Atlas · Step 4 of the method
Know exactly where to appear
The AIs answer from sources. The Atlas names them: the sites the engines actually cite and read on your market, ranked so you know which one to work on first.
A named list, in priority order
Every survey your trackers run reports the sources the engines cited or read to answer. The Atlas turns them into a map: one entry per domain, its AI Authority engine by engine, its detection count, and the date it was last seen. It reads as a work queue: the top of the list is where a mention pays off most, and the same list read through another engine puts other names on top.
- One entry per domain, with its AI Authority and its detection count on each engine
- Searchable, filterable by engine, sortable on any column
- Readable through the app, the REST API and MCP, so your AI assistant works the list with you
How you work the Atlas
Read the map
Your surveys fill it as they run. Sort by AI Authority, filter to one engine, and the sources that shape the answers on your market come to the top.
Go and appear
Take the list in order: a directory to be listed in, an editor to talk to, a review site to answer on, a reference page to be quoted in. Each entry is a real source, with an address.
Record what you obtain
An article, a listing, a quote: record the corroboration on its source, with its exact URL and its publication date. The source stops being a target and becomes a foothold.
How the two connect
It all starts with Tracking
Your Atlas is built by your own measurements. Each survey your trackers run reports the sources the engine cited or read to answer, and those sources are what the map is made of: it describes your market, in the terms of the engines that answer on it.
-
1
A survey asks
Your prompts go to the engines, word for word, at your cadence.
-
2
The answer names its sources
Every full answer is archived with the pages the engine cited, and those it read to get there.
-
3
Those sources become your map
Each domain enters the Atlas with its AI Authority and its detection count, and stays there as a series.
Which is why the Atlas grows with what you measure: adding a prompt or an engine to a tracker puts new sources on your map, and each one is somewhere you can go and appear.
The metric
AI Authority: the number that ranks the queue
A list of sources without an order is a list of chores. Each entry carries its AI Authority, from 0 to 100, where 100 is the source the engine leans on most. Read top down, the map becomes a plan: the effort you spend on the first entries carries further than the same effort spent lower.
Computed over every source Epovest knows
The 100 that sets the scale is the leading source of the entire corpus, all the surveys running on the platform, and not the top of your own list. So an AI Authority means the same thing on your map as on anyone else's, and a source that reaches 61 keeps its 61 as your own coverage widens.
- A rolling 30-day window
- The number reads the present: the detections of the last 30 days, recalculated every day. What an engine leaned on last spring stays in the history, and the ranking stays about now.
- One AA per engine
- Each engine has its own leading source, so its own scale. Filter to one and the order changes: the same market has as many maps as there are engines, and that is exactly the information you act on.
- A number you can act on
- AI Authority is a count of detections put on a scale, so an entry that climbs climbs because the engines cited it more. Next to it, the raw count says how many detections the score stands for.
Corroborations
Record what others publish about you
You obtain an article, a listing, a review, a quote in a guide. Record it on its source in the Atlas: the exact URL, the brand it speaks about, the date it went live. What you obtained stops living in an inbox and starts living on the map.
The exact address
A corroboration carries the URL of the page that carries it, not just a domain. It is rereadable, verifiable, and shareable with your client or your agency.
The date, on your curves
A corroboration is a dated move: it lands as an annotation on your citation curves, so you read what an obtained mention was followed by.
A life of its own
A page that comes down is archived rather than erased: "they spoke about us from March to July" is information, and your curves keep the annotation.
Official pages and corroborations are the two halves of the same story: a surface is your own address, a corroboration is a source where someone else spoke about you. The map shows both, side by side.
The tool the others open ground for
Tracking measures what the engines answer. Canon and Surfaces act on what you own outright. The Atlas is where you act on what you do not own, and everything the suite records converges on it.
Its place in the method
The Atlas is step 4: your canon says what should be repeated, your official pages carry it, and the Atlas says who else should be repeating it. It is also the only tool that works OUTSIDE what you own, which is why it is the one that opens ground.
Prepaid credits, no subscription: you pay for the checks that run. The whole grid is public.
See pricingCommon questions
Where should my business appear to be recommended by AI assistants?
In the sources the engines actually lean on when they answer your market's questions, which are rarely the ones you would guess. The Epovest Atlas names them: it collects the sources cited and read on every survey your trackers run, and ranks them by AI Authority so the first line of the list is the source where appearing carries furthest.
How do I find out which websites AI engines cite about my market?
You measure, and the list builds itself. Each survey asks your prompts to the engines and reports the sources they cited or read to answer; those domains become entries in your Atlas, with their AI Authority and their detection count. The map is the by-product of the measurement, so it describes your market rather than a generic list.
Which sources should I work on first?
Sort by AI Authority and start at the top, engine by engine. A source at 100 is the one that engine leans on most; the effort you spend there carries further than the same effort lower down. Filter to the engine that matters most to you, because each engine has its own ranking, and the top of one is rarely the top of another.
How do I track the mentions and articles I obtain for AI visibility?
Record each one as a corroboration on its source in the Atlas: the exact URL, the brand it speaks about, the date it was published. Each corroboration is pinned to its source on the map and lands as a dated annotation on your citation curves, so an obtained article stops being a line in a report and becomes a point you can read your measures against.
What is the difference between a surface and a corroboration?
A surface is a page where you speak in your own name: your site, your profiles, your listings, the owned half of your story. A corroboration is a page a third party published about you: an article, a review, a mention in a guide, the earned half. Both are managed in Epovest, and the Atlas is where the earned half lives, on the map of the sources the engines read.
Can I read the source list through an API?
Yes. Everything the Atlas shows is served by the REST API v1 and by the MCP server, with the same keys: the sources, their AI Authority per engine, their detection counts, and the corroborations you record. An AI assistant connected to your account can work through the list with you and record what you obtain as you go.
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.