The nine questions
The same nine questions, answered twice
Numbers in brackets point to the numbered sources at the foot of the page. The line under each question says what a complete answer contains.
1 · Where must the data live for questions and agents to work?
A complete answer names the systems the data has to be copied into, and the systems that are read where they stand.
CelonisCelonis builds a Process Intelligence Graph from event data drawn across a company's systems, and Data Core adds zero-copy, two-way links to Databricks and Azure data lakes, so process data stays linked to its source [3]. AgentC then carries this process context to agents built on other platforms [1].
Lima JunoYour rows stay in the systems that hold them. Juno registers each table through the Lima data fabric and reads it in place through the federated query engine, joining several sources in one answer. 2 · How far does it reach into SAP, Oracle and planning systems?
A complete answer names the interface used for each system and says which of them are live today.
CelonisCelonis describes its graph as reaching system-agnostically across SAP S/4HANA, Oracle Fusion, Salesforce, ServiceNow and Blue Yonder, among others named on the pages reviewed [2]. This reach covers seeing and connecting process data across these systems; a specific write back into one of them belongs to the Orchestration Engine, described separately.
Lima JunoMore than 40 connectors cover SAP ECC and S/4HANA, Oracle Fusion and E-Business Suite, planning suites, warehouse systems, lakes, databases, files and REST services. One question can join several of them, and the one agent in production, the PO cancellation agent, runs on federated ERP data the same way. 3 · How is the shared meaning, the semantic model or ontology, built, and by whom?
A complete answer says what arrives ready, what your team builds, and how long that modelling work takes.
CelonisCelonis mines its own shared structure from event logs across systems: the Process Intelligence Graph reconstructs how work actually moved through a company's processes, built and owned by Celonis's own mining technology [3]. Partner platforms building agents on this context draw their business definitions from their own studios, with Celonis supplying the process layer underneath [1].
Lima JunoLima’s supply chain ontology arrives filled in: materials, plants, suppliers, orders, stock and the relationships between them. Juno proposes a mapping for each registered column, and your team confirms it. 4 · How do business users ask questions, and is the working shown?
A complete answer shows the question, the query that ran, the sources read and the access rules applied.
CelonisCelonis frames its own pages around three steps, Analyse, Design and Operate, built on a context model that feeds the question interface a partner platform such as Copilot Studio or watsonx Orchestrate provides [5]. How that agent shows its reasoning to the person asking is a detail the partner platform controls, separate from the process context Celonis supplies [1].
Lima JunoPeople ask ChatLima in plain language. Each answer shows its working: the ontology terms used, the systems read, the seconds taken and the query itself, under the asking person’s row and column access. 5 · How are agents and rules defined, and by whom?
A complete answer says who writes a rule, in which language, and how it is reviewed before it runs.
CelonisAgents that use Celonis's process context are built in partner studios such as Copilot Studio, watsonx Orchestrate, Bedrock or CrewAI, each keeping its own way of defining an agent [1]. Celonis's MCP server is what supplies the process intelligence those agents call on while they run.
Lima JunoOne agent runs in production, the PO cancellation agent, built from a policy a planner stated in plain language. Lima’s agent blueprint designs 960 further roles across five chains, yet to be built the same way: a planner states the intent and the Planner Agent drafts the rules for review. 6 · How is autonomy set, and how do approvals work?
A complete answer gives the levels available, who approves what at each one, and how a level is raised and lowered.
CelonisCelonis positions most agent output as a recommendation for a person to action, with the Orchestration Engine, added from the Emporix acquisition, able to carry out a coordinated step across agents, people and systems when configured to do so [3]. A named, numbered autonomy ladder is a structure the pages reviewed leave to the client or partner platform to set.
Lima JunoEach agent carries a level you set, A0 to A4. At A2 a named person approves every action; at A3 the agent acts inside value, budget and confidence limits and sends exceptions to people. The level is one of eight settings on an agent, each set by a named person. 7 · How do actions reach the ERP, and is each write confirmed?
A complete answer names the interface for each ERP and says where the acknowledgement of a write is kept.
CelonisThe Orchestration Engine carries out the actions Celonis coordinates across agents, people and systems, reaching into applications such as SAP or Oracle when a client configures it to act there [3]. A named confirmation step logged against each individual write stays undocumented on the pages reviewed.
Lima JunoWrite-back uses each ERP’s own supported interfaces: released APIs and BAPIs for SAP, REST resources and interface tables for Oracle. It starts per application after a confirmed test write, and every write keeps the acknowledgement. The PO cancellation agent in production writes back this way, under human approval. 8 · How does it learn from people’s decisions, and what is audited?
A complete answer says what is recorded for each decision and how that record shapes the next recommendation.
CelonisCelonis's own strength is process-level traceability: because the graph is mined from event logs, a reviewer can see how a case actually moved through a process alongside any agent action taken on it [3][4]. How the platform's own agents learn from a person's override of a recommendation stays undocumented on the pages reviewed.
Lima JunoEvery accept, edit and override is stored with the rule version, the facts at that moment and the outcome, so the next recommendation improves. Drift detection flags when live data moves away from what a rule assumed. 9 · What is the pricing unit, is there a trial, and where do analysts place it?
A complete answer names the unit of charge, the route to try it, and the analyst reports that cover the category.
CelonisCelonis keeps pricing off its public pages and offers a free entry point alongside a scheduled demonstration for a fuller look [5]. Constellation Research names Celonis among the vendors on its 2026 ShortList for Decision Intelligence Platforms, a category that also names Aera, FICO, IBM, Pega and SAS [6].
Lima JunoJuno is licensed with the Lima platform and priced per engagement. The token manager budgets and meters every model call, per agent and per team. Twenty use cases are worked through on this site, one page each, and the analyst guide sets out the markets these products sit in.