Canon · Step 1 of the method
Decide what the AIs should say about you
When the AIs describe your business, do they say what you would say? The canon is where you decide it: one reference text, written once, reused everywhere as is.
One text, one voice, everywhere
The canon carries your one-liner, a short and a long description, and your category label, written in one language per project. It is the wording every publication reuses word for word: your site, your profiles, your listings, and everyone who speaks for you. The AIs read one consistent story instead of five variations.
- Four fields: one-liner, short, long, category
- Versioned: every change is a numbered, dated, immutable version
- Readable by your team, your agency and your AI assistant, through the app, the API and MCP
How you set your canon
Write it once
Decide what the AIs should say: your one-liner, your short and long descriptions, your category. Your words, one language per project.
It becomes version 1
The canon is recorded as an immutable version. It is meant to stay stable: revising it is a rare, deliberate decision, and the next version is numbered and dated.
Everything reuses it
Surfaces align to the current version, mentions reuse it as is, and the logbook records every version change in front of your curves.
Its place in the method
The canon is step 1: everything that follows measures against it. Your baseline tells you how far today's answers are from it, your surfaces align to it, and corroborations repeat it.
Prepaid credits, no subscription: you pay for the checks that run. The whole grid is public.
See pricingCommon questions
What is a brand canon?
A brand canon is the reference text of a business: what the AIs and everyone who mentions it should say, decided by the business itself. In Epovest it has four fields, one-liner, short, long and category, written in one language per project, and it is versioned: every change is a numbered, dated version that the rest of the suite refers to.
How do I correct what ChatGPT says about my company?
Start by deciding what it should say: that is your canon. Then align every surface you own with it, and get it repeated by the sources ChatGPT cites on your market. Your trackers measure whether the answers move. There is no switch to flip inside ChatGPT: you correct the sources it reads, and the canon is the text they should all agree on.
Why is the canon versioned?
Because it is quoted for years, on surfaces you do not fully control. Stability is the default; when a change is needed, it is a numbered, dated decision, and the registry shows which surface still carries an old version, so re-propagation becomes a checklist instead of a guess.
What belongs in the canon, and what never does?
What lasts: who you are, what you do, for whom, in your words. Never what moves: prices, feature lists, engine names, team sizes. A canon made of invariants survives your next release; a canon full of specifics is outdated a month later.
Make AIs recommend you
Create a tracker in minutes: your prompts, your engines, your cadence. First checks within 24 to 48 hours.
No subscription. Credits from $10, valid 12 months. Pause anytime: the history stays yours.