Observation Log: Typology of 'De-Excel' Discourse in Japanese and English Web Content, 2026

This dataset is an observational log of "de-Excel" discourse collected through natural exposure to Google Discover and selective deep-reading, conducted by a single researcher between May and June 2026. The collection includes 14 coded entries spanning Japanese and English web content, covering vendor articles, media reports, individual engineer posts, LinkedIn publications, and Google AI Overview responses. Each entry is coded using a five-category typology: A (total rejection of Excel), B (criticism of misuse), C (constructive continued use), D (misuse deflection onto Excel itself), and E (rejection of the de-Excel concept). Compound judgments such as A+D are permitted. The most frequent pattern observed is A+D, in which vendors simultaneously deny Excel's value while building products that presuppose Excel's existence. Key findings include: (1) no entry provides a logically coherent basis for total rejection of Excel; (2) sender attributes correlate with discourse type, with commercial vendors predominantly producing A+D discourse; (3) empirical data from the same period shows that 85.7% of companies with BI tools still use Excel as their primary data visualization tool, indicating a structural gap between discourse and practice; (4) the discourse space itself may be subject to algorithmic amplification via Google Discover, which the researcher acknowledges as a methodological limitation. This collection does not aim at quantitative distribution estimation. Its purpose is typological description and structural analysis of de-Excel discourse as a commercially produced linguistic phenomenon.