Twenty brands, one operator, and nothing bleeding between them.
The hard part of running content for many clients is not writing. It is holding twenty different sets of brand rules, twenty competitor sets and twenty publishing targets in one head without any of them contaminating another.
Multi-client is where content tools quietly fall apart.
Most tools are built for one brand and grown into many. The tell is a shared settings screen, a single tone-of-voice field, or a keyword list that belongs to the account rather than to a client. It works until an account manager pastes the wrong brief into the wrong workspace, and then it is a phone call.
The second problem is proof. An agency has to show a client what was done and what it earned, and reconstructing that from three tools and a spreadsheet at the end of a quarter is a day nobody has.
Separation has to be structural, and evidence has to accumulate on its own.
The website is the unit of everything.
- 1
A workspace per client
Brand, audience, competitors, prohibited terms, WordPress, Google properties and queue. Switching client switches all of it at once.
- 2
Separation enforced in the database
Row level security, not interface filtering. A build-time probe attempts a cross-tenant write and fails the build if one is ever accepted.
- 3
Roles and approval routing
Invite colleagues or the client. Owners and members differ, and publishing can require an owner's sign-off.
- 4
An activity log that writes itself
Every AI operation, every publish, every approval, per workspace, timestamped. The quarterly review assembles itself.
The things that make a portfolio manageable.
Every line here describes something the software actually does. If one of them turns out not to be true of your setup, that is a bug and we want to hear about it.
One account, one login, twenty separate briefs.
Switch website in the sidebar and every screen changes to that client: queue, calendar, library, analytics, visibility, settings.
A client's brand rules cannot reach another client's drafts.
Prohibited terminology, tone and claims are per workspace and are enforced at generation and again at publish.
Each client gets their own competitive picture.
The AI visibility sweep runs per website against that client's named competitors, so the report you take into a meeting is about them.
Approval can be required, per website.
Three positions from suggest-only to publish-on-schedule. A cautious client and an autonomous one can sit in the same account.
The evidence is there without anybody assembling it.
What was written, what it scored, when it published, and what it earned at thirty, sixty and ninety days.
Being straight with you
- There is no white-label mode. The product is Docket and a client invited into a workspace sees Docket.
- There is no client-facing report builder yet. The data is on the screens and in the activity log; assembling a branded PDF is still your job.
- Billing is not implemented, so there is no per-client invoicing, usage metering or seat management to speak of.
- Publishing is WordPress only, which for most agency portfolios covers most clients and will not cover all of them.
Answers, Not Hedging.
The parts either side of this one.
Bring one client and we will run it live.
Pick the account you are least happy with. We will run the visibility sweep and the opportunity scoring against their domain before the call and show you what comes back, which is usually a better conversation than a tour of the software.