Almirall unlocks decades of R&D data in seconds with Azure OpenAI in Foundry Models

At global pharmaceutical company Almirall, scientists spent hours manually digging through 50 years’ worth of data in 400,000 documents, slowing discovery and risking duplicated work that delayed treatment development.

The dermatology-focused company used Azure OpenAI in Foundry Models, Azure AI Search, and Azure Databricks to create a custom assistant that helps scientists instantly search R&D documents in three different languages and automate routine tasks.

Almirall’s scientists use the assistant to access decades of data in seconds versus hours or days, boosting productivity, reducing R&D cycles, and minimizing duplication.

Decades of data hold the key to future breakthroughs  

While skin is our main protective barrier, for patients with chronic skin conditions, it can become a barrier to living their best life. With more than 80 years of experience as a leading pharmaceutical company specializing in medical dermatology, Almirall understands that while its innovative treatments offer physical relief, real transformation occurs within. By addressing the unmet medical needs of patients, Almirall empowers them to face their day with regained confidence and improved psychological wellbeing.     

  “Making new discoveries is key,” says Francesc Fernandez, Senior Director of R&D Data Science at Almirall. “A new discovery later becomes a potential treatment for our patients, improving their quality of life and even potentially healing the disease.”  

  With more than 50 years of data, Almirall had a rich foundation to accelerate drug development, reduce attrition, and build on past research. But accessing that knowledge was slow and inefficient. Researchers spent hours, even days, searching scattered files and outdated systems. Valuable insights were often buried, lost with retiring colleagues, or duplicated unknowingly, wasting time and resources.  

Engineering an AI-powered path to discovery  

Almirall needed a faster, more accurate way for its R&D teams to access data. Microsoft provided a comprehensive suite of services that easily interoperated with Almirall’s systems. The Microsoft Industry Solutions Delivery (ISD) team partnered closely with Almirall’s scientists to co-design and deliver a custom, outcome-driven solution aligned with their needs to enable innovation at scale.

  “Our people are scientists. They have a specific language and know-how. Microsoft Industry Solutions Delivery had the Microsoft knowledge and expertise in pharma to speak their language. Together, we could quickly develop a solution that provides immediate value for our R&D team,” says Francesc Catalan, the CIO and Corporate Information Technology Director at Almirall.  

  Almirall built a custom assistant, using Microsoft Azure OpenAI in Foundry Models, Azure AI Search, and Azure Databricks to help researchers quickly find answers across 400,000 documents in English, Spanish, and Catalan. R&D team members refined it by testing prompts and validating results. Now scientists can locate past experiments and scientific documents in seconds, freeing up time for deeper, more innovative work. “The best value for scientists is to have them think, have ideas, challenge things, “says Fernandez. “When they’re stuck searching for documents, that’s not a good use of their expertise.”  

“Our people are scientists. They have a specific language and know-how. Microsoft Industry Solutions Delivery had the Microsoft knowledge and expertise in pharma to speak their language. Together, we could quickly develop a solution that provides immediate value for our R&D team.”

Francesc Catalan, CIO and Corporate Information Technology Director, Almirall

Since then, Almirall has rolled out the assistant to teams in the R&D department, with the goal of cost-effectively expanding the project’s architecture and internal know-how into other departments. “This has created the foundation where we can grow and use it for many different things,” says Catalan. “We won’t spend as much time or money in the future because the foundations are already created.”

Faster answers, faster treatments—and improved lives  

Users are already seeing productivity gains and have reported finding accurate answers to their queries around 80% of the time.     

“When we had an issue with one compound, I asked the assistant if we’d seen a similar issue with another compound. It found examples in minutes,” says Montse Erra, NCE Discovery & Early Development Director at Almirall. “Now I can spend my time on what truly matters, not half a day searching.”  

  In addition to the custom assistant that was developed, Almirall is also using Microsoft 365 Copilot to streamline repetitive and administrative tasks. Specifically, it assists with updating and managing the internal R&D governance handbook, summarizing and updating documents, and building PowerPoint presentations. This frees up time for strategic thinking and intensive problem solving that leads to innovation.   

  Ultimately, Almirall aims to build more efficient research processes that help it accelerate drug development and, therefore, drive real healing for people living with skin conditions. “This project fits right in with our values,“ says Fernandez. ”If we have a better use of existing data, we can make faster decisions, get to market earlier, and help patients sooner.”   

“This project fits right in with our values. If we have a better use of existing data, we can make faster decisions, get to market earlier, and help patients sooner.”

Francesc Fernandez, Senior Director, R&D Data Science, Almirall