After years of hands-on experience with AI projects – ranging from Natural Language Processing and Machine Learning to Deep Learning, generative AI, and agentic AI – I started asking myself a crucial question: Why do some Enterprise AI projects succeed while others stumble? Reflecting on this, I developed a simple, intuitive framework built around three key dimensions. It’s a framework that helps explain what really drives success in AI projects and what often gets overlooked.
The triangle of success: People at the heart of AI
The visual shows an equilateral triangle representing the three equally important dimensions of a successful AI project: Use Case, Technology, and Data. These three sides form a balanced structure, emphasizing that all dimensions must be present and aligned. If one side is missing or underdeveloped, the triangle – and therefore the project – loses its stability.
The center of the triangle holds the Personas. These aren’t just placeholders – they represent the people in the organization who define, use, or are affected by the AI solution. In other words, personas can also be thought of as key stakeholders. Placing them at the heart of the triangle sends a clear message: AI projects aren’t just about technology or data – they’re fundamentally human-centered. Each persona brings their own unique perspectives, needs, expectations, and skills which in turn shape every aspect of the project.
Personas don’t act as a fourth dimension – they’re the connecting force that gives meaning to Use Case, Technology, and Data. Their presence reminds us of a key truth in AI projects: Success only comes when the human perspective is considered across all three dimensions. That’s how a solution delivers real value, is technically feasible, and is built on trustworthy, relevant data.
So, when thinking about a successful AI project, it’s important to keep these three dimensions in mind – but also to understand who is involved in each. Let’s take a closer look at each dimension and the personas who typically drive it forward.
Meet the heroes behind each dimension
First, there’s the Use Case. This is where the real-world benefits of the project are defined. Typically, the Product Owner and Business Users / Domain Experts take the lead, shaping both the use case and business case, making sure the project is set up to deliver real, measurable ROI (return on investment).
Next comes Technology. Once the “why” is clear, the “how” takes center stage. Here, Low-/Pro-Code Developers and AI Engineers bring the solution to life, while the Enterprise Architect and IT Operations ensure that the infrastructure can support it. The Head of AI continues to act as the bridge between the business goals and technical execution.
Finally, the Data perspective ensures that the project has the right data foundation. Data Engineers manage and prepare the data – ensuring quality and quantity, while the Data Protection Officer handles compliance and legal requirements. Together, they ensure that the AI solution is built on trustworthy and usable (AI-ready) data.
Supporting all three dimensions is the Change Manager, who guides the team through the adoption process, ensures smooth cross-functional collaboration, and drives a change management approach that involves everyone in the project – and beyond, if it moves into production.
When the puzzle isn’t complete
These dimensions and personas1 work together like a carefully balanced system. But here’s the crucial question: What happens if one or two of these dimensions are missing when you plan your solution? Let’s explore this through three illustrative scenarios.
Focusing only on Technology: Imagine a team developing an AI agent, fully immersed in the technical side. The AI Engineers and Developers select the most advanced tools and frameworks, while the IT Operations team and Enterprise Architect ensures the infrastructure is solid. Everything looks impressive on paper – and may even shine in local setups, sandbox environments or MVPs (minimum viable product). But in the excitement over technology, the Product Owner and Business Users / Domain Experts are left out of the conversation. The actual business needs aren’t questioned, and the end users’ benefits remain unclear. Crucial data considerations are overlooked because the Data Engineer and Data Protection Officer haven’t been fully engaged. The result? The project delivers technical elegance but little real-world value – the team is trapped in a purely technical bubble, and the company gains no meaningful advantage.
Focusing only on the Use Case: Now imagine the business unit comes up with a creative and promising idea for using an AI agent to optimize customer service workflows. The Product Owner and Business Users / Domain Experts define the use case and business case in detail, highlighting potential benefits for the company. Excitement is high, but the AI Engineers, Data Engineers, and IT Operations team haven’t fully evaluated the technical feasibility or data requirements. During implementation, the team discovers that critical data is unavailable, incomplete, or restricted due to privacy regulations. The technology envisioned cannot easily integrate the required datasets. Despite a compelling use case, the project falters because the technical and data dimensions were not properly aligned – good ideas alone aren’t enough without the right technical skills involved and infrastructure supporting them.
Focusing only on Data: Finally, imagine the business unit is eager to analyze specific client data, which is considered a valuable knowledge source for an agentic AI solution. The Data Engineer doesn’t see any hurdles in aggregating the data, and the Data Protection Officer has no issues from a legal perspective with analyzing the data within the company’s infrastructure. However, the Enterprise Architect and IT Operations team recognize a critical constraint: Company policy mandates that all AI applications must be hosted in the cloud. On-premises hosting isn’t possible, the infrastructure isn’t available, and the budget cannot cover it. As a result, even this promising use case cannot be realized. The potential value is lost because the technology and operational dimensions weren’t fully considered, showing that even excellent data and legal compliance aren’t enough without alignment across the other dimensions.
Why balance matters in AI projects
Looking at all three dimensions of this framework, one thing becomes clear: An AI project is far more likely to succeed when each dimension – Use Case, Technology and Data – is given equal attention. Overlook even one, and the risk of falling short – or failing entirely – increases dramatically. That’s why it’s crucial to keep all three perspectives in mind, along with the key personas involved, to give your project the best chance of being successful.
Ultimately, their presence serves as a reminder of a fundamental truth in AI projects: Success comes only when a human-centered perspective – one that respects the talents and skills of the people involved – is integrated into every dimension.
1 The personas mentioned are exemplary and intended as a guide. They are not guaranteed to be exhaustive, and not all of these roles may exist in every organization. Depending on the size of the company, responsibilities may overlap, with one person taking on multiple roles. This section is therefore meant to provide a flexible reference that can be adapted to your organization’s specific structure and roles.
💬 Comments or suggestions? I’d love to hear from you! You can leave a comment on LinkedIn, send me a direct message or connect with me, or use the contact form on my blog. I look forward to hearing your story!
💡Liked what you read? Subscribe to my blog and get new posts delivered straight to your inbox – so you never miss what’s next.