The Invisible Infrastructure of Innovation: Why Data Governance is Foundational to the Future of Healthcare
Co-authored by Sebastian Sbirna, Novo Nordisk
August 26th, 2025
In the race to transform healthcare, technology steals the spotlight, AI-driven diagnostics, personalised medicine, and digital therapeutics dominate headlines. But behind the scenes, there is a quieter, less glamorous foundation upon which all of this innovation stands: Data Governance.
Without strong governance, even the most advanced tools risk becoming untrustworthy. “People often think of governance as bureaucracy,” Sebastian says. “But in reality, it’s the framework that enables digital transformation to be sustainable, scalable, and ethical.”
Sebastian’s experience spans banking, energy, and pharma, three heavily regulated industries where data accuracy can mean the difference between compliance and catastrophe. But it’s in healthcare, particularly at Novo Nordisk, that the human stakes are most tangible.
“Data Governance is about clarity,” he explains. “It’s the roles, responsibilities, and processes that ensure data is well managed. That includes everything from how data is collected to how it’s used, interpreted, and ultimately disposed of.”
His current role involves working at the enterprise level, supporting different business lines, from “Manufacturing & Supply Chain” to “International Sales & Commercial Operations”, in order to embed Data Governance principles that reduce risk and drive insight.
The Domino Effect of Small Errors
What makes Data Governance so critical, yet so invisible, is how small data errors at the source can cascade through an organisation, creating widespread consequences.
“Imagine a sales representative logging a doctor visit,” Sebastian says. “If the form allows for free text input without validation, someone might type an incorrect address or misspell a hospital name. That data then flows into analytics, KPIs, forecasting, and even AI models.”
And that’s where the real problem begins: “Business leaders might look at a dashboard and say, ‘These figures don’t make sense.’ By then, it’s too late, the error has propagated through the system.”
In healthcare, these missteps aren’t just operational, they can be life threatening. Incorrect patient records, misinterpreted clinical trial data, or flawed drug usage projections can compromise both care delivery and regulatory compliance.
As healthcare increasingly embraces artificial intelligence, the risks of ungoverned data are magnified. AI models, particularly large language models and machine learning systems, rely on data quality to deliver meaningful results.
“AI doesn’t fix bad data,” Sebastian warns. “It accelerates whatever data you give it. If there are errors, they’ll be embedded in the output. Now those flawed insights are arriving faster, more confidently, and influencing high stakes decisions.”
Take Novo Nordisk’s internal commercial initiative, “Next Best Action (NBA)”. It’s a machine learning model designed to help sales representatives determine their most strategic move, perhaps suggesting a webinar invitation after a face-to-face meeting with a healthcare professional. “Sounds simple,” Sebastian says, “but the number of variables involved is staggering. Without structured, accurate data, the model’s recommendations can be misleading.”
Governance Is Not One Size Fits All
One of the most compelling takeaways from our conversation is that Data Governance doesn’t have to be rigid or heavy-handed, especially for startups.
“I don’t think full governance programs are necessary for every company,” Sebastian admits. “If you are a 5-person team building a digital health product, focus on what’s essential, like data quality, traceability, and having basic metadata to understand your data’s origin.”
He cautions against adopting complex frameworks too early, which can stifle innovation. Instead, emerging health tech startups should apply “just enough governance” to avoid chaos while maintaining their agility.
Crucially, he also distinguishes Data Governance from data privacy. “Privacy is about consent, ownership, and who has access,” he explains. “Governance is about how data moves through systems and how it’s interpreted. They’re linked, but they’re not the same.”
Another key theme? Data contracts. As smaller health startups collaborate with large enterprises, defining how data is shared, stored, and transformed becomes essential. “If you send your data to a bigger partner, they need to know what they’re receiving, in what format, under what conditions,” Sebastian says. “Having a data contract helps create trust. It also saves time for engineers and analysts who need to ingest and use that data safely.”
In short, good Data Governance isn’t just about reducing risk, it’s a commercial enabler that fosters more seamless partnerships and scalable operations.
Recommendations for Getting Started
For those seeking to build governance into their digital health roadmap, Sebastian points to the Data Management Body of Knowledge (DMBOK), created by DAMA International, as a go to resource. “It’s a gold standard, used also by our company, Novo Nordisk”, he says.
As we wrapped up our conversation, Sebastian emphasised a message that resonates deeply with the mission of electronRx: Data should empower, not mislead. “We often think of AI, analytics, and digital health as futuristic. But none of it works without data we can trust,” he says. “Governance may feel like an overhead, but in reality, it’s the foundation that keeps the whole system upright.”
In a landscape where decisions are increasingly data driven, the cost of errors, especially in healthcare, is measured not just in lost efficiency, but in lost timely ability to directly help patients. Getting it right starts with governance.
