As workforce turnover reshapes the industry, how can organisations preserve the knowledge that drives scientific progress?

The biopharma industry continues to endure some of the most turbulent and difficult market conditions in a decade. Influenced by geopolitical, technological and financial pressures, the landscape has become one of extremes. For early-stage companies, the funding environment is a desert. Rainfall (VC investment) is hard won and appears only infrequently within oases; fewer, larger deals that tend to concentrate within later-stage companies who have shorter paths to clinic with lower risk roads to return. This pattern also extends to mergers and acquisitions, where preclinical valuations have declined significantly and growth has been primarily driven by marketed assets.1 Big pharma is experiencing the same extremes, albeit differently. Here, a macroeconomic drought on one hand has met a GLP-1 driven deluge on the other, and the outcome has been a shift in strategic direction and R&D priorities.

Within such a turbulent environment, staffing changes have been an essential conversation. While turnover is not inherently a bad thing, it would be insensitive not to acknowledge that in recent years, the majority of staff turnover within the biopharma industry has been involuntary, and the cost at both an individual and organisational level enormous.

A staff turnover metric is often used as a proxy measure for the cultural health of a company, but assuming that low turnover is always a positive signal can be misleading. On an individual level, voluntary turnover can indicate career development and new opportunities for growth. For the organisation, it can be an opportunity to refresh talent and address missing skill sets, or introduce new ideas into the business. Finally, at a community level, the frequent movement of skilled staff between companies can accelerate innovation within a collective ecosystem.2 Yet even in these positive scenarios, there’s a cost to change beyond the direct financial impact of recruitment.

Biotech acquisitions are a perfect example of how, even in positive scenarios, there’s a cost to change. The innovation and intellectual property that contributes to a biotech’s valuation is usually generated by inventors within the company, some of whom will inevitably move on post-acquisition. In their 2025 study, Verniger and Riccaboni studied data from 15,318 inventors over 1,375 acquisitions, and found that acquisitions typically result in a 13.5% drop in inventor retention, and perhaps more importantly, a 35% drop in citation-weighted patent productivity from those inventors who remain with the organisation.3 There can be multiple reasons for this, including team restructures, changes in resource allocation or shifts in strategic priorities, but regardless of the cause, this must certainly be accounted as a cost of change.

The loss of core institutional knowledge (the collective expertise that’s specific to an organisation, or even a role) is another such cost. Explicit institutional knowledge is relatively easy to document in standard operating procedures (SOPs), employee handbooks and company databases, but there’s a wealth of tacit knowledge in any company that is rarely properly documented. In a biotech setting, this kind of knowledge might include the original reference for a DNA sequence that’s been used in a particular plasmid design for years, ‘the knack’ of getting a complicated machine to behave properly, the reason why a protocol is always run one way and not another, or why an ‘obvious’ experiment isn’t worth doing, because although it’s never been reported, someone remembers that it was done once (or three times) before and didn’t work.

Taken in isolation, these examples may be trivial. Collectively, they pose a significant operational risk – one that compounds with scale. A major workforce survey found that 42% of role-specific knowledge is unique to the individual doing the job, and that 66% of employees have some level of difficulty accessing the information they need to work effectively.4 In most industries, including biopharma, this translates to lost time and duplicated effort, both of which have financial implications.

The International Data Corporation estimates that a company of 1,000 knowledge workers loses up to $2.5m per year to time spent searching for information that is not centrally stored, and a further $5m per year duplicating work that already exists somewhere in the organisation.5 In a scientific context, those duplicated efforts are failed experiments, additional reagent costs and months of research time consumed on questions that have already been answered. This contributes to a pattern of avoidable waste that has been estimated to consume up to 85% of biomedical research investment.6

The loss of institutional knowledge is one of the problems that ‘digital transformation’ promised to solve. Yet in many cases, the solutions companies implement to improve communication and efficiency ultimately end up doing the opposite. There’s too much data, spread across too many tools. Is that document on the local drive, on One Drive, or saved as a Google Doc? Was it shared on Slack, on Teams or via email? Is it uploaded to the project management portal or Dropbox? And who can say with any degree of certainty which the latest version is? This is another superficial example, but reference to the figures above demonstrates how costly this can become when it happens at scale. The reality is that most biopharma companies are dealing concurrently with two separate, but related, problems, both of which become more apparent during times of change. First, how to capture and document tacit knowledge that may not usually be recorded, so that the company’s science and collective experience is protected. And secondly, how to store it in such a way that it can be easily found and retrieved.

It’s time to re-examine the idea of digitisation to improve scientific resilience, but this time with a twin objective: to amplify tacit knowledge, while simultaneously cutting through the resulting noise of too much information.

In order to be successful, the new generation of laboratory management software must be designed to weave all the different strands of a typical laboratory workflow together, rather than adding more technology to an already overcomplicated stack. This means connecting the in silico design of a construct to the protocol used to make it, the samples produced in the process and the data generated along the way. In short, these solutions need to be able to build and preserve a robust scientific record that includes all the experimental context and entwined relationships between samples that might otherwise never have been captured.

The first test of any such system is simple. Can a new team member find out, in minutes rather than days, why the lab uses this vector and not that one? Can a senior executive see, at a glance, which projects are progressing as planned, and how resources are allocated? Can a company preparing for regulatory submission easily demonstrate a complete, unbroken chain of evidence from sequence design, plasmid and cell line to viral vector and associated data? In a well-implemented system, the answer to all three should be yes.

In a highly regulated environment like biopharma, staffing changes create specific compliance vulnerabilities: keeping records contemporaneous with physical operations when headcount is low, ensuring that training records don’t fall out of alignment with electronic signature authorisation and ensuring that adherence to SOPs doesn’t drift once the original author moves on. A connected digital record helps mitigate these risks by making audit-ready documentation a by-product of the lab’s day-to-day operations.

The second test of a new digital solution is whether scientists actually use it. If software is difficult to navigate, clunky to operate, slows the pace of the lab or requires repetitive manual data entry, it won’t deliver on its promised solution. The real advance in second-generation software is usability; that is, becoming a technology that scientists enjoy using, that adds value to their work and frees up their time, because no matter how good the software is, it’s the people in the team who will both deliver and derive the value from it.

When this sort of resilience is baked into the scientific workflow, change becomes easier to manage. Team members moving on to new opportunities don’t take years of unwritten knowledge with them, or spend their last days in the lab frantically scribbling down details of where their latest results can be found and what they mean. New team members inherit a full scientific history, and a clear understanding of the direction of travel, rather than a folder of disconnected documents and a vague reference to an overstuffed freezer drawer. Their learning curve becomes shallower and the time it takes to get up to speed shortens.

These are difficult times for the biopharma workforce. For those scientists navigating a job market that has rarely been harder, the professional and personal cost has been significant. But these are exactly the moments when small cracks in the scientific record can start to widen. Those scientists, teams and leaders who now have firsthand experience of what it costs when institutional knowledge walks out the door, are precisely the people best placed to build a more resilient future. Markets evolve and funding cycles turn, but good science holds firm. A robust digital infrastructure should be considered an essential element of a company’s scientific foundations to make sure this holds true.

References: 1. Visit: bcg.com/publications/2026/reimagining-businessmodels-biopharma-trends 2. Visit: jstor.org/stable/144358 3. Visit: sciencedirect.com/science/article/pii/ S0148296325003960?via%3Dihub 4. Visit: panopto.docsend.com/view/pzjpwm6 5. Visit: computhink.com/wp-content/uploads/2015/10/ 6. Visit: thelancet.com/journals/lancet/article/PIIS0140-6736(09)60329-9/abstract

Sophie Lutter PhD, is head of Marketing at Lab Thread, an integrated laboratory management platform for life sciences R&D. A developmental biologist by training, she has spent 15 years working at the intersection of science and strategy – most recently as Scientific Strategy Director at That's Nice, a US life sciences marketing agency, and previously with the founding commercial team at OXGENE, which was later acquired by WuXi AppTec. She writes about the systems, structures and cultural habits that determine whether scientific knowledge endures – or walks out the door.