FAQ

Questions, answered plainly.

What agencies ask before putting AI in front of clients — with the mechanism behind each answer, not adjectives.

What agencies ask before putting AI in front of clients — with the mechanism behind each answer, not adjectives.
On this page

What Tempera is

What is Tempera?
The operating system for marketing and communications agencies: clients, brands, campaigns, and activities in one governed workflow, a dedicated AI agent per client, and a human review gate before anything is client-ready.
Who is it for?
Agencies first — from a single studio to a multi-agency network, an installation runs under your own name and logo, and your clients' reviewers ride along on free seats. The same governed workflow also serves independent consultants (the Solo plan) and in-house brand teams (a flat monthly plan): a workspace mode reweights the vocabulary and defaults, and the review gate stays.
Is it another AI content tool?
No. Content tools generate; Tempera runs the job — a structured brief in, options grounded in that client's own history, a named sign-off, and delivery as real files: Docs, Slides, PDF, XLSX, PSD.

Explore the platform

The AI

Isn't AI marketing content just slop?
Left ungoverned, mostly yes — generic prompts into generic models produce work no client would pay for. Tempera is built against that: each client's agent learns from what that client approved and rejected, and a human verdict stands between every draft and the client. Work either passes the gate or it doesn't ship.
Which AI models does it use?
The ones you choose — tasks route across 17 direct providers on your own keys, per client, per agent, or per task. When a better model ships, you point the routing at it. Nothing is welded to one vendor.
How do you keep AI costs under control?
By engineering, not by hoping: each task runs on a right-sized model, the prompt carries only that client's relevant context, and every call is metered to the client, campaign, user, and model — with monthly budget caps that stop spend before it happens.
Does AI work go straight to our clients?
On a team, no. The moment a second person is active, whoever ran the work cannot approve it — the server rejects self-approval — and exactly one accountable owner signs off before anything is client-ready. Working solo, the review step still runs and your own verdict is recorded and audited.

See the controls

Your data

Do we have to move our files into Tempera?
No. Working files stay in storage you point at — your own disk, bucket, drive, or SFTP — in a readable client/brand/campaign layout your other tools can reach. The database stores references, never the bytes.
Do you track visitors on this website?
No. This site runs no analytics, no advertising pixels, and no cross-site tracking — the same restraint the platform brings to client data.

See the controls

Pricing

How does pricing work?
On agency plans, you pay a platform fee for each client with at least one AI task that month — dormant clients cost nothing and seats are unlimited. AI usage is metered separately, or bring your own provider keys and pay the provider directly.
What happens if we stop paying?
After a grace period the installation goes read-only — viewing and exporting always keep working. Non-payment never locks you away from your own data.
Can we evaluate it without keys or a card?
Yes. Mock mode runs the entire workflow — briefs, options, the review gate, delivery — with no AI keys and no client data, and the 30-day Studio trial runs the real thing without a card.

See pricing

Running it

Can we self-host?
Yes — hosted or self-hosted is the same product at the same price. Our cloud is simply an installation where we are the operator; yours runs on your infrastructure, with an offline option for the strictest environments.
Is there a mobile app?
iOS and Android. The review gate in your pocket: your waiting-on-you queue, approve or send back with notes, offline browsing, and QR pairing in seconds — with free seats for your clients' reviewers.
Is there an API?
Yes — a versioned public API with scoped keys and OpenAPI references, so portals and dashboards plug into the same governed core.
Which languages does it speak?
English and French ship built-in across the product, this site, and the mobile app; more languages can be activated on a running installation without a redeploy.

Explore the platform

Prove it on one client

See one client move from context to approved work.

Run a real brief through Tempera in mock mode. Follow it from grounded options to a named human verdict and client-ready work.

No AI keys. No card. Nothing reaches a client without review.