**Slack message -- #analytics-requests** **Eunji Cho** 9:14 AM > hi! I'm Eunji, head of merchandising analytics at Glow Republic. We're a K-beauty retailer -- 12 stores in Seoul + e-commerce. > > we have a demand forecasting problem and it's costing us real money. last quarter: 180M won in expired inventory write-offs + 250M won in missed sales from stockouts. K-beauty trends move SO fast -- a product goes from zero to viral in two weeks and we can't keep up. > > I have: 2 years of daily sales data (~2000 SKUs), social media mention counts (Instagram + TikTok), influencer tag data, and seasonal calendars. all in CSVs. > > what I need: predict demand by SKU at least a week out so our buying team can order the right quantities. and flag products that are starting to trend before they peak. > > the sunscreen/seasonal stuff is different from the viral stuff (like when that centella ampoule blew up last year from one TikTok video). they probably need different approaches. > > can you help? I can share the data today. budget isn't an issue, we just need it to work.