The bioprocessing industry has a digital execution problem

Why the industry’s biggest challenge is no longer innovation, it’s digital execution

Biopharma has never had more scientific potential. New modalities are expanding what is possible in medicine, and manufacturers have access to increasingly advanced science and equipment, from automated bioreactors and single-use systems to predictive analytics and other digital solutions, like automation and artificial intelligence (AI).

Despite all this, adoption lags far behind other industries. Surveys have revealed that, even in 2025, over a third of biopharma companies are still in “early exploration” or have not started their digital transformations (1). And, even if an organization commits to digital adoption, implementation often becomes fragmented by business priorities, bioprocessing silos, and workflow stages. Indeed, the industry still struggles to execute consistently and at scale.

In a recent discussion with Ashley Howard, Senior Product Director for Automation and Digital at Cytiva, one theme came through: “The industry is not limited by innovation; it’s limited by execution,” she says. “Execution breaks down not at the point of strategy, but when organizations decide whether to change.”

Digital execution is the bottleneck at the heart of modern bioprocessing and it’s becoming harder for biopharma to ignore. As therapies become more complex and increasingly personalized, manufacturing timelines are being squeezed—and the pressure to reduce time-to-market will only intensify. Delays driven by inefficiency or slow decision-making don’t just increase cost, they impact competitiveness and, more importantly, delay patient therapies. “It’s the ability to embrace the capabilities into day-to-day operations that will begin to differentiate organizations,” Ashley says.

When you look at day-to-day manufacturing processes, the same challenges persist—limited visibility, fragmented data collection and processing, and a surprising level of manual decision-making. That reliance on human intervention isn’t just by design; it reflects a lack of confidence in the systems themselves.

Improving how processes are designed and executed is no longer optional, it’s critical to delivering therapies faster and more predictably.

Scaling digital programs: a system that is data-rich, but insight-poor

In many facilities, process data still sits across numerous disconnected equipment and systems. Advanced sensors are often treated as optional add-ons rather than part of the initial design. Automation is implemented to meet immediate needs, rather than enable long-term process intelligence.

The result is an environment that is data-rich, but insight-poor. Many organizations still operate within “islands of automation with limited connection between them,” Ashley says. Data exists, but it doesn’t flow in a way that consistently supports more confident decision-making and faster drug production.

As processes become more complex, that gap becomes harder to manage. Small changes in materials, conditions, or equipment behavior can create a downstream impact, making it harder to detect deviations and respond before variability turns into a risk.

The problem is not lack of digital technology—it’s when and how to use it

Process analytical technology (PAT), advanced sensors, hybrid modeling, digital twins, and predictive analytics are not new concepts. The value is understood, and in many cases, the pathway is clear. The challenge is when and how these capabilities get applied.

In practice, digital and automation investments are often introduced too late, after processes are locked, validation pathways are defined, and change becomes expensive. What should have been foundational becomes retrofit. And by that point, the opportunity is already reduced.

Organizations often know how to improve performance, but they still do not act. As Ashley put it, “The cost and risk of change often outweigh the perceived benefit.” This is not a technology problem. It is an operating model problem.

Overcoming adoption barriers to automation and digital technologies

Biopharma has historically been able to absorb inefficiency in ways that other industries simply could not. High margins and strong portfolios meant manufacturing did not need to operate at the same level of efficiency or digital maturity. “For years, the strategic investment was in R&D, not manufacturing scalability,” Ashley says.

That operational mentality is beginning to change, mainly out of necessity to succeed in an evolving biomanufacturing landscape, where new modalities, faster timelines, and flexible bioprocessing are key elements to getting drugs to market faster. As pipelines evolve and competition increases, manufacturing performance—increasingly defined by speed, scalability, and reliability—is becoming a point of differentiation. The conditions that allow inefficiencies to persist will necessarily slowly disappear. And this is where the existing model demonstrates its limitations.

Without integrated data, teams spend more time retrieving information than acting on it. Without observability, variability is detected too late. Without interoperability, systems remain fragmented and without a clear path to implementation, allowing digital initiatives to stall between ambition and execution.

In simple terms, the industry isn’t short of ideas, it’s short of connected capability.

Learning from other industries, without copying them

Other manufacturing sectors are further ahead in this journey and beyond the issue of adoption and execution. In industries like semiconductors and automotive, digital wasn’t layered on, it was designed in. Sensing, automation, and data and process control operate as an integrated system, not a collection of add-ons. Biopharma operates under different constraints, but the comparison is useful. As Ashley observed, “When you walk into the factories of different industries, you can quickly see who’s doing this better.”

Closing that gap isn’t just down to manufacturers. Equipment providers and technology partners, including companies like Cytiva and its peers, are part of the equation as well. In many cases, the industry hasn’t made it easy to adopt digital capabilities. Solutions are often positioned around what is technically possible, rather than what is realistically achievable in practice.

Biopharma needs to do better. We can move forward by simplifying adoption, remaining genuinely vendor-agnostic, and providing clarity on not only the value of embracing automation, but how it can be implemented. In other words, the industry doesn’t lack ambition, it lacks a repeatable path to execution.

Enhanced process observability isn’t the issue, adoption is the challenge

For years, the industry has talked about moving beyond basic monitoring toward real-time monitoring of critical parameters using sensors. The US Food and Drug Administration’s process analytical technology (PAT) guidance from 2004 (2) was meant to encourage organizations to adopt real-time advanced sensors, but this has been slower than hoped, particularly at commercial manufacturing scale (3).

In practice, the challenge is simply embedding core measurement and monitoring capabilities consistently across processes. As Ashley put it, “While widely discussed, PAT is not consistently embedded in practice. It’s far harder to implement and costly at scale than many assume.”

Even core measurements including pH, temperature, dissolved oxygen, and flow are not always standardized or scalable across facilities, not because the technology does not exist, but because the practical barriers are real: cost, operational burden, contamination risk, and regulatory overhead.

Adding sensors doesn’t just create more data—it creates more complexity, validation, and responsibility. And without a clear business case, the trade-off often doesn’t stack up. While the ambition is to move towards predictive, model-driven processes, many organizations are still working through the fundamentals of implementing real-time monitoring.

The next competitive advantage will come from integrated intelligence

The future of bioprocessing will not be defined by individual technologies. It will be defined by how well sensing, automation, data, modeling, and workflows come together. Despite the growing focus on AI, our future as an industry doesn’t start with AI, it starts at the ground level.

“You can’t just throw AI at it—you have to build the foundations first,” Ashley says.

That means connected systems, usable and actionable data, and having access to scalable architecture. It also means taking a more practical approach to change: “Stop trying to design the perfect system—just get started,” she advises.

The shift from ambition to action is what unlocks progress.

From technical possibility to operational reality

The industry no longer needs to prove that digital can create value. That’s already been established. The real challenge is unlocking that value consistently, and that requires a shift in mindset. Automation and digital cannot remain a later-stage enhancement. Likewise, process observability cannot be optional. Data cannot stay trapped in systems that limit its usefulness, and digital transformation cannot rely on large-scale programs that never progress beyond planning stages.

In short, the risk isn’t doing too much, it’s not doing anything at all. Says Ashley, “We often start with the big vision—and never actually get started.”

The organizations that move forward won’t be those with the most technology. They will be the ones that design for integration early, reduce the cost of change, and act on insights faster. In modern bioprocessing, the biggest limitation is no longer what the industry can imagine, it’s what the industry can consistently execute. And right now, that execution gap remains the industry’s biggest constraint.

Future focus: moving from innovation to digital execution

Closing the execution gap requires a deeper look across the lifecycle—from how data is generated, to how decisions are made and how systems scale. Increasing observability in bioprocessing, steering a practical course toward digital transformation, ensuring reliability in process performance, and shifting toward model-driven manufacturing are four key initiatives that will help many biopharma organizations move from idea to true digital execution.