How MAG Unified Data with Snowflake to Improve Visibility and Reduce Costs

Beverage maker Mark Anthony Group (MAG) is using Snowflake to unify fragmented enterprise data and scale AI across its global operations, to achieve faster insight-to-action cycles, improved data accessibility, and reduced total cost of ownership across the business.

The initiative targets long-standing issues with disconnected systems across regions and functions, which limited visibility into inventory, demand, and customer behavior and slowed analytics, Snowflake said in  a blog post.

The company produces a portfolio of beverage brands, including White Claw Hard Seltzer and Mike’s Hard Lemonade, making timely access to accurate data critical for managing demand, distribution, and product performance at scale.

Building a Unified Data Foundation

The first step in the transformation was consolidating data from disparate systems into a single platform.

Sam Wong, senior director of data, analytics and AI at Mark Anthony Group, said: “Typically, a company has one marketing department, one finance, one operations and one HR. In our case, we have multiple finance and marketing departments, sales organizations and others that collaborate, each relying on different CRMs, ERPs and market data.”

MAG standardized how it shares data with external providers, replacing flat-file transfers with direct, real-time data integration.

This shift reduced total cost of ownership, improved reliability, and made it easier to identify and resolve data quality issues at the source. It also removed common issues such as file format inconsistencies and manual handoffs.

As a result, teams gained faster access to trusted data, enabling more timely and informed decision-making across supply chain, merchandising, and finance.

Customizing Snowflake Intelligence for Enterprise Use

A key challenge in scaling AI was ensuring the system could interpret business-specific language consistently across regions and functions.

MAG addressed this by developing a semantic layer that standardizes internal terminology and aligns data definitions across systems. By integrating its business glossary and data catalog into Snowflake, the organization is working to ensure that queries return consistent and meaningful results, regardless of how users phrase them.

With its data foundation in place, MAG began piloting a customized implementation of Snowflake Intelligence tailored to its enterprise environment.

The company developed a global application that acts as a wrapper around the Snowflake Intelligence engine. Currently in pilot, with a broader rollout underway, the application is designed to give business users direct access to data through a simplified interface.

Using text-to-SQL capabilities, users can query data in plain English, and in some cases via voice, without needing to understand underlying data structures or write queries. This shifts access to insights closer to business teams while reducing dependency on technical resources.

Key capabilities of MAG’s implementation include:

• Explainability and context: The system provides descriptions of datasets and underlying logic, helping users understand not just results but how they are derived.
• Mobile and voice access: The application is web-enabled and mobile-friendly, allowing users to access and query data from different devices, including on the go.
• Collaboration through existing tools: Rather than introducing a separate interface, MAG is integrating the application into Microsoft Teams, enabling users to access insights within existing workflows.

Building on these capabilities, MAG also focused on moving from insight to action more quickly.

With integrated data and AI capabilities, the organization began automating responses to key business signals, such as adjusting inventory strategies or responding to supply chain disruptions.

This reduced manual intervention and improved the speed and consistency of decision execution across the organization.

Translating Data and AI into Business Impact

MAG has already begun using AI and machine learning models to support sales performance analysis and enrich existing datasets with additional attributes. These capabilities are expected to reduce the time required to generate insights and act on them, improving responsiveness across business units.

“I now have quicker access to all my data, which will help me across so many different initiatives,” Wong said. “What are some new revenue opportunities? What are some operational inefficiencies we can target? How do I improve product quality now that I have greater insights into it? It’s going to trigger a new utilization of data that we haven’t had before,” he added. “That’s going to fundamentally change our business processes and workflows, bringing to life a vision of agentic enterprise.”

Over time, this approach is expected to help MAG find new revenue opportunities, reduce inefficiencies, and make faster, more informed decisions across the business.