Stop Asking for an AI strategy: 3 questions clinical leaders should ask instead

By Matthew Harrington, Global Head, Clinical Product, Syneos Health

Artificial intelligence (AI) now features prominently in boardroom and executive discussions across the life sciences industry. The attention is understandable: AI has the potential to fundamentally reshape how companies discover, develop and deliver new therapies. Its effect on both business performance and patient outcomes will be substantial and far-reaching.  

Yet alongside this ambition sits a more practical concern: can AI deliver returns commensurate with the investment and attention it has received? Findings from a PwC Global CEO Survey underscore this tension. While most organizations report operationalizing AI to some degree, 56% of CEOs say their AI investments have yet to generate measurable cost savings or revenue gains._¹

In this context, “What’s your AI strategy?” is a reasonable question, but perhaps the wrong one. It invites answers that focus on naming AI platforms, vendors and services rather than addressing the real issues that matter. This mirrors a point-in-time when organizational leaders once asked the question, “What’s your cloud strategy?” The underlying platform choice was far less consequential than the business outcomes it enabled: cost optimization, people productivity and job efficiency.

Three Questions for the AI-Curious

Think about AI as a tool for driving measurable impact across business operations. Rather than asking about an “AI strategy,” clinical leaders should focus on how this tool can be applied to improve decision-making, optimize performance, and accelerate execution. Ask yourself these three questions:

• What’s your prediction strategy? How are you applying AI to anticipate risks, forecast timelines and deliver insights that inform earlier, better decisions?
• What’s your optimization strategy? How are you improving resource allocation, study planning, forecasting accuracy and operational performance across programs?
• What’s your automation strategy? How are you using AI to eliminate manual effort, streamline repetitive tasks and generate structured content that accelerates workflows?

Questions like these will lead to a deeper, more revealing conversation and sharpen focus on your specific challenges and tech-enabled, but expert-led, options for solving them. 

For example, site contracting and budgeting remains a classic, costly technology challenge, characterized by fragmented data, manual workflows and non-integrated processes. AI offers a transformative path forward. By reducing complexity and improving predictability, it enables faster, smarter decisions across the entire drug lifecycle. 

To drive value in the site contracting and budgeting process, for example, Syneos Health experts designed and deployed AI solutions that improve both speed and quality across critical steps, including: 

• First-draft site contract authoring, where generative AI supports both straightforward template alignment and more complex harmonization based on historical agreements.
• Site contract and budget negotiations, where AI-powered review tools compare redlines, deploy acceptable fall-back language and contracting playbooks, and flag materials for expert human assessment and streamline approval workflows.
• Site payment processing, including invoicing and reconciliation, where AI now automates contract-to-event comparisons, identifies discrepancies and enables structured human review prior to payment release.
• Site payments forecasting, using predictive algorithms to improve projections of study budgets, with imputation logic, derived from both historical and forward-looking measures.

With a more integrated process, sponsors gain transparency through unified data across EDC, CTMS and contracting systems, along with predictive insight that strengthens forecasting confidence.

Beyond contracting and payments, other pragmatic, AI-enabled solutions focus on:

• Protocol optimization, using advanced analytics to design smarter, more feasible trials and reduce amendment risk. 
• Country and site mix optimization, identifying the optimal footprint for speed, quality and representativeness.
• Role-based dashboards, assistants and agents that generate real-time insights and prioritized actions to drive study efficiency.
• Structured Document & Medical Writing, where first drafts are deterministic instead of probabilistic, resulting in predictably accurate and high-quality content.
• Safety and pharmacovigilance case processing, automating intake and classification to improve throughput and compliance and quality.

Human expertise is non-negotiable. AI can generate plausible outputs, but domain experience is required to evaluate accuracy, context and risk. Judgment and accountability must remain in professionals’ hands. 

Building the ROI Case: Make it Measurable

One reason AI investments struggle to demonstrate return is that they are framed too broadly. “AI transformation” is difficult to quantify. Automation, prediction and optimization are not.

Consider these metric-driven use cases as you evaluate potential investments:

• Predictive applications of AI reduce risk and help avoid delays. They can help increase enrollment forecast accuracy; reduce variance between projected and actual activation timelines; lower the proportion of underperforming sites; decrease protocol amendments linked to feasibility gaps; and improve budget and timeline adherence.
• Optimization applications of AI improve execution and resource performance. They can help increase the percentage of studies delivered on time and on budget; reduce time to first patient in; improve site productivity and monitoring effectiveness; decrease variability across countries and regions; and increase portfolio-level resource utilization.
• Automation applications of AI minimize manual cost and cycle time. They can help shorten contracting and negotiation timelines; reduce manual processing hours per study; decrease error rates and duplicate payments; lower administrative cost per trial; and reduce rework across documentation and reconciliation workflows.

Framing AI investments this way strengthens the business case, but disciplined execution ultimately determines whether those projected gains are realized.

Avoiding Common AI Execution Traps

When PwC asked CEOs to pick the question that currently concerns them most, one rose to the top: “Are we transforming our business fast enough to keep up with technology, including AI?” 

Organizations often stumble not because they move too slowly, but because they overgeneralize AI’s capabilities and treat it as a strategy in itself rather than grounding it in defined prediction, optimization and automation priorities. Here are three more pitfalls executives must watch out for:

• Undermining human expertise. AI strengthens performance when it augments experienced teams, not when it replaces critical judgment and oversight.
• Overlooking near-term ROI. Balance long-term ambition with measurable short-term value. Define what success looks like now, next year and five years from now.
• Outpacing operational readiness. Technology can move quickly; organizations often cannot. Align deployment with governance, change management and workflow integration.

The “AI is everywhere” narrative creates urgency without always providing clarity. The way forward is to demystify AI by tying it to specific business problems. Ask what it could predict, how it could optimize performance and what it could automate within your existing workflows. Have candid conversations about risk, trade-offs and measurable return. Then anchor investment decisions to a defined roadmap—one that includes reskilling your workforce at scale and deliberately designing a technology environment built for integration, governance and sustained performance. 

If you’re ready to move beyond the question of “AI strategy” and focus instead on prediction, optimization and automation within your clinical development programs, talk to a Syneos Health expert. We help sponsors turn AI ambition into structured, accountable strategies that drive performance across the development lifecycle.