For agencies

A marketing agent built for agencies, not for one account

Every client in its own workspace, every ad account under one seat, every change approved by a person and logged for the client.

Powered by Anthropic's Claude

Free to install · No card · v

Northline Studio › All clients › Overview Synced 2m
Clients Last 7 days Add client
OverviewApprovals · 5ReportsKnowledge
ClientPlatformsSpendPurch.CPAvs last wk
Harlow & FinchMeta · Google£6,120241£25.39−8%
Ridgeway CoffeeMeta · Google · TikTok · Pinterest£7,960478£16.65−14%
Verve FitnessGoogle£3,410102£33.43+21%
Otto MenswearMeta · TikTok · Pinterest£4,280199£21.51+2%
Blue LanternLinkedIn · Google£2,90037£78.38−5%
Pale Fox WinesMeta · Pinterest£1,76088£20.000%

Just ask

Ask across the whole roster, or one client at a time.

Each client's data, knowledge and history stay in that client's workspace. The agent respects the boundary.

Which clients need attention this morning? Write every client's weekly report What did we change on Verve last week, and who approved it? Which client has the most budget headroom? Onboard Pale Fox: what should I know before proposing anything? Show me every approval pending across all clients Which accounts are behind pace? Compare CPA across all clients on Meta

How it works

How an agency runs on it

Step 1

One workspace per client

Create a client, connect its ad accounts and — for e-commerce clients — its Shopify catalogue. Connections are scoped to that client, so an agent working on one client can never see or touch another's data.

Step 2

Teach it the client once

Knowledge cards hold what a new account manager would need to know: target CPA, campaigns that must never be paused, banned claims, the landing pages in play. The agent reads them before every proposal, and reports say which cards were applied.

Step 3

Approve, and hand the client the trail

Every applied change records who approved it and what the platform returned. The weekly report lists every action approved in the period alongside the results, so the client sees the work as well as the numbers.

Approvals

One queue for every client

Approvals from three different clients, side by side. Each is scoped to its own workspace and its own constraints.

Needs approval 4m ago

Verve Fitness · pause the broad-match ad group

The broad-match ad group spent £5,410 in 14 days for 8 conversions. The exact-match ad group on the same terms converts at £24. Knowledge card applied: target CPA £30.

Ad group · Broad termsPAUSED
Ad group · Exact terms£40 → £70/day
Recovers £4,900 / month
Approve both Decline
Ready 18m ago

Ridgeway Coffee · weekly report drafted

Spend £7,960, purchases +14% week on week, seven actions approved. Two alerts: Meta retargeting CPA and pacing at 111%.

Summary + platform splitDRAFTED
7 actions approvedLISTED
2 alertsFLAGGED
Ready in Reports · Ridgeway Coffee
Export PDF Decline
Needs approval 1h ago

Blue Lantern · pause Trade Buyers — EU

£840 spent in six days with no leads. Knowledge card applied: LinkedIn lead target £60. The UK campaign on the same offer is converting at £52.

Trade Buyers — EUPAUSED
Trade Buyers — UK£60 → £90/day
Stops £4,200 / month
Approve both Decline

Why agencies need a different shape of tool

An in-house marketer has one account and one boss. An agency has twenty accounts, twenty sets of constraints and twenty people to justify every change to. Adease is organised around that: an organisation with clients inside it, each client with its own connected accounts, knowledge cards, sessions and reports. Team members hold seats in the organisation and work across clients; the agent works inside one client at a time.

The audit trail is the client relationship

When a client asks why the budget moved on Tuesday, the answer is in Adease: the proposal, the reasoning, who approved it, when, and what Meta or Google returned. The weekly report carries the same list. That record is what lets an agency use an AI agent on client money without the client feeling that a machine is spending it — a person always is.

Priced per person, not per account

Agencies are trained to expect a fee per ad account or a percentage of spend. Adease charges per seat — $50 a month for Pro, $200 for Max with five times the agent allowance — and every seat has unlimited clients, ad accounts and campaigns. Winning a new client does not raise the software bill.

Where it fits

A good fit if you

  • run paid media for several clients and want one place to see all of it
  • need a human decision before any spend changes, with a record of who made it
  • are tired of paying per ad account or per pound of spend
  • need to show clients exactly what changed, why, and who signed it off

Not a fit if you

  • want an agent that acts on its own without review — Adease will not do that
  • work on Linux, or need a dashboard a distributed team opens in a browser
  • need client-facing logins — clients receive reports, they do not log in

Questions people ask

Can a client see their own workspace?

Not today. Clients receive the reports you generate — as PDF or edited text — rather than logging in. Seats are for the agency's own team, who can work across every client in the organisation.

Is client data kept separate?

Yes. Ad account connections, synced campaign data, knowledge cards, conversations and reports are all scoped to the client they belong to. The agent operates inside one client's workspace at a time and cannot read another's.

Can it change my campaigns without asking?

No. There is no autonomous write path to a live ad account. Every change the agent proposes is queued in Approvals as a diff and applied only once a person approves it, inside the spend guardrails set for that workspace. Rejecting a proposal sends nothing to the platform.

What does it cost?

Adease is free to install and needs no card. The Pro plan is $50 per seat per month with unlimited ad accounts, clients and campaigns. Max is $200 per seat per month with five times the agent token allowance. You pay per person, never per account or per pound of spend.

Which AI model is behind it?

Adease is built on Anthropic's Claude models. The agent that reads your campaign data, investigates performance and drafts changes runs on Claude. Adease is an independent product made by Adease LTD.

Free to install

Bring one client in this week.

Connect their accounts, write three knowledge cards, and ask what is wrong.

  • No card required
  • Nothing changes without your approval
  • Signed and notarised builds
  • Cancel any time

v · macOS 12+ and Windows 10+