Dassault Systems

Dr. Harsh Shah

A clinical technology project can meet its launch date while leaving the underlying work almost unchanged. Staff may still reconcile information manually, study teams may still wait for decisions, and participants may still face confusing instructions. The system is available, but the expected benefit has yet to reach the people doing the work.

This is the last mile of life sciences innovation: translating a technical capability into a dependable way of working. For leaders responsible for research and clinical development, the practical question is whether technology helps teams make better decisions and execute studies with less avoidable friction. Answering it requires attention to workflows, accountability and evidence long after implementation.

Define the improvement before choosing the technology

An investment should begin with a specific operational problem. “Improve efficiency” is too broad to guide implementation. A more useful objective might be to shorten the time between identifying a data discrepancy and resolving it, or to reduce the number of times a site must enter the same information.

Each objective needs a baseline, a responsible owner and a measure that reflects the work itself. Teams should also identify constraints, such as staffing, study complexity or an external dependency. This prevents a technology project from being held responsible for every delay while still making its contribution open to scrutiny.

Clear objectives give implementation teams a basis for choosing which features matter, which processes need redesign and where limited resources should go first.

Follow the work across organizational boundaries

A process rarely belongs to one department from beginning to end. A study update may pass through a site coordinator, a monitoring team, a data manager and a sponsor decision maker. Improving one task can have little effect if information then waits at the next handoff.

Before deployment, teams should map a representative workflow from the initial request to the final action. Where is information created? Who checks it? What triggers a decision? Where do people copy data, seek clarification or maintain separate spreadsheets?

This exercise should include the people who perform those tasks. Their workarounds often reveal a missing requirement or an unclear responsibility. Removing an unnecessary step may produce more benefit than digitizing it. A useful implementation plan therefore specifies both the system changes and the changes in how people coordinate their work.

Make routine use practical

Training attendance and user logins provide limited evidence of meaningful adoption. A person can complete training yet remain unable to handle a difficult case. A frequently accessed platform can coexist with a parallel manual process that carries the real workload.

Support should reflect the decisions users must make. A site coordinator may need help managing an incomplete submission; a study leader may need guidance on interpreting an exception. Short, relevant instructions and an accessible escalation route can make a new process easier to sustain.

Leaders should examine whether the intended workflow is being followed and ask why exceptions occur. Some exceptions indicate an appropriate response to local circumstances. Others expose a design problem. Treating every workaround as resistance risks missing an opportunity to improve the process.

Make information usable for decisions

Connecting systems is only part of integration. Teams also need agreement on what information means, when it becomes available and who can act on it. A shared dashboard offers little help if departments interpret a status differently or cannot identify the source of a reported issue.

Data definitions, ownership and escalation rules should be established alongside technical connections. Decision makers need enough context to distinguish a genuine problem from an incomplete update. Access should also match the user's responsibilities and the applicable controls.

For AI applications, the same discipline matters. Teams should define the intended task, examine the suitability of the inputs and establish how outputs will be reviewed. An apparently useful recommendation should not acquire authority simply because it appears in an automated workflow.

Measure progress without overstating impact

Evaluation should connect usage to operational results. For example, a team might examine whether a new query management process reduces resolution time while maintaining review quality. Counting activity alone cannot establish that the process improved.

Comparisons also need context. Differences in protocol complexity, site experience or staffing can influence results. Where feasible, teams can test a defined workflow in a limited setting before expanding it, recording what changed and what remained constant.

Patient impact requires particular care. Reducing administrative effort may support a better participant experience, but it does not by itself demonstrate improved clinical outcomes. Claims should match the evidence available. Measures such as fewer repeated requests or clearer participant communications can be valuable without being presented as proof of a treatment benefit.

Keep ownership beyond launch

The operational owner should remain accountable after the project team steps back. Regular reviews can bring together users, process owners and technical colleagues to examine exceptions, check results and decide which adjustments are worthwhile.

These reviews should produce decisions rather than another reporting obligation. If a feature is unused, investigate whether it addresses a real need. If a process creates extra work, revise it. If an improvement holds up under review, document the conditions that made it successful before extending it elsewhere.

Life sciences innovation reaches its last mile when a useful capability becomes reliable practice. That requires a clear problem, workable processes and honest measurement. Giving this work sustained attention helps organizations turn investment into improvements that research teams, sites and participants can actually experience.