Copied facts
The warranty, the price, the battery life: written in ten places, updated in three.
Brand context
What is true about your brand, written once.
You state who you are, what you sell, how you speak and which facts are approved. Proposals, the assistant, the public profile and the AI answer checks read the same version, and every output records which one it read.
The problem
Every tool invents its own brand.
A prompt here, a document there, a number copied by hand: every text produced starts from a different version of who you are. And AI answers are judged without a common yardstick.
Inconsistent tone
Every draft sounds different, and the words to avoid keep coming back.
No yardstick
To say an AI answer is wrong you need to know what is true. In writing.
What it contains
Six cards, the facts, the tone, the documents.
The context collects what anyone writing for your brand needs, with a reference for every sentence: an approved fact, a document or the gap that makes it incomplete.
Company
Who you are, where you operate, what you do.
Audience
Who you speak to, product by product.
Offer and products
Lines, features, terms.
Competitors
The same as in monitoring, with the topic where each one is strongest.
Positioning
Why choose you, in a sentence that can be verified.
Strategy and priorities
What matters in this period.
Approved facts
Each fact with its source, who approved it and a review date.
Tone and style
Writing rules, words to use, words to avoid.
How it works
One version, many outputs.
Every change creates a new version of the context. Outputs already produced stay tied to the version they read: none rewrites itself, and none composes the brand on its own.
| Output | What it reads | How |
|---|---|---|
| AI answer accuracy | Approved, unexpired facts | Always the current version |
| Intervention proposals | The full context | Records the version read |
| Ask TrustInsider | The full context | Records the version read |
| Public profile | The facts you choose to publish | Always the current version |
| Workflows | The full context | Records the version read |
| API and MCP | Approved, unexpired facts | Records the version read |
The gaps
Where the context falls short, we ask you.
If AI answers give different prices for a product and no data is approved, nobody can say which is wrong. We flag it as a gap: what is missing, in which card, how many answers it involves. Answer one question at a time; the answer counts only once a person approves it.
The safeguards
Nothing goes in without a person.
The context guides analyses and proposals, so it must stay yours: declared, approved, versioned. Changing a fact doesn’t rewrite past observations.
The tone judge
A draft with a word to avoid doesn’t reach approval: it comes back with the reason.
Honest comparison
Facts and answers are compared on the words: a paraphrase stays “not verifiable”, never counted as an error.
Reference documents
PDF, text, Markdown and CSV up to 2 MB, with no formats that contain macros or code.
Only what you choose
Internal conversations, email and chat don’t enter the context.
Frequently asked questions
Questions about the brand context.
How a sheet entry differs from an approved fact, when an uploaded document enters the context, what happens when you change a fact and which files we accept.
What is the difference between a card entry and an approved fact?
A fact has a source, who approved it and a review date, and it is used to judge the accuracy of AI answers. A card entry guides whoever writes, but it doesn’t judge any answer.
Does an uploaded document enter the context immediately?
No. A document is a reference: nothing enters the context without a person’s approval.
What happens when I change a fact?
A new version of the context is created. Outputs already produced stay tied to the version they had read; past observations are not rewritten.
Which documents can I upload?
PDF, plain text, Markdown and CSV, up to 2 MB each. No formats that contain macros or code.
Related paths
After “Brand context”, the next step.
Each feature answers a specific question. From here you move on to related features, to the methodology behind their measures and to help for using them in everyday work.
Explore every intelligence tool
AI visibility, AI Trust Score, search, reputation, competitors, actions and execution: a feature for every question about your brand.
ReadFrom question to action
How TrustInsider works in six steps: brand, questions, quote, evidence, page-by-page intervention and a check on the same questions.
ReadA transparent method, not a black box
How every TrustInsider measure is built: sampling of AI answers, sources and rights, index calculation and declared limits.
ReadFind the right answer
Help with access, credits, collections, integrations and data: guides, FAQs, the site assistant and support requests from your workspace.
Read