Ecommerce Cash Flow Forecasting:
The 13-Week Model DTC Founders Should Run

Short answer

Build a rolling weekly model of cash in and out over one quarter, tuned to two ecommerce quirks: card and marketplace payouts land days to weeks after the sale, and you pay for inventory weeks before it sells. Forecast receipts by channel at real payout dates, then subtract inventory POs, ad spend, fees and payroll for each week's closing cash.

You can earn a margin on every order and still watch your bank balance slide toward zero. That is the ecommerce cash flow forecasting problem: the money you make and the money that reaches your account run on different clocks. A generic 13-week cash flow forecast assumes a sale becomes cash when it happens; in a DTC business it doesn't, because card settlements lag, marketplace payouts lag more, and inventory swallows cash weeks before it sells. Here is the DTC version: a copyable 13-week skeleton and the two lines founders always get wrong.

Why ecommerce cash flow forecasting needs its own model

Every online brand lives inside two timing gaps, and both push cash out before it comes in.

The payout gap. A sale is not a deposit. Card payments settle two to three business days later; marketplace channels pay on a rolling schedule of a week or two; wholesale sits on Net 30 to Net 60. The revenue shows in your dashboard today; the cash arrives at three or four different speeds.

The inventory gap. You pay for stock long before it earns — a deposit at the purchase order, the balance on shipment, then weeks in transit and on the shelf before it sells. As J.P. Morgan's cash conversion cycle work notes, inventory is usually the largest slice of the days your money is tied up — the math behind your cash conversion cycle.

Put them together: cash leaves for inventory in week 1, the sale lands in week 8, the money clears in week 10 — a forecast that treats revenue as cash misses all of it.

How do you build a 13-week cash flow forecast for an ecommerce business?

Thirteen weeks is one quarter — long enough to see the next inventory buy, short enough that weekly estimates stay honest. Build it on the direct method: real cash in, real cash out, week by week. Five steps.

1. Set 13 weekly columns, opening cash at the top, closing cash at the bottom. Closing cash in week 1 becomes opening cash in week 2 — that chain is the model.
2. Forecast receipts by channel, at their real payout date, not the sale date. Put website card sales, each marketplace, and wholesale on their own lines, and place the cash in the week it clears. This is what separates an ecommerce forecast from a generic one.
3. Forecast inventory outflows by purchase order: deposit in the week it leaves, balance in the week it ships. Lumpy is the point.
4. Layer the costs that move with sales: ad spend, fulfilment and shipping, processing fees, and a returns allowance that claws cash back after the order.
5. Add fixed costs, financing repayments and tax/VAT, compute closing cash, flag any week under your minimum, then reconcile weekly and add week 14.

A 2026 DTC P&L benchmark is a useful gut-check — it puts ad spend and fulfilment at a large share of DTC revenue and returns near a fifth of orders. If those lines run light in your model, it is too optimistic.

The 13-week template: copy this structure

Numbers are illustrative; the structure is the template. Note how receipts land after the sales that created them, while the inventory PO lands well before.

Two build rules. Name each channel and large customer on its own receipt line — a lumped "revenue" line hides the payout timing you built the model to see. And set your payout lags from recent bank statements, not a channel's stated terms; the two rarely match.

The two lines founders get wrong

Everything above is bookkeeping. The edge is in two lines a spreadsheet won't fix for you.

Inventory, dated wrong. Founders log a purchase order the week they place it, then get caught when the balance clears on shipment eight weeks later. Book cash on the dates money actually leaves — deposit one week, balance another — and the model stops lying about your mid-quarter position. This is also the biggest structural fix: buy less dead stock. McKinsey's work on AI in distribution operations found stronger demand forecasting and inventory optimization cut inventory by 20–30%, and even a basic demand-planning tool and tighter reorder points free cash that would otherwise sit on a shelf.

Returns, forgotten. Refunds pull cash back after you have counted the sale, so booking the order but not the refund makes every line above it fiction. Put a returns allowance in at your real rate and watch it drag closing cash down after a promotion.

What this means for you

If your runway is under twelve months, one channel is more than a quarter of collections, or you are about to place a large seasonal order, run this forecast — this week, in a plain spreadsheet. One person owns it; the weekly update takes under an hour.

The forecast earns its keep twice. It shows a gap early enough to fix cheaply — the difference between arranging inventory financing without giving up equity on your terms and scrambling on someone else's. And it is the backbone of planning cash and inventory for Q4, when both gaps stretch widest just as you are spending most.