<![CDATA[This study presents the first systematic comparative evaluation of eight forecasting models applied to Google Trends data specifically for Indonesian digital MSME product categories, addressing a critical gap in accessible demand prediction tools for resource-constrained enterprises. Employing quantitative comparative experimental design, we analyzed 18 product keywords from fashion and traditional food sectors with weekly temporal resolution (2020-2024), evaluated via 80/20 train-test split with MAPE. Exponential Smoothing achieved superior overall performance (20.33% MAPE), while Moving Average proved optimal for traditional food products (16.21% MAPE). Category-specific model selection enables MSMEs to anticipate demand peaks, time market entry windows, optimize inventory positioning ahead of seasonal cycles, and prioritize promotional spending, translating search trend signals directly into actionable penetration decisions. These findings provide an evidence-based, zero-cost framework for demand-informed market strategy, accessible to resource-constrained enterprises without specialized infrastructure or advanced technical expertise.]]>
Copyrights © 2026
