Demand Forecaster
Upload your daily sales history. We will pick the best forecasting method for each product and email you the forecast as Excel files.
Input format
Download a sample: sample_input.csv · sample_input.xlsx
| Column | Required | Description | Example |
|---|---|---|---|
product_id | yes | Product / SKU identifier | SKU-1001 |
date | yes | Sales date, YYYY-MM-DD | 2026-03-01 |
sales_units | yes | Units sold that day (whole number, 0 or more) | 14 |
metadata | no | Optional JSON text; leave empty if unsure | {} |
- One row per product per day. Several rows for the same product and day are added together.
- Provide at least 35 days of history; more history (a year or longer) gives better forecasts.
- Products with fewer than 35 days of history get a simple default forecast.
- Processing time depends on the number of products: from a few minutes to several hours for thousands of products.