Clear ownership, predictable capacity.

No. You are buying a recurring planning function. Reorder Desk prepares the plan; your team approves purchases.

An independently replenished SKU that receives a forecast, inventory-position calculation, reorder decision, risk classification, and accuracy tracking each cycle.

Not without explicit approval. Supplier negotiation, final PO approval, logistics, and inventory accounting remain with your team.

At minimum: dated SKU sales, on-hand and allocated inventory, unit costs, supplier lead times, safety stock, MOQs, and case packs.

Growing Shopify brands with physical, independently replenished products and repeatable demand. Common fits include beauty, personal care, home fragrance, pet products, household goods, stationery, craft supplies, and everyday accessories.

Yes. Planned date ranges and expected uplift are applied to the affected forecast weeks and documented in the plan.

The cycle produces a management exception instead of silently inventing a value. Source-data accuracy remains the customer’s responsibility.

No single method is assumed to fit every SKU. Up to 24 candidate forecasting methods from the demand-planning literature — including Croston's method and its variants for intermittent demand, exponential smoothing, and the Theta method that won the M3 forecasting competition — are backtested against each SKU's own held-out history with a rolling-origin design across multiple horizons. A challenger only replaces the simple baseline when a Diebold-Mariano significance test, Bonferroni-corrected for the number of methods compared, confirms the improvement is real rather than backtest noise. WAPE and bias are reported by SKU every cycle, and seasonal-naive models only join the pool once a SKU has enough history (52+ weeks for monthly/quarterly cycles, 104+ for annual) to test for seasonality honestly.