People + AI roles
Who decides, what AI assists with and what agents may do alone.
Business AI Transformation
We connect SAP Business AI, Joule and agents to your processes, data and rules, and raise autonomy step by step, with governance from day one.
The problem
Today
With the transformation
Commercial model
No standard price: it depends on your ambition, domains, landscape and data.
Phase 1
Readiness, data and an opportunity portfolio.
Assessment
Phase 2
Operating model, architecture, governance, business case.
Blueprint
Phase 3
First capabilities live, results measured.
First domain
Phase 4+
More processes, domains and autonomy.
Expand
The approach
Need one document automation or one workflow? An Automation Sprint is the better fit.
From automation to autonomy
People read, check SAP, decide and update.
Level 0
Rules handle predictable steps.
Level 1
AI reads the email and shows the SAP context.
Level 2
An agent investigates and proposes a fix.
Level 3
Agents run the process within policy.
Level 4
Not every process should reach the top. The right level depends on risk, data quality, regulation and how reversible a decision is. A fully autonomous domain (level 5) is a goal only where it pays off.
Autonomy is earned step by step.
How it works
Strategy, SAP landscape, data, current AI and constraints.
Use cases ranked by value, feasibility and risk.
Target process, roles for people and agents, governance.
A few capabilities live, with measured results.
More processes and domains as trust grows.
Think enterprise. Deliver in steps.
Foundations
Who decides, what AI assists with and what agents may do alone.
Joule, SAP and custom agents built in Joule Studio, agentic workflows.
Live SAP context and business data agents can trust.
Identity, permissions, approvals, audit, monitoring and cost control.
APIs, events and agent protocols, only where the process needs them.
Postings, checks and SAP changes stay deterministic. Agents reason and recommend.
SAP capabilities depend on your edition and SAP’s release plans. Read our analysis of Joule agents and the Autonomous Enterprise.
Examples
Procurement
Buyers chase confirmations, delays and claims across systems.
Agents monitor, follow up and prepare; buyers negotiate and decide.
Finance
Invoice exceptions, reconciliation and collections fill the week.
AI gathers context and proposes; people approve.
Supply chain
Disruptions are found and coordinated by hand.
Detect, understand, act within policy, monitor.
Customer service
“Can we deliver by Friday?” needs five teams.
Inventory, ATP, logistics and rules in one answer.
Illustrative target pictures, not customer results.
Business case
Think of one area, such as invoice exceptions, supplier follow-up or customer questions.
Your numbersChange any value. The result updates as you type.
Illustrative, productivity only: faster cycle times, fewer errors and better decisions come on top. Runs in your browser; nothing is sent.
Use these numbers in my requestFAQ
To let them focus on the work where judgment and expertise matter most. AI takes the repetitive coordination.
Where reasoning, context or ambiguity justify them. Many steps are better served by plain automation.
Yes, when the process needs it, through open protocols such as A2A and MCP and a governed gateway.
Yes. We check data readiness first; some initiatives start at once, others after groundwork.
Yes, and it is often the best way: procurement, finance or supply chain first, then expand.
Free, 30 minutes
Start by understanding where AI creates value, and where it does not.
Google Calendar loads only when you click. Open in a new tab