AI draft can be checked against the definition objectively instead of by gut feeling.
Why does Brand Voice matter?
Roughly half of new articles on the web are now AI-generated and, left unsteered, sound alike. In the Klaviyo/Datalily survey (8,000 consumers, December 2025) only 7 per cent said visibly AI-generated content increases their trust in a brand - 31 per cent said it decreases it. A recognisable voice is a measurable trust and differentiation lever.
Brand Voice in practice
- 01Innocent Drinks: playful, slightly silly, very British - the same voice from bottle label to tweet, much copied and rarely matched.
- 02A B2B provider defines three attributes with one write-this/not-that pair each - afterwards sales emails, support replies and the website sound like one company for the first time.
- 03AI workflow: the voice definition travels into every content tool as a system prompt - drafts arrive in the brand’s register instead of the model’s default tone.
From the journal
- The tone of a rejection: the message your business sends most oftenEvery business says no far more often than yes. Yet there is a template for the yes and none for the no. 64 per cent of job seekers never get a final answer, and in the trades up to two thirds of enquiries go unanswered. What belongs in a good no, how fast it has to go out, and where the friendly sentence turns into a legal problem.
- Content recycling with AI: twelve assets from one piece, without the filler95 per cent of B2B marketers use AI, according to the Content Marketing Institute, and 87 per cent got more productive. Only 39 per cent saw their content perform better. Why that gap exists, and how to turn one researched piece into many formats without diluting it.
- Brand consistency with AI content: style guides machines can followA language model does not read a brand manual. And it forgets rules measurably: once three constraints apply at the same time, accuracy drops by more than 40 per cent. What that means for your style guide, where machine-readable brand rules are already standard, and what the law adds from 2 August 2026.
