Linesight hosted its inaugural Life Sciences Capital Benchmarking and Project Performance Conference in Frankfurt at the end of March 2026.
The event brought together 48 participants and speakers from 19 life sciences organisations for a full-day discussion on the issues shaping capital delivery today.
The discussions focused on real-world and strategic challenges, moving beyond day-to-day detail to address the critical issues shaping capital delivery at a programme and enterprise level.
Conversations centred on how organisations can drive the greatest impact across their teams and portfolios - enhancing performance, enabling better decision-making, and delivering more effective outcomes on complex life sciences projects.
Speakers and attendees focused on what improves predictability in complex capital programmes, from early estimating and benchmarking through to integrated project controls, risk management, and the emerging role of AI in day-to-day workflows.
The tone was pragmatic, with many debates returning to the same common constraint: if the data is fragmented or inconsistent, decisions become harder to defend.
1. AI is an operating model shift, not a tool rollout
The AI session framed adoption as a structural change in how teams plan, govern, and deliver projects.
While leveraging AI for straightforward, everyday tasks offers valuable quick wins, it’s equally important to thoughtfully adapt existing workflows to ensure long-term positive impact and effective integration.
The ultimate goal in both cases is to implement AI to better support human-centered decisions, faster reporting cycles, and clearer accountability.
A consistent point was that capability and usage are not the same thing. AI tools can already support a wide range of project work, but most organisations are still using them in narrow, ad hoc ways, often without a clear view of risk, governance, or return on effort.
One line from Jonathan Phillips, Principal, Cost Management, captured the scale of the shift:
AI as a systematic shift is comparable to electricity or the internet in how it rewrites industries.
2. Risk is moving from register to quantified decision support
Risk management came through as a delivery discipline, not a compliance exercise. The discussion moved quickly from the mechanics of risk registers to how teams quantify uncertainty, connect schedule risk to cost exposure, and use iterative analysis to steer decisions earlier.
The Quantitative Risk Analysis (QRA) case study reinforced why iteration matters. The approach described was not a single workshop that produces a static output. It was a repeated cycle that tests assumptions, updates to risks as delivery conditions change, and supports targeted mitigation where it moves the needle most.
What also came through was the link between risk discipline and real delivery constraints. When programmes are challenged from design maturity gaps, vendor timelines, and supply chain volatility, teams need scenario thinking. It helps leaders separate structural risks from transient shocks, and avoid locking in decisions based on early optimism.
In pictures: Linesight's Life Sciences Capital Benchmarking and Project Performance Conference
3. Digitally supported project controls can become a single source of truth, from estimate to final account
A key takeaway from the Project Controls session was the critical role of the digital golden thread in elevating project controls from a reporting function to a fully integrated, data-driven capability.
Rather than relying on disconnected tools and fragmented data, organisations are embedding a single, trusted flow of information across cost, schedule, risk, and change - linking these controls directly to design, procurement, construction, and validation.
This digital golden thread enables near real-time visibility of impacts, consistent decision-making, and full traceability from requirement through to delivery, which is essential in the highly regulated life sciences environment.
The discussion made clear that establishing a digital golden thread within project controls is fundamental to driving greater certainty, control, and performance on complex capital projects.
4. Benchmarking works best when it drives decisions, not just validation
The overall discussion focused on benchmarking being treated as infrastructure, not a late stage sense check. The benchmarking panel focused on how benchmarking can improve early investment decisions, but only when leaders are clear on intent, context, and data quality.
Several examples highlighted where benchmarking breaks. Location factors, changing market conditions, and limited comparable projects can undermine confidence. If we don’t present the solutions to these challenges as part of the outputs, confidence can be undermined. This is why a range is presented rather than a definitive figure.
There was also strong interest in moving beyond static benchmarks to live performance insights. The point was not to collect more data for data’s sake. It was to make delivery data usable across the portfolio, so leaders can challenge schedules, contingency drawdown, and productivity assumptions using evidence from completed work.
- Treat AI as workflow design, and start with a small set of repeatable use cases that reduce manual reporting and improve decision quality.
- Fix data foundations first, including standards, governance, and consistent structures across cost, schedule, risk, and change.
- Quantify risk early and revisit it, so leaders can see how uncertainty is trending and where mitigation is changing the outcome.
- Use contingency actively, tied to known risks and tracked against progress, rather than holding it as an untouchable buffer.
- Build the digital golden thread, so the estimate, schedule, risk, and cost data connect cleanly from first assumptions to final account.
- Use benchmarking to shape choices, especially in early phases when site, utilities, supply chain, and delivery strategy decisions have the biggest long term cost and schedule impact.
Closing and next step
The discussions in Frankfurt made one thing clear. The organisations improving predictability are not chasing single tools or isolated fixes.
They are building a joined up system where benchmarking informs early choices, project controls connect the progress and warning signals across delivery; risk is quantified and revisited; and AI is introduced with strong data discipline and governance.
If you want to compare where your programme sits today, and what practical steps would reduce cost and schedule risk, Linesight can support with benchmarking, project controls, and risk capability built for life sciences delivery.
