Cytiva Exec: Biopharma's Bottleneck Is Execution, Not Science

More than a third of biopharma companies had not started, or were only beginning, their digital transformation journey in 2025 — the sharpest number in an argument from Nicolas Pivet, vice president and general manager of manufacturing and digital solutions at Cytiva, that the industry's real constraint is manufacturing execution, not scientific innovation.

RNA technologies and cell and gene therapies have advanced dramatically. Manufacturing has not kept pace. "Our ability to manufacture these therapies reliably, repeatedly, and at scale remains a major constraint," Pivet told GEN.

What's actually causing the bottlenecks?

The symptoms are familiar to any plant manager: paper batch records, disconnected LIMS deployments, fragmented automation architectures, isolated ERP systems, and data silos. Pivet's claim is that these are symptoms, not causes.

"The deeper issue is that many manufacturing environments evolved over decades and were never designed as connected systems," he said. "Digital capabilities were often layered on to solve local problems rather than integrated around end-to-end manufacturing performance. The result is an environment that is increasingly data-rich but insight-poor."

The practical consequence for operations teams: actionable insights arrive too late to affect outcomes. It's a visibility problem before it's an analytics problem.

Why observability comes before AI?

Pivet's prescription is sequenced and blunt. "You cannot control what you cannot see," he said. Early visibility into process deviations enables earlier intervention, which he calls the foundation of advanced manufacturing.

Only after connecting process, equipment, analytical, and workflow data should organizations layer on predictive analytics, digital twins, and AI. "Biopharma does not lack information. It lacks accessible, contextualized, and trusted information," he said.

The warning for R&D managers tempted to start with AI use cases: "AI does not fix weak foundations. It exposes them."

The barriers are organizational more than technical. Biomanufacturers often perceive change as riskier than the status quo, and trust — in Pivet's framing — is the rate-limiting step, earned only through reliable performance. The organizations moving fastest, he observed, are those demonstrating value on real manufacturing problems, validating outcomes, and scaling successful approaches gradually rather than attempting sweeping transformations.

How severe is the skills gap?

The workforce numbers are stark. The Global Biopharma Index reported that 29% of organizations experience severe shortages in digital and AI-related skills, according to Pivet. "Workforce readiness is one of the defining challenges of this decade," he said.

The roles changing span the full org chart: equipment operators, process engineers, data scientists, quality professionals, and executives. Pivet lists the skills that matter now — data literacy, systems thinking, automation fluency, model interpretation, and cross-functional collaboration — and notes staff are shifting from collecting information and reacting to events toward interpreting data and optimizing processes.

His five-year forecast: a wave of hybrid roles blending process and digital expertise, with titles such as digital process engineer, manufacturing data scientist, process intelligence specialist, and manufacturing systems architect.

He also points to immersive learning — mixed and augmented reality environments where operators and engineers troubleshoot realistic scenarios digitally before entering live GMP suites, where hands-on practice is constrained by production schedules and equipment availability. "Immersive learning may become as important to workforce development as flight simulators became to aviation training," he said.

Why do tech transfers keep failing?

Pivet identifies a failure mode that isn't technical or human: process understanding fails to transfer along with process instructions. Receiving organizations inherit specifications and operating procedures but not the context — why parameters matter, how the process behaves under changing conditions.

Digital twins narrow that gap by letting teams "test assumptions virtually before testing them physically," he said. The result: "uncertainty is removed rather than relocated."

Where should a limited budget go first?

If an organization can prioritize only one area, Pivet says choose digital infrastructure, because people and processes increasingly depend on information. Without trusted information flows, organizations struggle to scale knowledge, replicate success, or improve decision quality.

He predicts high-quality information flows will be a major determinant of manufacturing competitiveness in the coming decade, and that the best-positioned companies will pair AI initiatives with strong connected-data and observability foundations. "Biopharma does not have an imagination problem," he reiterates. "It has an execution problem." Closing that gap, in his view, is the industry's biggest opportunity to accelerate patient access and unlock the value of the science already in development.