--- title: "What Are Machine Readable Brands?" description: "Machine Readable Brands: brands whose identity, language, and design logic exist as a specification that people and systems can work with. Definition, three stages, boundaries." author: "Robert Haase" datePublished: 2026-07-19 dateModified: 2026-09-05 inLanguage: en url: https://robert-haase.de/en/machine-readable-brands.html translation: https://robert-haase.de/machine-readable-brands.html.md keywords: ["Machine Readable Brands", "Machine-readable brand", "Brand Infrastructure", "Brand Specification", "Agentic Brand Readiness", "AI Search", "Agents", "Brand Strategy"] --- # What Are Machine Readable Brands? ## Definition Machine Readable Brands are brands whose identity, language, and design logic exist as a specification that people and systems can work with equally. **Specification** means rules that can be applied, not examples that must be interpreted. **Equally** means the same document for the agency and for the model, not two versions sharing a name. **A stage of maturity, not a property.** Three stages lead there: being read, that is, findable at all in AI search and by agents · producing with AI, that is, using the brand as a given rather than a matter of interpretation · becoming applicable, that is, providing the brand as a logic systems work with directly. Only the third stage is meant. **Distinctions.** Not the same as structured data, which marks up individual facts but says nothing about stance and limits. Not the same as a design system, which governs execution and leaves meaning open. Machine Readable Brands answer whether a system can *apply* the brand without interpreting it. They are the counterpart to [Agent Authority](https://robert-haase.de/en/agent-authority.html): one makes the brand legible outward, the other binding inward. **Why the question exists.** Across 135 home pages of German listed companies, 456 of 13,527 controls carry no name, and [52 of those pages have no main-content landmark](https://robert-haase.de/en/evidence.html#dax-landmarken). The figures, with their limits, are in the [evidence collection](https://robert-haase.de/en/evidence.html#dax-benennung). Used and defined in this sense by Robert Haase since March 2026. Related terms: [Brand Specification](https://robert-haase.de/en/glossary.html#brand-specification) · [Agentic Brand Readiness](https://robert-haase.de/en/agentic-brand-readiness.html). ## The Second Audience Brands have acquired a second audience. AI search summarizes them, agents compare and recommend them, teams produce in their name with AI tools. All three work only with what they can process. A brand book describes a brand. A [specification](https://robert-haase.de/en/glossary.html#brand-specification) can be applied. That difference decides whether a brand exists in systems or only in presentations. One example: the brand book says "We communicate at eye level, approachable and clear." Three readers, three readings. The specification says: "Address the reader as you, never as the customer. At most one exclamation mark per text. Technical terms are explained on first use." A person can apply that, and so can a machine. [Machine View](https://robert-haase.de/en/machine-view.html) shows it live: the same page side by side, once for people and once for machines. ## The Three Stages **A brand becomes machine readable in three stages. The first two are groundwork.** → **Stage one: being read.** AI search and agents find the brand, summarize it correctly, and can recommend it, for instance when ChatGPT describes it on request. Whether a brand reaches this stage is measurable: [Agentic Brand Readiness](https://robert-haase.de/en/agentic-brand-readiness.html) is the maturity level for it, self-test included. → **Stage two: producing with AI.** Teams generate on-brand output, an ad built in Canva for instance, because the brand exists as a specification, not as text everyone reads differently. Without this stage, AI production sounds like no one or like everyone. → **Stage three: becoming applicable.** The brand becomes a logic that systems work with directly: language, claims, facts, approvals, boundaries. The agent answering a customer request in the brand's name works at this stage. Brands here are Machine Readable Brands. ## What Can Be Structured, and What Cannot **Not everything about a brand belongs in a specification.** Meaning, attitude, and trust remain what they are: human, meaning told and experienced. What can be structured are the parts that must be applied repeatedly and consistently: the positioning in one sentence, the language rules, the verifiable facts, the design system, the boundaries of what may be promised in the brand's name. Who defines what an agent may promise, and who stands behind it, is covered in [Agent Authority](https://robert-haase.de/en/agent-authority.html). ## Boundaries **Three common mix-ups.** **SEO and [GEO](https://robert-haase.de/en/glossary.html#geo)** optimize content for findability. Machine Readable Brands start earlier: with the brand itself as an applicable structure. Findability is a consequence, not the goal. **The brand book** describes the brand for people, usually as a PDF; it can be read but not applied. **The design system** structures the execution layer, that is, how the brand looks: components, colors, typography. At IBM, Porsche, and Deutsche Bahn it is public and versioned. How widespread that is overall has not been measured. What is missing is the same structure for the layer of meaning, that is, who the brand is: positioning, language, decision logic. The closest example for that comes from government: the Swiss Federal Chancellery provides its writing rules in a form language models can apply directly, described in [Open Brand Repo](https://robert-haase.de/en/open-brand-repo.html). A brand that runs its layer of meaning fully as a specification is not known to me. That is exactly where Machine Readable Brands begin. ## Frequently Asked Questions **Where does the definition come from?** From me. I have used the term in this sense since March 2026: as the third stage of a maturity model. **How does a brand start?** First test the current state, for example with the [Agentic Brand Readiness self-test](https://robert-haase.de/en/agentic-brand-readiness.html). Then formulate the brand as a specification: who it is, what it stands for, how it speaks, which facts hold. And make that definition machine-readable and maintained, via structured data and versioned documents. **Are there examples?** For the execution layer, many; for the layer of meaning, none yet from business. Both cases are covered above under Boundaries. --- *This file is generated from the page: https://robert-haase.de/en/machine-readable-brands.html. If the two ever differ, the page is authoritative.*