About
Quantore is a large office-supplies wholesaler and cooperative operating in a highly competitive and digitizing market. Hixton helped Quantore explore how Generative AI can reduce manual work in high-pressure commercial processes.
Quantore used Generative AI to match customer product descriptions to catalog items, helping teams respond quicker and more accurately to complex tenders.
Implementation & results
The challenge
Quantore handles extensive tender requests where product lines are often provided without EAN numbers. Matching customer descriptions to the right catalog products was manual, slow and difficult to scale.
In a market with increasing cost pressure, this process needed to become faster and more reliable.
Hixton approach
Hixton built a proof of principle showing that AI can determine the best product match from customer descriptions. The approach considered practical business criteria so results aligned with how Quantore evaluates tenders.
- Python-based proof of principle with GPT-4o Mini
- Matching customer descriptions to Quantore source data
- Support for private label preference
- Scoring based on criteria like sustainability, quality and price
Impact
The proof of principle demonstrated a concrete path to faster tender preparation and more consistent product matching. This helps teams focus on value-added commercial decisions instead of manual lookup work.
The initiative gave Quantore a practical foundation for expanding AI-supported workflows.