adanse.ai logoadanse.ai
Documentation

How Adanse works, end to end.

Reference material for the people who configure, govern and answer for an AI workforce.

Getting started

What Adanse is

The operating layer between your business and the AI acting on its behalf.

Workspaces and tenancy

Every record belongs to one workspace; membership decides access.

First hour

Connect a system, review Business DNA, hire one AI employee at R1.

Roles

Member, decider and owner — deciders can approve, reject and undo actions.

Understanding the business

Business Memory

Typed, sourced facts with freshness and confidence, linked to entities.

Business DNA

Declared behaviour compared against observed behaviour over time.

Intelligence Graph

Entities, systems, employees and decisions as nodes and typed edges.

Knowledge spaces

Curated document sets scoped to departments and employees.

Workforce

AI employees

A named worker with a department, remit, skills and an autonomy boundary.

Skills and procedures

Reusable, versioned procedures an employee is permitted to run.

Teams

Hybrid human and AI teams with shared objectives and escalation paths.

Capabilities

Registered tools and actions an employee may call, with scopes.

Governance

Autonomy R0–R5

From suggest-only to fully delegated, applied per employee and per action.

Policies

Conditions that must hold before an action executes, evaluated at decision time.

Decision records

Reasoning, inputs, confidence basis, policy, approver, result and value.

Audit

Append-only evidence written as work happens, linked back to its decision.

Reference

Autonomy levels

R0
Observe only. The employee reads and reports, never proposes an action.
R1
Suggest. Proposals are drafted for a human to take or discard.
R2
Act on low-risk work with post-hoc review.
R3
Act within limits; anything outside the limit escalates for approval.
R4
Act broadly; only exceptions and threshold breaches escalate.
R5
Delegated. Acts alone within policy, with audit and outcome review.

Levels are enforced per action, not per employee alone — a single employee can be R4 for reporting and R1 for payments.

Adanse Credits

One unit of execution. No tokens to reason about.

Customers should not have to price model tokens, workflow steps, browser hours or sandbox CPU seconds. Execution is converted into a single unit, so the commercial contract does not change when the infrastructure underneath it does.

What draws on your allowance
Simple AI task
Business-data lookup
AI decision
Complex reasoning
Business action
Browser operation
Code execution
Complex autonomous run

Every run shows the credits it used in your workspace, so you can see exactly where the allowance went.

Standard and premium execution

Reasoning, workflow execution, API and MCP actions, knowledge and memory retrieval and evaluations draw on the standard allowance. Browser execution, code sandboxes, premium models and large document processing draw at higher rates or against their own allowance.

Hard budgets, not surprises
Soft warning at 70% · second warning at 90% · allowance reached at 100%.
At the limit the organisation chooses: stop execution, continue with overage, or draw down a prepaid balance.
Budgets can be set per organisation, department, AI employee and automation.
Return

Cost is only meaningful next to outcome.

Billing in Adanse is not a credit meter. Spend is presented against the measured value the workforce produced in the same period, from the same records.

Adanse — illustrative month
Platform cost$999
AI execution$186
Total Adanse cost$1,185
Business outcomes
Hours saved327
Estimated labour value$9,810
Revenue recovered$14,300
Errors prevented17
Invoices resolved1,842
Average resolution time↓ 61%
Unauthorised actions0
Estimated measurable value$24,110
Estimated return20.3×

Illustrative figures. In your workspace these come from the outcome ledger.

Enterprise

Committed capacity, not a fixed ceiling.

Enterprise agreements are composed rather than listed, and priced annually against committed execution.

Included with Enterprise
Committed execution capacity, agreed annually
Dedicated or isolated environment
Enterprise identity, SCIM provisioning and data residency
Custom roles and custom audit retention
Named support with agreed service levels
Onboarding, assurance review and quarterly outcome review
What shapes an enterprise agreement
Committed execution capacity and negotiated unit pricing at volume.
Required environments — production, staging, dedicated or isolated.
Security and compliance requirements, including identity, provisioning and residency.
Support model and service levels.
Talk to us