{"slug":"kantar-worldpanel-fine-tunes-genai-models-on-databricks-to-generate-market-insight-training-data-faster","url":"https://findausecase.com/use-cases/kantar-worldpanel-fine-tunes-genai-models-on-databricks-to-generate-market-insight-training-data-faster","title":"Kantar Worldpanel fine-tunes GenAI models on Databricks to generate market-insight training data faster","description":"Kantar Worldpanel used the Databricks Data + AI Platform and MLflow to experiment with Llama, Mistral, GPT-4 and GPT-3.5 for a proof of concept linking receipt descriptions to product barcode names. GPT-4 produced the most accurate outputs (94%), which the team used to automatically generate a training dataset of about 120,000 receipt-to-barcode pairs in a couple of hours to fine-tune a smaller production model.","company":"Kantar Worldpanel","industry":"Technology & Software","aiCapabilities":["Generative AI"],"technology":["Databricks","MLflow","Unity Catalog","GPT-4","Llama","Mistral"],"deployment":"Public Cloud","problemStatement":"Kantar Worldpanel's legacy systems were inflexible, resource-intensive to maintain, and required specialized, outdated programming skillsets, limiting data democratization and experimentation with new AI-driven use cases.","solutionApproach":"Kantar Worldpanel used the Databricks Data + AI Platform and MLflow to manage the ML lifecycle, experimenting with Llama, Mistral, GPT-4 and GPT-3.5 to fine-tune a model linking receipt descriptions to product barcode names, downloading models via Databricks Marketplace, exploring Databricks AI Search for description comparisons, and using Unity Catalog to govern data sharing across teams.","businessValue":"Kantar Worldpanel automatically generated a training dataset of about 120,000 receipt-to-barcode description pairs at 94% accuracy in just a couple of hours, letting manual coding teams focus on discrepant results and freeing engineering resources for core development, while streamlining data scientists' workflows.","evidence":{"band":"high"},"sourceUrl":"https://www.databricks.com/customers/kantar-genai","dates":{"publishedAt":"2026-09-15T09:05:40.607Z","publishedAtSource":"pipeline","updatedAt":"2026-09-15T09:05:40.607Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/kantar-worldpanel-fine-tunes-genai-models-on-databricks-to-generate-market-insight-training-data-faster. Bulk republication requires permission."}