Blog

New tech, old habits: Your Data & AI Operating Model needs an update

Many organizations still struggle to realize value with Data & AI. Navigating this era requires a critical update to their Operating Models.

Fanny Kassapian

February 25, 2026

5 minutes

This image may be AI-generated. The rest of this blog is not.

Ever been frustrated by an impractical kitchen? One where there's no counter space, the expensive espresso machine can’t be plugged in, and two people can't prep a meal without bumping into each other?

You’ve experienced a failing Operating Model. In this context, your Operating Model is the blueprint for your cabinetry: The logic of how you organize the space to serve your needs.

The Broken Engine Syndrome

Despite investments in people and tech, organizations are hitting a wall. Our latest Data & AI Monitor shows they struggle to create value, and while AI integration is a top priority, the essential building blocks—governance, culture, and experimentation—remain unevenly developed.

Companies have invested in Data & AI capabilities, but haven't changed the way they operate, and are now facing an unprecedented paradigm shift.

We’ve identified 10 trends that are putting Data & AI Operating Models under pressure. Each of these requires rethinking how we work:

When a Data & AI Operating Model is no longer aligned with strategic goals, the result is a widening gap between the organization's ambitions and its abilities to realize them.

Look for these red flags:

The ROI Ghost: After years of investment, the room goes silent when the CFO asks for measurable returns. Data has become synonymous with cost, not value.

Strategic Disconnect: The business sees the potential of AI, but execution is so slow and unreliable that they’ve stopped waiting. What’s left is a massive trust deficit.

Data teams as a Bottleneck: Data teams are reactive firefighters, patching a broken engine of legacy processes. The work is ungratifying and transactional, leaving no room to innovate.

Governance as a Barrier: Data governance and management policies are viewed as chores rather than capabilities. People bypass the system to get work done, and Shadow AI is everywhere.

The hard truth: Without an updated Operating Model, expensive tools and brilliant hires remain investments that never pay back. Funding will eventually stop, but market pressure will keep rising.

The good news: You’re hardly alone. Most organizations are navigating 2026 AI demands with 2018 operating structures. The engine isn't broken, it just needs a (significant) tune-up.

Updating your Data & AI Operating Model is the most impactful move to ensure tech starts working for you, rather than the other way around.

The Tune-Up: Reevaluate your Blueprint for Value

The perfect kitchen doesn't exist – only the one that fits your life: Are you a solo baker? Do you host dinner parties? Are there children to consider? And then, beyond the big oven or large countertop: What rules and safety standards do you need to enable, like making plates accessible for kids to set the table, or enforce, like keeping cleaning products safely out of reach?

Similarly, a functional Data & AI Operating Model must answer three strategic questions to bridge the gap between your organization's capabilities and its value agenda:

1. What opportunities lie ahead?

Don't start with the tech; start with the value proposition: What is the gap between your strategic agenda and the value delivered? What do your customers want in 2026 that they didn't need two years ago? Your Data & AI Operating Model must be built to deliver that value. It all starts with a clear vision translated into concrete use cases.

2. What do you need?

What do you need to deliver value with Data & AI? Consider core capabilities (e.g., cloud engineering, LLMOps, ...) as well as enabling capabilities (such as AI governance, data literacy, or change management): What do you need to make the tech work for you?

3. Who does the work, and how?

Now is the time to organize those capabilities to deliver value. As AI is revolutionizing knowledge work, organizations must revisit the way they operate through six dimensions:

Let's say you need AI-powered insights; this will allow business stakeholders to query data themselves and support their day-to-day decision-making. You've selected a tool that fits your requirements; now you need a functioning Data & AI Operating Model to drive value.

To move from a successful pilot to a value-generating asset, you must tune the six gears of your blueprint by answering the questions: Who does the work, and how?

Way of working and structure: How is this capability embedded into core business processes? Who owns the results? Who governs its usage? How are teams and processes designed to scale?
Customer & Product: What specific metrics define success and realized value? How is feedback collected for continuous improvement?
Technology: Who implements and maintains this tool? How is it integrated within the IT landscape? How are scalability, costs, and compliance monitored and enforced?
Data: Who ensures data is FAIR (Findable, Accessible, Interoperable, and Reusable) and AI-ready? How is accuracy validated? Who owns this data? How are privacy policies enforced?
People & Expertise: Who needs to do their work differently, and how are they taken along? Are incentives and personal goals aligned with the new behavior? How are we closing the talent gap?
Leadership & Culture: How does this capability align with organizational goals? Who owns and sponsors the vision, the budget, and the change narrative? How is this change prioritized?

Regardless of the capability in question, these 6 dimensions are the interconnected gears of your organization. They work holistically—one cannot be changed without rethinking the others. Now is the time to fine-tune them.

The Bottom Line

Your Data & AI Operating Model is the blueprint for your growth engine. It isn’t just an org chart; it’s the definition of how your company organizes its capabilities to deliver value. If you solely focus on technology and overlook the Operating Model, you’re just buying more expensive shelfware. As Conway's law best describes, a technical system reflects the organization that produced it.

In a follow-up blog, I will explore modern Data & AI Operating Models with concrete examples—Stay tuned!

Meanwhile, would you like to discuss your organization's Data & AI Operating Model? Please contact us! Our AI maturity scan can be used to assess specific aspects of your company. You will get recommendations to get your organization to the next level of AI maturity and get started with your AI Strategy.

Tags: