Appearance
Voice, messages and naming
Answered, 5 October: all three verbal-identity recommendations accepted. These are parts of an identity, not new content types. A writing guide, a product page and a video keep their own forms.
The recorded decisions preserve your approval and its limits. No further answer is needed on these three choices. The examples and alternatives below remain the record of what you reviewed.
| Choice | Accepted recommendation |
|---|---|
| Voice and tone | Keep Brand Voice distinct from Tone Variations within verbal identity. |
| Messages and taglines | Keep Key Messages distinct from Tagline System, with source wording and scope preserved. |
| Naming and word usage | Keep Naming System distinct from Terminology. |
What is already settled
An identity program can include verbal and strategic work. A source's stated strategy remains distinguishable from our interpretation. One piece can support several research paths without becoming several duplicate pieces. These rules stay settled.
Your earlier request was to extract and review the identity vocabulary. It did not approve every extracted row. This page separates three boundaries from that older bundled question. Purpose, values and promises are also answered. The remaining strategy list and audio voice still need their separate review.
Calls to action and offers already have a decision. A requested action, price, benefit and terms remain searchable with their wording, subject and source location. They can occur in an image, page or video, with or without an end card. Offer is not a whole-piece type. The recorded answer is not being asked again here. Exact vocabulary and implementation still need checking.
1. Voice and tone answer different questions
Accepted recommendation: keep Brand Voice for the stated communication character and principles; keep Tone Variations for how that voice is adapted to a situation. Tone can respond to audience, channel, purpose or the reader's circumstances. Neither label requires a brand to publish formal guidance.
Checked example: Mailchimp's voice and tone guide distinguishes a relatively consistent voice from tone that changes with context and the reader's state of mind. It explains why the same humor is not suitable for every situation.
Counterexample: a serious service announcement does not prove that the brand has abandoned a generally playful voice. A cheerful video is evidence of that piece's tone; it does not, alone, establish the company's stated communication principles. Voice here also does not mean the narrator's vocal performance.
Alternative: keep one Voice and Tone description with the distinction explained inside it. That is simpler to maintain, but makes it harder to compare a brand's general guidance with its adaptations. I recommend two meanings within one verbal-identity area; this does not require two separate screens.
What changes downstream: let someone move from stated guidance to supported examples and see which situation each example belongs to. Analysis must identify whether it is reading a rule or describing a piece. A product can have its own supported guidance; it should not inherit every company attribute automatically. Changes over time need dated evidence, not an assumption that voice can never change.
2. A message is not the same as its exact wording
Accepted recommendation: keep Key Messages for the main ideas a source says it intends to communicate, and Tagline System for recurring short expressions and their usage rules. A message may have several expressions. A tagline may serve a company, product or campaign. Preserve that supported scope rather than treating every memorable sentence as a company-wide tagline.
Checked example: NC State calls Think and Do its brand mantra and gives rules for its wording. Its guidance distinguishes the phrase from the broader ideas it expresses, and permits it to serve as a tagline without requiring that use. This is a useful boundary case: preserve the source's term and supported role rather than automatically treating mantra and tagline as synonyms.
Counterexample: an analyst's summary of a video's message is not a quotation or an approved brand message. A one-off headline is not automatically a tagline system. A button asking someone to register can remain a call to action; calls to action and offers remain distinct under the approved decision.
Alternative: combine all of these under Messaging, keeping wording and usage in descriptions. That is compact, but can mix what is being said with the exact expression someone wants to study. I recommend the distinction so a researcher can compare message themes and then inspect the actual lines used.
What changes downstream: retain exact text separately from paraphrases, with language, source and applicable company, product or campaign when known. Multiple pieces may express the same supported message; one piece may express several. A match in wording alone does not prove common campaign membership. Search and Ask should reveal the evidence behind the interpretation.
3. Naming rules are not just a glossary
Accepted recommendation: keep Naming System for rules used to form and organize names; keep Terminology for preferred words, meanings, spelling and usage. These can live in one guide while answering different research questions.
Checked examples: Red Hat's brand architecture guidance describes different naming treatment for products and components. This supports a naming-system example, not a universal structure for every company. Mailchimp's word list instead gives usage and spelling guidance, including the distinction between the noun “signup” and the verb “sign up.” That is terminology guidance, not a rule for creating product families.
Counterexample: a list of product names does not by itself establish the rules that created them. Shared words do not prove that two products are versions, siblings or part of a bundle. A glossary entry about an industry term does not create a product or feature.
Alternative: combine both under Language Guidelines and keep any distinctions in prose. I recommend separate meanings because people may research naming patterns without wanting general copyediting rules, or inspect preferred vocabulary without studying product architecture.
What changes downstream: preserve actual names, supported aliases and naming guidance separately. Connect a rule to the products, services or features it demonstrably covers; do not manufacture those relationships from a prefix. Captures and later reanalysis should preserve the names and rules that were valid in the source's context. Exact storage and compatibility remain audit work.
What this does to the existing library
The selected live data and local code already use voice, tone, tagline and message-related descriptions. That does not prove they carry the approved distinctions or reliable source evidence. No existing record has been relabeled. The evidence note separates observed storage and local consumers from the complete audit still required.
Your reply accepts these three boundaries. Remaining identity and strategy rows stay open where not already decided. This review does not approve a schema change, new extraction prompt or automatic publication.
Return to the morning review desk for the other independent packets.
Prepared 5 October 2026.
The review date below records a check of the current choices against the recorded answers and linked text examples. It does not certify the complete database or pipeline.
Last reviewed 5 October 2026.