TELECY Presents Causal Inference for TV CM Measurement at YANS2026
The booth's most-discussed topic was how to estimate TV CM effects, a problem researchers noted is harder than digital ad measurement, and TELECY's answer is a patented difference-in-differences method.
Reporting from 1 source: ASCII.jp.
TELECY, a CARTA HOLDINGS group company, presented its data science work at YANS2026 in Sendai. Visitors focused on how the company estimates TV CM effectiveness with difference-in-differences, comparing broadcast and non-broadcast areas to build counterfactual numbers. TELECY also helped judge and award the CARTA HOLDINGS Award to Toyota Technological Institute researchers for a QA data augmentation method that preserves the original dataset's evaluation perspective.
TELECY's booth centered on two patented methods: Sequential Response Analysis (patent 7112816) and Direct Response Analysis (patent 7651769), which power the TELECY Analytics TV CM measurement tool. The company also showed Lightning MMM, a Marketing Mix Modeling tool that spans offline and digital advertising, and an AI agent that collects past broadcast data and drafts proposal materials for pre-broadcast analysis.
The CARTA HOLDINGS Award went to Ryo Kojima and Makoto Miwa of Toyota Technological Institute for a controllable QA data augmentation method that keeps the evaluation perspective and distribution of the original dataset. TELECY engineers joined the judging, and the selection cited the method's schema design across category and LLM performance axes.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.