Völur Transforms Customer Data Securely and Effectively with Snowflake

The global meat processing industry is responsible for a significant amount of the world’s carbon emissions, and ethical concerns to reduce waste are putting considerable pressure on processors to optimize their production. But meat processing is a traditional industry. Data sits in different places, and most organizations use human expertise to make production designs. These factors mean that AI has not been used to enhance decision-making at scale — until now. 

Norwegian startup Völur was created with one specific mission: to help meat processors make better decisions that lead to the optimal use of all animals. The company uses the power of data and AI, in combination with deep domain knowledge, to enhance production planning decisions and reduce inventory costs and overall waste in the industry.

“Coming from the meat processing industry, Völur’s co-founder saw the potential of using data and AI to empower meat processors to make optimal decisions,” says Rebecca Wiborg Seyfarth, VP of Engineering at Völur. “The data tools that meat processors currently use are mostly for general-purpose scheduling. In an industry with a complex value chain, nuance can make a big difference.”

According to Völur’s team, deep industry knowledge is key to making AI algorithms meaningful for their customers. “The meat industry is still very reliant on individuals’ knowledge when it comes to inventory planning, cutting processes or demand forecasting,” Seyfarth says. “Not every meat processor can use all the data they collect. That said, most processors know they need to make better use of their data, which presents a significant opportunity for Völur.”

As Völur onboarded more customers, it needed a scalable platform that would simplify data application development and offer the necessary tools to rapidly respond to market changes. Völur’s objective was to develop an AI platform supported by a proposed data application, requiring rapid data processing to feed its AI model and deliver quick time to insight. Using Microsoft Azure to host its solution, along with third-party tools like dbt for data engineering and Terraform for infrastructure, Völur also needed a platform that could offer seamless third-party integrations with almost any vendor.

Völur built its data platform on Snowflake, forming the heart of its data ecosystem. It covers everything from various animal characteristics and customers’ business constraints to product catalog pricing and demand forecasts.

For Völur’s customers, this means rapid but protected time to insight based on daily production plans, supply forecasts and market demand data. From the moment Völur’s customer data hits its web application UI, several secure data pipelines are triggered and executed inside Snowflake, feeding Völur’s AI model. The AI model’s output is delivered seconds later in the form of unique insights based on millions of variables that help support factory-level decision-making.

This fast-paced string of data events relies on seamless, secure data sharing capabilities — which Völur now has in droves. Seyfarth explains how the company receives shared data with one of its customers quickly and securely: “As a Snowflake user, one of our customers had the advantage that we could easily set up an account in the same data center region to begin data sharing. The company then provided cloned data with real-time updates. Everything is backed up, and we use Dynamic Data Masking to protect sensitive market pricing information. Snowflake allows us to focus on expanding our platform without worrying about the complexities of data extraction.” 

Völur’s application in Snowflake delivers optimized data pipelines in just minutes. Seyfarth sees this as a major advancement for meat processors, highlighting Snowflake's ability to enhance traditional data handling methods and cultures. Seyfarth says, “The key to succeeding in this industry is to integrate with how our customers work rather than force any significant changes.”

From its early development stages when Völur had yet to build its data application, the company took advantage of the Powered by Snowflake Startup Program — a toolbox of resources designed to help companies build and grow their data applications on Snowflake. This included a dedicated team of Snowflake data engineers and startup experts who provided timely support and advice during development and deployment.

“We used Snowflake’s startup support a lot in the beginning,” says Seyfarth. “As we scaled our AI and data teams, we relied heavily on Snowflake’s data engineering support. Snowflake helped us achieve critical milestones and was very fast at providing answers — something that’s vitally important for startups.”

Looking ahead, Völur intends to expand its customer base and incorporate more data from current customers’ factories into its platform. The company believes this will be streamlined by Snowflake’s ability to seamlessly create new customer data accounts with reliable data protection and security protocols.