How Microsoft AI Agents Support Cactus Life's Human-Led Work
The healthcare sector is experiencing a surge in complex clinical data, placing significant pressure on medical communications agencies.
To help alleviate the burden of manual data management, Cactus Life Sciences, a global agency, is modernising its operations by integrating Microsoft 365 and the Copilot AI platform. The organisation has embedded automation into its workflows while upholding the strict security standards required for sensitive pharmaceutical data.
Cactus Life Sciences employs over 350 professionals with advanced degrees in science. The agency converts its large volumes of clinical and scientific research into practical insights for healthcare professionals, payers and patients.
Previously, this work relied on manual and time-consuming processes. Key tasks such as document reviews and structured data extraction were well managed, but challenging to scale efficiently.
Microsoft introduced its 365 Copilot through a phased deployment to address these operational bottlenecks while maintaining data integrity.
“We didn't just want to automate tasks, we wanted to reimagine how work gets done,” says Odity Mukherjee, Lead AI Transformation, Cactus Life Sciences.
“With tools like Copilot, we've been able to rethink our workflows from the ground up, creating new efficiencies that free our teams to focus on what truly matters: delivering exceptional science to our clients.”
Security was the primary concern. As the agency often handles proprietary insights, it requires tools for trusted authentication and dedicated project environments to protect client data.
Deploying these tools within familiar Microsoft applications allowed rapid adoption within Cactus Life Sciences.
Custom agents to focus on specific processes
The rollout included more than 30 custom automation agents, which, instead of replacing entire roles, take on and automate smaller components of the overall function.
This enables automation to manage repetitive data processing, allowing Cactus Life’s writers to focus on the quality of their analysis.
The custom agents handle the retrieval and structuring of information from scientific literature. Additional automation tools support quality control tasks, such as checking abbreviations, ensuring uniform formatting and ensuring compliance with regulatory and publishing standards.
These agents can also extract and compare insights across multiple complex documents.
Cactus Life says that structured data extraction is now 35% to 50% faster than with previous workflows.
Project managers use the agents to summarise lengthy email threads and auto-generate task lists, while scientific writers process large volumes of articles more efficiently.
To manage the change, the company launched a Copilot Champions programme to identify internal AI advocates who would foster innovation and encourage experimentation.
Integration and education programmes
Supported by a dedicated training team, Cactus Life implemented modular education programmes to teach employees both the capabilities and limitations of AI tools. A centralised repository was also created to share effective prompts, agent ideas, and best practices.
“We built our Copilot Champions programme about six to eight months ago to create a vibrant community,” adds Odity.
“This helps drive experimentation and allows new ideas to emerge, whether it’s developing agents or finding new ways AI and automation can augment our work.”
The agency says it maintains a strict human-centric approach to technology. To address concerns about AI use, processes remain unchanged.
Technology is employed only to support human professionals, and all automated outputs are reviewed.
