A wholesale forecast built around your biggest customers runs into an immediate problem: those customers rarely share their actual purchasing plans. Retailers guard their open-to-buy numbers, buying calendars, and reorder timing closely — leaving the supplier to forecast a relationship they can only partially see into.
Why this is harder than a direct-to-consumer forecast
A retail or ecommerce forecast can lean on your own historical demand data and marketing calendar — you control most of the variables. A wholesale forecast depends heavily on decisions made inside someone else's business: their inventory position, their own sell-through, their promotional calendar, their vendor consolidation decisions. None of that shows up in your own systems until a purchase order actually arrives.
What to build instead of waiting for visibility you won't get
- A rolling history of actual order patterns by customer — timing, size, and seasonality — even without their internal plans, past behavior is a real signal
- Sell-through proxies where available — POS data feeds, reorder velocity, or even informal check-ins with buyers can substitute for direct visibility
- Scenario ranges rather than a single number — a base case built on trailing patterns, plus upside/downside cases tied to known risk factors like a customer's own public performance signals
- Customer concentration flags — if a small number of accounts drive most of wholesale revenue, the forecast needs to show that dependency explicitly, not average it away
The goal isn't certainty — it's a usable range
A wholesale forecast built this way won't be as precise as one built on full visibility, and it shouldn't pretend to be. The goal is a defensible range grounded in real order history and known risk factors, updated as actual orders come in — enough to plan inventory, cash, and staffing decisions around, without waiting on information your customers were never going to hand over.