Licences and roles
Why you see the screens you see - how packages, accounts, environments and roles decide what is available.
Two things decide what you can do in the console. The licence decides which features your organisation has bought. Your roles decide which of those features you can see and change. A screen is in your navigation only if both allow it.
Accounts and environments
Your organisation has an account. The account owns one or more environments, also called tenants: separate spaces with their own profiles, events, configuration, team and API keys. A typical account has a production environment and one or more sandboxes. Environments are isolated from each other. (The console's Workspace settings are an environment's own settings.)
- The licence belongs to the account, not to an environment.
- Some roles apply to the whole account (
account_admin); others apply inside one environment. - The tenant switcher in the top bar moves you between the environments you have access to. A Sends suppressed badge marks an environment whose outbound mode is off or sink, so it sends nothing to real people; a Sandbox or Development badge appears instead when the backend does not report the mode.
The licence
A licence is feature packages plus a usage band. Packages decide what exists; the band decides how much is billed.
| Package | Adds | Screens it unlocks (examples) |
|---|---|---|
| Core (always) | Profiles, events, schema, segments, identity, signals, computed attributes, data contracts, custom datasets, analytics, dashboards, metrics, privacy and consent, preference centre, delivery receipts, file import and export, notifications, the content library, tracking, connected AI apps | Profiles, Events, Custom datasets, Tracking, Analytics, Dashboards, Metrics, Signals, Segments, Schema, Identity, Privacy, Notifications, Content, Usage |
| Campaign | Campaigns, journeys, message templates and documents, delivery channels, fatigue policies, silent hours, response policies, control groups, coupon codes, attribution, languages, split tests, pre-flight checks, external systems, approvals | Campaigns, Message templates, Documents, Channels, Contact policies, Coupon codes, Attribution, Languages, External systems, Change log |
| Decision (needs Campaign) | Next-best-action, shared offer catalogues, item recommendations | Offer catalogues, Recommendations |
| Activation (needs Campaign) | Audience sync to advertising platforms | Activation |
| Connect | Streaming, Kafka, change data capture, data connectors, warehouse queries, private networks | Data connectors, Private networks |
| Digital | On-site content, banners, landing pages and web forms, in-app messages | On-site: Digital content, Forms |
| AI assistants | Plain-language audiences, copy and image generation, journey drafting, performance prediction and insights, your own AI models | AI models, plus assist buttons throughout the console |
| Agents (needs AI assistants) | Background AI workers, with the features Custom agents and Built-in agents | Agents |
Custom agents and Built-in agents are independent features of the Agents package: an account can have either or both. Built-in agents that read campaign data appear only when the licence also includes Campaign.
A contract can leave out individual features inside a package, for example Campaign without journeys. The core package is never narrowed this way.
If a screen or button described in these pages is missing for you, the usual reason is that the package is not part of your account's licence. The console hides what you cannot use as a courtesy, and the platform refuses the action in any case. Ask your account administrator.
Usage
The Usage screen under Settings → Workspace shows what your account is metered at against the band it contracted:
- Billable profiles: real people with a durable identifier such as an email address, a phone number, a CRM id or a loyalty number, counted after duplicates are merged and active within a recent window. Anonymous visitors and cookies do not count.
- Stored events and Operations.
Nothing blocks when you go over a band. You get a warning and an alert email, and an invoice line only if it persists. Sandbox and development environments are shown but marked Not counted. The headline ratio, profiles per customer, is also a data-quality signal: a figure far above the contracted constant usually means identity resolution is not merging duplicates, not real growth.
Roles
A role is a named set of permissions. You can hold several, and they add up: each role only adds access, never removes it. Built-in roles follow the pattern <area>_<level>, so each area has a viewer, an editor and sometimes an approver. The console shows built-in roles by a name in your language, for example Campaign viewer for campaign_viewer; the API and these pages use the code.
| Area | Viewer | Editor | Approver or admin |
|---|---|---|---|
| Customer data | data_viewer | data_editor | |
| Data model | data_viewer (reads it) | schema_editor | |
| Analytics | analytics_viewer | analytics_editor | |
| Audiences | campaign_viewer (reads) | audience_editor | |
| Campaigns | campaign_viewer | campaign_editor | campaign_approver, campaign_admin |
| Contact policies | campaign_viewer (reads) | contact_policy_editor | |
| Content and messaging | content_viewer | content_editor | content_approver |
| Activation | campaign_viewer (reads) | activation_editor | |
| Privacy | privacy_viewer | privacy_officer | |
| Integrations | integration_viewer | integration_editor | |
| Access | team_manager, api_key_manager | admin | |
| Cross-cutting | auditor | engineer | account_admin |
A few worth knowing:
adminhas full access to everything in the environment, including its team and API keys.engineeris the technical operator: data model, customer data, integrations, tracking, content approvals and the Campaign screens.auditorcan read everything and change nothing.agents_editorbuilds agent definitions and creates, starts, stops and switches agent instances, but cannot read their runs or log.agents_auditorreads definitions, instances, runs and the log and changes nothing. See Optima Agents.privacy_officerhas the whole privacy area: purposes, consent policies, erasure and deletion requests, data-subject exports, purpose grants and PII levels.team_managermanages the environment's members and invitations only;api_key_managermanages its API keys only. Together they are the two halves ofadmin's access management.campaign_editorcan build campaigns but only read channels, external systems, activation and contact policies. Those are changed byintegration_editor,activation_editorandcontact_policy_editor.campaign_adminholds all of them and also approves campaigns.account_adminmanages every environment of the account: members, custom roles and creating environments.
Your role also decides what happens on a screen you can reach. The console shows View only when your role can see an area but not change it, and No access when it cannot see it. If you try something your role does not allow, it tells you to ask an administrator of the environment.
Custom roles
Under Settings → Access → Roles, an account administrator can define custom roles by picking the API routes and methods a role may call. Built-in roles can be edited only by the platform operator. A role that is still assigned to people cannot be deleted. Reassign them first.
Managing access
Tenant administrators manage who has access.
Team (Settings → Access → Team) lists the people with access to the current environment:
- Select Invite member, enter their email address and choose their Roles.
- They receive an email with a link to join. See your first sign-in. Pending invitations appear under Pending invites and can be revoked until accepted.
- Use Edit roles to change someone's roles, and Remove from team to take away access to this environment. Removing a member does not delete their user account.
For each member you can also set Purpose grants (which purposes they may act under) and a PII level (how sensitive the fields they see may be). See consent and purposes.
A person's account can be locked, disabled or have its password reset by an administrator. Disabling or locking ends their sessions at once.
API keys (Settings → Access → API keys) are credentials for integrations. A key acts with its own roles, purposes and PII level and can do only what those allow, so give each integration a dedicated key with only the roles it needs. The key list also shows Connected AI apps, such as AI assistants someone connected using a key. Disabling, rotating or deleting a key cuts off everything using it.