The evidence suggests almost all business leaders are piloting or investing in AI initiatives, and biopharmaceutical giant Boehringer Ingelheim is committed to investing in emerging technology that could have life-altering consequences.
The company’s 55,000 employees focus on developing innovative therapies that can improve lives in areas of high unmet medical need, with AI and data playing an increasingly crucial role in their work.
Also: 4 ways business leaders are using AI to solve problems and create real value
Global CIO Markus Schümmelfeder told ZDNET that emerging technology can open all kinds of possibilities when its adoption is accompanied by organizational change: “AI together with big data availability and access to the right capability is the real game-changer.”
So, how can business leaders drive successful organizational change in an age of AI? Schümmelfeder and his colleague Oliver Sluke, head of IT research, development, and medicine at Boehringer, told ZDNET their four best-practice tips for AI-enabled business transformation.
<!–>
1. Create a data environment
Most digital leaders agree: before you start tinkering with technology, you must ensure your data is managed, sorted, and accessible.
Boehringer has a data ecosystem called Dataland, which has been in place since 2022. Schümmelfeder said the ecosystem collates data from across the enterprise, allowing professionals to run simulations and data analyses safely and securely.
“To be able to execute use cases and analytics, you need a successful data environment, so we created that.”
He explained how the ecosystem is about much more than storing data. Dataland also includes some critical data management and analytics systems.
“We have dozens of tools sitting on top of it, like Snowflake and Collibra, to catalog the data, make use of it, and bring information into AWS.”
Also: Is your business AI-ready? 5 ways to avoid falling behind
Sluke said another key element of Boehringer’s data environment is the One Medicine Platform, powered by the Veeva Development Cloud, which combines data and processes, enabling Boehringer to streamline its product development.
“We previously had 55 individual small systems that did the work of Veeva. It was very fragmented, as you can imagine. It was not a harmonized data model,” he said.
The Veeva platform works with Dataland to form what Sluke referred to as a state-of-the-art technology stack.
The result is a consistent approach to IT and integrated insights for life-changing research.
“IT and medicine came together with this transformation,” said Sluke. “This shift goes way beyond just replacing a tool, it’s also a different way of working.”
2. Build an AI platform
With enterprise information consolidated in Dataland, Boehringer uses the platform to explore and exploit AI.
“We have the data environment and the tools on top,” said Schümmelfeder. “We have a stack for all the machine learning and AI topics, and we’ll provide more tools as the technology develops.”
The company’s specialist approach to AI, called Apollo, allows employees to select from 40 large language models (LLMs).
To an outsider looking in, 40 models sounds like a lot of choice. However, Schümmelfeder said this range is important for performance and efficiency reasons.
“That approach means that, when you have a use case, you can run different LLMs against your data and get specific answers,” he said.
Also: The top 20 AI tools of 2025 – and the No. 1 thing to remember when you use them
Boehringer does not develop models internally. Schümmelfeder said the rapid pace of AI development makes it more sensible to dedicate IT resources to other areas.
Alongside mainstream models like Gemini and ChatGPT, the company uses niche models that are more appropriate for research than general models.
“Certain LLMs are better for specific use cases than others,” he said. “Efficiency is also an issue. You cannot use super-expensive models for every question. That approach does not make sense.”
3. Use an Agile approach
Companies that want to exploit their data platforms and models must have professionals who can work on these foundations.
Sluke said Boehringer recognized at an early stage that it needed a new way of working.
“Over the last five years, we’ve been on a software engineering journey,” he said. “We recognized it’s not just about data. Our IT organization also needed to have capabilities to build applications using a state-of-the-art technology stack.”
Sluke said the aim was to establish Agile and continuous delivery in software engineering, allowing the organization to produce code quickly and effectively.
“We saw from the beginning that data was just one element – we also needed to put algorithms on top, which was a good decision, because then two years ago or so, when all this AI hype started, we were immediately able, with our software engineers, to start using these technologies,” he said.
Also: 5 ways to manage your team more effectively in the AI-enabled enterprise
Schümmelfeder said making the shift to an Agile way of working might sound easy, but it isn’t.
“Nothing’s more uncomfortable when you tell someone, ‘You did it this way yesterday, but you’ll do it another way tomorrow.’ People will say, ‘I was already successful without that approach. Why should I change?'”
His team made this shift to Agile through communities of practice, where people across the IT organization learned new skills through hands-on activities.
The organization now runs about 80% of its projects via an Agile methodology.
“Scrum is a buzzword,” he said. “But in this case, we’re proven you change how the organization works, and not just in which boxes the organization works.”
4. Identify strong use cases
The other key element that drives organizational change is focusing on AI use cases that help the business exploit its data.
Schümmelfeder outlined three specific AI-enabled use cases. First, Smart Process Development, which uses machine learning and genetic algorithms to improve biopharmaceutical processes, such as capture chromatography.
Second, he pointed to Genomic Lens, an AI-based process that the company uses to generate insights that help scientists discover new disease mechanisms in human DNA.
“It’s a more precise approach and provides faster identification of new therapeutic concepts based on genetic patterns,” he said.
“We use machine learning, big data processing, and predictive algorithms. We take data from various biobanks, and the AI detects novel genetic patterns and disease mechanisms.”
Also: 5 ways to turn AI’s time-saving magic into your productivity superpower
Finally, the company uses algorithms and historical data to identify populations for clinical trials. Sluke gave more details.
“It’s crucial for us to identify the right population before we run a clinical trial. Based on our historical data, we run an algorithm, and we can speed up the entire process of finding the populations by roughly four weeks,” he said.
“This increase in speed can make a big difference to certain patients, especially when there is nothing out there on the market that does a similar job. So, that’s another example where AI has helped us to make a difference, not just in our company, but beyond the enterprise and for patients.”
Get the morning’s top stories in your inbox each day with our Tech Today newsletter.
