When I first opened OpenAI’s Agent Builder, I didn’t expect it to feel so much like conducting a symphony. Yet, that’s what it turned into — an orchestration of intelligent agents, each with its own role, personality, and purpose, collaborating to transform an idea into a structured business model.
In my experiment (see video below), I modeled my first process: a four-agent workflow designed to analyze, design, and report on business ideas. And to put it to the test, I used one of my favorite teaching case studies — the Hotel Mobile Check-in. The result? Surprisingly convincing.
This experiment was inspired by a recent brainstorming session with Alex Osterwalder, where we discussed the future of business model platforms like Strategyzer and how AI could transform the way we analyze, design, and test business ideas. The idea was to explore whether intelligent agents could collaborate — much like human experts — to support entrepreneurs and educators in developing, iterating, and validating business models in real time.
From a Single Prompt to an Agent Orchestra
In OpenAI’s Agent Builder, each agent can be configured individually, much like creating your own GPTs in the ChatGPT app. You can define their role, expertise, tone, and instructions — and then connect them into a workflow that feels alive.
For this first experiment, I designed a process where each agent plays a specific business role:
-
BN – Business Narrator: Sets the stage, describes the business idea, and ensures narrative coherence.
-
BA – Business Analyst: Dissects the idea, explores feasibility, risks, and potential improvements.
-
BD – Business Designer: Shapes the concept into a table that can be used to complete for example a Business Model Canvas or Value Proposition Canvas.
-
BR – Business Reporter: Summarizes everything clearly, preparing a professional report ready for presentation.
Each of these agents communicates with the next in the sequence. The flow passes through user approval gates, allowing me to review, tweak, or reject outputs before moving forward. It’s like guiding a team of virtual consultants — fast, structured, and collaborative.
The Workflow: Visualizing Intelligence in Motion
The process starts with the Business Narrator generating a coherent description of the business idea. Once approved, it hands off to the Business Analyst, who evaluates the strengths, weaknesses, opportunities, and threats.
Next, the Business Designer synthesizes everything into actionable structures. Finally, the Business Reporter turns the findings into a crisp, human-readable summary.
The workflow feels remarkably natural, almost like project management software meeting strategic consultancy — except the “team” works 24/7, never forgets context, and improves with every iteration.
Testing It: The Hotel Mobile Check-In Case
To validate the model, I used a familiar case: Hotel Mobile Check-In, a service concept I often use in my classes to illustrate innovation in tourism and hospitality.
The agent orchestra handled it with ease. The Narrator described the customer experience, the Analyst identified operational bottlenecks, the Designer translated insights into a coherent service blueprint, and the Reporter wrapped it all in a clear strategic summary.
It felt like having a multi-disciplinary consulting firm — powered by AI — at my fingertips.
Why It Matters
This approach isn’t just a novelty. It represents a new paradigm for entrepreneurial education and innovation management. Instead of static templates or one-off analyses, we can now simulate collaborative intelligence ecosystems where each agent specializes, learns, and co-creates.
For teachers, it’s a revolution in experiential learning. For entrepreneurs, it’s a new way to iterate business ideas faster. And for researchers, it’s a live experiment in organizational AI design — the art of composing digital teams that think together.
Next Steps
My next step will be to scale this approach:
-
Integrate more advanced agents (e.g., a Customer Researcher or Financial Forecaster).
-
Add memory between sessions for continuous project development.
-
Test the orchestration with real student projects in class.
The journey has just begun. The future of business analysis, design and leadership might not be about one AI, but about many AIs working together — harmonized like a digital orchestra.