> For the complete documentation index, see [llms.txt](https://help.newtone.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.newtone.ai/documentation/core-concepts/brand-guidelines.md). # Brand guidelines The encoded form of your brand. Generation guidelines govern content written from product data; translation guidelines govern content carried between languages. Your guidelines are the encoded version of how your brand writes. They are not a document the platform reads for inspiration — they are structured rules, terms and examples applied on every run, in every language. This is what makes output publishable rather than draft-quality, and it is configured before your team starts working. [Open Guidelines in Newtone](https://app2.newtone.ai/configure/guidelines) ## What brand guidelines contain ### Generation Used whenever content is generated from scratch — PDP copy, category and listing content. **Templates** define the shape of an output: which fields are produced, in what order, at what length, with what structure. A template is the skeleton; the rest supplies the voice. **Rules** are the testable statements about how the brand writes. Register, sentence construction, what may and may not be claimed, what must appear for a given category. **Lexicon** is your brand's own vocabulary: the terms you use for materials, techniques, product families and categories, the house terms that are never paraphrased, and the words you refuse to use. **Training samples** are examples of correct output. They carry what rules cannot articulate — rhythm, restraint, the feel of a paragraph that sounds like you. ### Translation Used whenever approved content is carried into another language — PDP translation, decks, documents, web copy. **Rules** govern how translation behaves: how closely to follow the source, how to handle length differences between languages, register in each target market, what must never be adapted. **Glossary** holds source-to-target term pairs per language, plus do-not-translate entries: product and collection names, house terms, trademarks that stay in the source language everywhere. **Training samples** are approved translations that demonstrate the target voice, which is rarely a literal rendering of the source. ## How guidelines are structured Guidelines are scoped rather than global. Three levels, matching the structures defined in your catalog. **Brand.** The top level. An organization can hold several brands, and each holds its own guidelines. Nothing is shared implicitly between them — a house with a main line and a diffusion line keeps two separate voices on purpose. **Content type.** Within a brand, guidelines are defined per content type: PDP copy, category and listing copy, a document type. This is where most of the configuration lives, because a rule that is right for a product description is usually wrong for a category page. **Category.** Optional, and used where a product category needs its own treatment — different mandatory claims, different terminology, a different length. Fragrance and leather goods rarely want the same instructions. ## What encodes well, and what does not Aspirational adjectives do not encode as rules. "Elegant", "aspirational" and "timeless" mean something to your team and nothing as an instruction. Testable statements encode as rules. "Never use exclamation marks." "Never describe a price as low." "Always *maison*, never *brand*." "Second person in English, third person in German." The useful test when reviewing rules: could someone who has never worked with your brand check whether this was followed? What fails that test usually still belongs in the configuration — as a training sample rather than a rule. Voice that resists being written down can still be demonstrated. If your team keeps saying "it just doesn't sound like us," that is a signal to add samples, not to write another adjective. ## How guidelines get built The Newtone team encodes them during the configuration phase, working from your existing brand book, published content you consider correct, examples you consider wrong, and a working session with the people who own the voice. Your team reviews the result rather than builds it. See How a project runs. The wrong examples matter as much as the right ones. Much of what makes a brand recognizable is what it refuses to do. ## Changing guidelines Guidelines are not frozen at go-live. Raise changes through your Newtone [contact.](/contact-us/contact.md) The signal to watch is repetition. A reviewer correcting the same thing across many items is not finding content errors — they are finding a gap in the configuration. Fixing it at the guideline level removes the correction from every future run. See [Quality and review](/documentation/core-concepts/quality-and-review.md). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://help.newtone.ai/documentation/core-concepts/brand-guidelines.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.