In less than five years, Cube RM revolutionized tender management in the life sciences sector, attributing much of its success to Microsoft solutions. Using its tender management software, clients filter through and locate relevant tenders from over 40 countries. Combining a Power BI interface, Azure Cognitive Services, Azure OpenAI Service, data processing and visualization capabilities, Cube RM has helped its customers discover the most relevant tenders among thousands based on their product portfolio, which has resulted in less missed opportunities and an average revenue increase of 10 percent for customers.
Every year, public entities publish countless tenders for pharmaceutical products or medical equipment with detailed requirements. With unstructured, hard-to-find data, bidding companies struggle to discover and manage tenders that align with their portfolio. In 2018, Cube RM set out to change this, building a tender management software that consolidates the necessary tender data from over 40 countries in a customized format.
“Our strategy is to help companies secure more business from tender opportunities and enhance productivity in the process,” says Philip Kytinos, Chief Technology Officer at Cube RM. “With an advanced, user-friendly artificial intelligence (AI)-based software and Microsoft Power BI as our primary technology interface, we extract key details from tenders to deliver only the relevant ones to our clients. Consider a global pharma company that would otherwise need to manually crawl through multiple platforms to identify suitable tenders among thousands available. With our service, they save considerable time and are assured of the quality of data they receive.
Building on Azure, growing with Microsoft
From the get-go, Kytinos championed Microsoft’s extensive suite of products and services in his work and the company’s technical infrastructure. “Having one stack was very important for me, and that’s what I love the most about Microsoft’s ecosystem. Our developers are using Visual Studio, Azure DevOps for collaboration, GitHub for source code management, and SQL Server for data management. We also use Microsoft 365 for company-wide operations,” says Kytinos. “Everything runs on Azure, the integration is seamless, so adopting and using the different tools is extremely easy.”
Cube RM was also propelled into a larger business community through joining the Microsoft for Startups program, opening avenues for exploration and innovation. “Being part of the Microsoft ecosystem has been a huge advantage for us. The extensive community support means we can easily find new solutions to our challenges. We've established strong connections not just in Greece, but globally, and this network has been instrumental in our journey,” shares Kytinos. “Whenever we required support, whether it was architectural, cloud-specific, or commercial there was always someone available to assist. The level of collaboration has been unparalleled.”
One of the most significant advantages for Cube RM has been Azure Cognitive Services, which it uses to tag and recognize relevant tenders. “Azure Cognitive Services are user-friendly and straightforward for both me and my administrative team to manage,” notes Kytinos. Cube RM uses optical character recognition (OCR) in combination with translation services to collect tender data globally and assess source relevance in the native language.
Digging up the most relevant tenders among 1 million
In order to increase the accuracy of tender search, Cube RM sought to bring more granularity to its algorithm. “When we wanted to extract details such as molecules, understand the packet type, form, and dosage, Azure Cognitive Services couldn't deliver the level of accuracy we looked for,” recalls Kytinos. In exploring solutions, Cube RM combined the natural language processing (NLP) technologies and models that it has built with Azure OpenAI Service, and thus was able to bring much more precision to the process. “Azure OpenAI Service's use of large language models, which are pre-trained with vast amounts of data, eliminated the need for us to spend time training models from scratch."
By capturing and filtering data with more efficiency and customization, Cube RM has brought improved outcomes, visibility, and revenue to its clients. Importantly, the company also reduced “noise”, a critical metric, by filtering out the most relevant tenders to its clients, while the rest is considered noise. “For example, one client was manually reviewing and evaluating one million tenders annually. With our software, backed by Azure OpenAI Service, they only need to review 5,000 a year, which boils down to a 99 percent reduction in noise,” Kytinos remarks. “With better visibility on tenders that might have been missed or overlooked, our customers have also seen an average of 10 percent rise in revenue.”
“Previously, we faced a plethora of inconsistencies in our methods. Now, with a robust infrastructure, inconsistencies have been considerably reduced. Our processes remain consistent, ensuring high-quality data delivery,” explains Kytinos.
“One client was manually reviewing and evaluating one million tenders annually. With our software, backed by Azure OpenAI Service, they only need to review 5,000 a year, which boils down to a 99 percent reduction in noise.”
Philip Kytinos, Chief Technology Officer, Cube RM
Looking ahead, Cube RM aims to further enhance its offerings, with a primary focus on data architecture. “We hope to enhance our search capability with technologies like vector search, which enables semantic search,” Kytinos explains. “We're also contemplating the integration of Azure Data Lake, Synapse, Data Factory, and Analysis Services. These will be the cornerstones of our data architecture efforts as we aim to establish a data warehouse to facilitate our applications.”
In this journey, Cube RM continues to rely on its long-time partner. “With guidance from Microsoft experts, we achieved a 53 percent cloud cost reduction without compromising quality,” affirms Kytinos. “It's been an exhilarating partnership with Microsoft, and I'm hopeful it continues to flourish.”
“With guidance from Microsoft experts, we achieved a 53 percent cloud cost reduction without compromising quality. It's been an exhilarating partnership with Microsoft, and I'm hopeful it continues to flourish.”
Philip Kytinos, Chief Technology Officer, Cube RM
Follow Microsoft