Use case · Agencies
Your clients ask what the AIs say about them. Answer with a measure.
One project per client, their canon in their language, their trackers on their market's prompts. You read everything from one account, and every deliverable traces back to dated checks.
Views from the app. Your figures replace the sample at the first survey.
One client, one project, watertight
Each client lives in its own project: its canon, its trackers, its surfaces and corroborations, its logbook. Nothing bleeds from one client to the next, a finished engagement archives without erasing its record, and the account-level views reassemble the whole book of clients when what you are organizing is your own morning.
What you measure
Per-client baselines
Answer rate, tone or share of voice per client, engine by engine: each engagement measured on its own terms.
A cross-client task board
Quests from every project in one view: what each client file asks for next, machine-suggested or posed by you.
Dated proof of work
Every action in the client's logbook faces the curve it moved: reporting quotes dated records.
One prepaid balance
A single balance funds every client's measurement, topped up when you decide: you pay for the checks that run.
The recipe
How to run a client portfolio on Epovest
The full working guide for this use: watertight projects, onboarding in one sitting, the cross-client board, reporting that proves the work, and the API for the repeatable parts.
One project per client, watertight
The project is the unit of an engagement: the client's canon in the client's language, their trackers, their registers, their logbook.
Keep the boundary clean: a client's competitors are keywords in THEIR project, never a project of their own. The canon belongs to the client and speaks the language their market reads. When an engagement closes, archive the project: it stops generating work, keeps its record intact, and a returning client finds their history where it was left.
Onboarding a client, in one sitting
Three gestures make the intake, and the third produces the first deliverable of the engagement: a dated baseline.
- 1 The audit, by surfaces Register the client's pages as surfaces, as they stand: site, profiles, listings. The unchecked boxes are the audit: what the engagement will fix is on the table at the first meeting.
- 2 The canon, from their pages Draft the canon from what the client already publishes, then settle it with them: category, one-liner, facts, perks. One sheet, their language; every later gesture inherits it.
- 3 The baseline tracker Their name and their market's buying prompts, weekly in hd, discovery on. The first checks are the baseline report: what the engines say before the engagement, dated.
The morning board: every client, one view
Quests across all projects, in one list: that is the agency's morning. Each client file states what it needs next.
The file fills itself from the measures: surfaces to realign after a canon revision, watched corroborations to reread, candidates to sort, discoveries to accept. Add your own errands per engagement, a listing to obtain, a page to get corrected, and the board becomes the day's route through the whole book of clients, without opening each project to ask.
Reporting that proves the work
The client's question is always the same: what did you do, and what did it change? The logbook answers in that exact shape.
Date every action in the client's logbook: the publication obtained, the listing corrected, the campaign. The report then reads action by action against the curves, and every figure quotes a dated check the client could go read. Pull the raw series by API or MCP into your own template: the numbers arrive already attributable.
Script the repeatable parts
The gestures of this page repeat for every new client: they are exactly what the API and the MCP tools carry.
Connect your assistant and hand it the intake: create the project, draft the canon from the client's site, register the surfaces, open the baseline tracker. It reads the board across clients, drafts the monthly reports from the logbooks, and flags what changed since last week. Top-ups stay yours: an assistant prepares, the person approves.
Worked example
Ten clients, one morning
The board lists what moved overnight: a watched comparison changed for one client, a canon revision left four surfaces outdated for another, two candidates await sorting for a third. One hour, in one view, orders the whole book: the comparison is reread and its change dated, the surfaces requeued, the candidates recorded. The month's reports will quote each of these gestures with its date, and the next checks will say what they moved.
With your AI assistant
Your AI assistant runs this recipe with you
Connect Epovest to the assistant you already use, one-click connector or API key. It creates trackers, reads the results and keeps the registers with you; a payment only ever happens with your approval.
The instruction to give it
Connect to Epovest and onboard my new client: create the project, draft its canon from [client site], register their pages as surfaces with their checklists, then open a weekly baseline tracker on their name and their market's buying prompts. Every Monday, read the cross-project quest board and summarize it per client.
$0.10
per prompt, per engine, per check
Pricing
Whatever you configure, the unit never changes. Your total shows live in the configurator before anything runs; you start whenever you want, from a $10 top-up. No subscription: only the checks that run are debited.
Engagements evolve, projects follow
A scope grows: add trackers or keywords in place. An engagement ends: archive the project, its record stays readable and it stops asking for work. Reopening it later brings the same history back.
The other use cases
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.