Lab Manager, a trusted resource for laboratory professionals worldwide, recently published a new article from Ryan Bernhardt, General Manager of Virscidian at Dotmatics, on why the standard definition of lab orchestration falls short.
Most orchestration conversations stop at the physical layer: instruments, robotics, and schedulers moving samples through a queue. Bernhardt argues that this leaves out the part that matters most. When experimental intent gets separated from experimental results, scientists end up as the connective tissue holding the lab together, hand-carrying context between systems that were never designed to talk to each other.
An orchestrated lab connects three domains at once:
Physical: instruments generating data, robotics executing workflows, scheduling platforms sequencing tasks and sample movement.
Logical: protocols, workflow design, and business rules. This is where scientific intent lives before it becomes action.
Digital: where results land alongside their context and stay connected across teams, across time, and across the inevitable turnover in tools and methods.
The cost of getting this wrong looks different depending on where you sit. Scientists spend a day or more writing up a single experiment because the data has to be pulled together by hand. Digitalization leads maintain vendor integrations that were never meant to interoperate, patching work that never finishes and scales badly. R&D directors watch technology investments underdeliver because each one lands on a fragmented foundation.
The piece also puts AI at the end of the sequence rather than the beginning. Agentic AI is only as trustworthy as the data beneath it, and fragmented, decontextualized data produces outputs no one can act on. In a Dotmatics study of scientists working with CRO-generated data, 80 percent said the workarounds required to turn data into meaningful output are hurting their work, and nearly 70 percent reported compromised decision-making as a result.
Bernhardt closes with four questions R&D leaders can use on their own environment today, including how much accumulated knowledge leaves when a scientist moves to a new program, and whether an AI initiative could genuinely reason across the full scientific record: the intent behind experiments, the decisions made along the way, and the context connecting them.
Read the full Lab Manager article for the full argument, including the reference architecture Bernhardt uses to separate today's integration-dependent lab from a connected one.