United StatesConsumer goods2025 – ongoingMattel
Gemini-based classification of customer feedback
Mattel cut customer-feedback analysis from a month to a minute with Gemini and BigQuery
Problem
Mattel collects feedback from reviews, social media and direct customer contacts, often in datasets of tens of millions of points. Analysis took about a month, which slowed the discovery of product issues.
Approach
Mattel's data team built a feedback classification system on Google Cloud. BigQuery stores and processes the data, and Vertex AI with Gemini models trains a model that classifies feedback types and sentiment. The work runs from ingestion to AI inference in one pipeline.
Results
Takeaway
Mattel reported the gains in processing capacity, time and cost. Google Cloud published the account, and the article gives no method for the savings figure.
Categories
- Size
- Large enterprise
- Industry
- Consumer goods
- Business function
- Production and qualityInternal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Built in-house
- Evidence
- Vendor case study
Sources
Source summaries
Google Cloud's customer story describes Mattel's system for classifying feedback with BigQuery, Vertex AI and Gemini. It reports analysis time falling from a month to a minute, a 100x gain in processing capacity and about $1 million in savings.
Full details are in the original sources.