Best Practices

The Camunda Best Practices distill Camunda's experience with BPMN, DMN, and agentic orchestration on the Camunda toolstack, drawing on consulting engagements, community feedback, and customer projects. They offer a blend of conceptual and practical guidance, generalized from current project experience. They are not definitive: learning is ongoing, and how well a practice applies depends on your situation.

Note that Camunda extends the same guarantee to Best Practices as to the core product. However, Camunda cannot ensure the absolute accuracy or timeliness of the information provided, and disclaims any liability for damages resulting from applying these recommendations.

Project management

Following the customer success path

Follow proven steps for evaluating and introducing process automation successfully.

Doing a proper POC

Use a proof of concept to check whether your approach and technology fit your needs.

Architecture

Data flow

Understand how data moves through Camunda 8 and why it matters for sizing your environment.

Deciding about your stack

Our recommended technology stack for building Camunda solutions.

Size your environment

Understand sizing considerations for Camunda 8, for both SaaS and Self-Managed.

Understanding human task management

Use Camunda task management features, or implement your own requirements for readable models.

Development

Connecting the workflow engine with your world

Learn how to connect the Zeebe workflow engine with your application or remote systems.

Service integration patterns with BPMN

Choose the right BPMN modeling approach when integrating systems and services.

Writing good workers

Implement job workers that perform the work behind your service tasks.

Dealing with problems and exceptions

Handle exceptions, leverage retries, and use incidents to deal with problems in your processes.

Handling data in processes

Understand how to work with process variables and associate data with process instances.

Routing events to processes

Choose the right technology to start a process instance or route a message to a running one.

Testing process definitions

Test your executable BPMN processes with automated, fast in-memory workflow engine tests.

Local development with element templates and Camunda 8 Run

Use Camunda 8 Run with element templates in your local development environment.

Modeling

Creating readable process models

Create visual process models that are easy to read, discuss, and remember.

Naming BPMN elements

Name activities, events, and gateways in your BPMN diagrams from a business perspective.

Naming technically relevant IDs

Properly name technical element IDs in your BPMN diagrams for executable flows.

Modeling beyond the happy path

Model the happy path first, then incrementally introduce problems and exceptions.

Modeling with situation patterns

Document recurring patterns and find satisfying solutions for modeling them.

Building flexibility into BPMN models

Build flexibility into process models to handle operational problems or allow human intervention.

Choosing the DMN hit policy

Understand the different ways to evaluate rules in a DMN decision table.

Choosing the resource binding type

Understand the differences between latest and deployment binding for linked resources.

Operations

Versioning process definitions

Understand how Camunda handles evolving process definitions through versioning.

Reporting about processes

Leverage historical process data collected by the Camunda engine to generate relevant reports.

CI/CD guidelines

Element templates at scale

Provision element templates at runtime and make them available across your organization and Desktop Modeler.

Agentic orchestration

Processes that hand a step to an AI agentAI agentAn addressable execution of an LLM-driven loop with shared memory context across iterations. An agent runs an agent loop where the model decides what to do next, which tools to invoke, and when to stop. run on the same engine, variables, and audit trail as fully deterministic ones, so the practices above still apply. For specific guidance on designing, tuning, and evaluating agents, see the agentic orchestration documentation:

• Design and architecture covers deciding where an agent belongs in a process, blending deterministic and AI-driven steps, and guardrail patterns such as human-in-the-loop escalation.
• Model recommendations covers writing tool descriptions, managing the context window, sanitizing tool output, and prompting.
• Choose the right model covers benchmarking candidate LLMs against your own requirements.
• Evaluate AI agents covers testing non-deterministic flows, monitoring agents in Operate, analyzing token and tool usage in Optimize, and detecting agents that go off the rails.