BPM (Business Process Management) is a methodology that enables the design, execution, control, and optimization of business processes to improve efficiency and organizational outcomes.
A few days ago, I wrote in a LinkedIn post that “BPM as we know it” is changing. Not because it has died, but because a new layer has emerged that is revolutionizing the way we design and manage processes: the agentic layer.
This transformation is not only technical but also strategic. To better understand it, I had the opportunity to speak with Gastón Milano, CTO of GeneXus, who provided me with a key perspective on the topic.
Traditional Processes vs. Agentic Processes
Until now, many processes were designed as structured, step-by-step flows, where each stage was rigidly defined. This worked well for predictable and highly regulated processes. However, with today’s complexity, this approach is losing effectiveness.
Gastón Milano explained to me that today it might make more sense to start with a dynamic, agentic process, without rigid BPM, allowing the system to discover the best path from point A to point B. Only afterward, if necessary, is the flow formalized and structured.
Why Change the Traditional Approach?
There are several reasons to adopt agentic processes:
- Flexibility: The agent can determine the appropriate course of action based on the context and available tools.
- Efficiency: It avoids forcing an overfitted process when the agent can find better routes.
- Adaptability: The system adjusts to changes and exceptions in real time, without relying on rigid rules.
The Ideal Combination: Deterministic Processes and Dynamic Tasks
Of course, not everything can be improvised, especially in sectors with strong regulations and legal requirements. In these cases, structured BPM remains vital to ensure traceability and regulatory compliance.
But even within these processes, it is possible to incorporate an agentic approach to handle specific tasks, meaning:
- Structured processes that contain dynamic tasks managed by intelligent agents.
- Agents that know when to follow rules and when to adapt or “break” them in a controlled manner.
This approach, which combines symbolic and probabilistic elements (also called neuro-symbolic), represents the most robust evolution for BPM:
- Structured processes
- Tasks solved by intelligent agents
- Integrated human validation
- Traceability as an integral part of the design
A New Mindset for a New Era
This evolution is much more than technology: it’s a mindset shift. It invites us to rethink how we design processes, placing adaptive intelligence and human-machine collaboration at the center.
The question is no longer processes vs. agents, but how and when to combine both to achieve the best results.
At GeneXus Consulting, we support organizations in their digital transformation processes. We design and implement solutions that combine the best of both worlds: structured processes when control is needed, and intelligent agents when flexibility is required. Let’s explore together how to apply this in your organization. Let’s talk.