United StatesFood and restaurants2025 – ongoingBoston Beer Company
Generative AI extraction of ingredient specification data
Boston Beer processed about 5,000 ingredient spec files in a week instead of six months
Problem
Boston Beer needed allergen, kosher, density and storage data from thousands of vendor ingredient specification files, mostly PDF and DOCX, to load into its master data system. Doing it by hand would have taken about six months.
Approach
Boston Beer worked with Syniti, whose generative AI reads the unstructured files, extracts the data points and produces files for loading, integrated with its master data tools and SAP. Extracted data is held for review and validation before it enters the system. The first wave followed a pilot and covered about 5,000 files and about 400 ingredients.
Results
Takeaway
The AI did the reading and staff kept the review step before data was loaded. The timing figure is Syniti's.
Categories
- Industry
- Food and restaurants
- Business function
- Production and qualityLogistics and supply chain
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Vendor solution
- Evidence
- Vendor case study
Sources
Source summaries
The article repeats Syniti's customer story. About 5,000 vendor spec files covering about 400 ingredients were processed, and a project that would normally take six months took one week. The figures come from the vendor.
Full details are in the original sources.
