AI case study

Kantar WorldpanelReceipt product matching

Manual coding bottlenecked receipt analysis. Synthetic data trained models to automatically link paper descriptions to barcodes.

Published|1 year ago

Key results

Creation Time
~2 hours
vs manual coding
Pairs Generated
120k
Model Accuracy
94%

Result highlights

Unlock 3 result highlights

The story

Context

A prominent international market research company analyzes consumer behavior data in the fast-moving consumer goods (FMCG) sector to provide actionable insights for manufacturers and retailers.

Challenge

Identifying purchased products required linking paper receipt descriptions to barcodes, a process bottlenecked by inflexible legacy systems and...

Solution
Unlock full story

Quotes

Unlock 10 more quotes

The company

Kantar Worldpanel logo

Kantar Worldpanel

kantar.com

Consumer behavior research and shopper data analytics platform.

IndustryConsumer Products
LocationLondon, England, UK
Employees10K-50K
Founded1992

The AI provider

Databricks is a Big Data company that offers a unified analytics platform for data science, engineering, and analytics teams.

IndustrySoftware & Platforms
LocationSan Francisco, California, United States
Employees10K-50K
Founded2013

Similar Case Studies

Related implementations across industries and use cases

109 AI case studies in Consumer Products

1,356 AI case studies in Product Engineering