How to Forecast Cash for a Product Launch (2026)

Every product launch is a cash bet before it is a marketing event. You wire a deposit to your manufacturer months before a single unit sells, pay the balance and freight when the goods land, then spend on ads to move inventory that public beauty brands hold for a median of 168 days before sell-through returns cash. The launch deck talks about revenue. The bank account talks about the deposit, the freight invoice, and the ad spend that all left first.

That gap between cash out and cash in is where launches kill brands. The good news: it is predictable, which means it is forecastable. When I talk to founders planning a launch this size, the question they ask is "can we afford it?" and they almost always mean "can the P&L afford it?" The P&L is the wrong place to look. Below is how I build a driver-based launch cash forecast, what the curve actually looks like, and how to stress-test the downside so a slow start does not sink the company.

A launch is a cash bet, not a revenue line

Here is the trap. Your P&L shows revenue when it is earned, not when it is collected, and expenses when they are incurred, not when they are paid. A launch widens every one of those timing gaps at once. The deposit is cash gone before anything is made. Production and freight are cash gone before anything sells. Ad spend drafts daily while customer cash lands days later. By the time the P&L shows a profitable launch, your bank account has already been through the worst of it.

So forecast the launch the way the bank account experiences it: as a sequence of cash events over time, with the deposit at the front and recovery at the back. The simplest version of this discipline is the one we tell every operator to start with: look at your bank balance from one week to the next, write the difference on a Google sheet, and watch it every single week. A launch forecast is just that habit projected forward across the deposit, the landing, and the slow climb back. Revenue is an output of the model. It is not the input.

Here is the sequence the bank account actually sees. The deposit and the landing balance are the two big bets. Ad spend and refunds keep drawing cash down for months while the launch is live.

Build it from drivers, not a guess

A single revenue guess cannot be stress-tested. A driver-based forecast links every line to an operational lever you actually control, so when you change one assumption the whole cash curve recomputes in seconds. The way we explain it to founders is plain: you break revenue into its components. One row is ad spend, the row below it is your cost to acquire a customer, the row below that is customers. Ad spend divided by acquisition cost equals customers, customers times AOV equals revenue. Now every number on the page is a lever you can pull, not a wish. For a launch, these are the drivers that matter:

Notice that the supplier deposit alone is typically 30 to 50% of the order, out the door on Day 0. The 8% returns input above is a category-light placeholder. Brand-owned DTC sites blend closer to 14 to 15% returns, and apparel runs higher still at 22 to 30%, so calibrate this driver to your own category before you trust the recovery curve. The sell-through curve is the single driver that swings the whole model, which is exactly why it deserves a base case and a downside case rather than one optimistic line.

What the launch cash curve actually looks like

Take a representative $20M DTC brand placing a $300K production order, 40% deposit, sourced from Asia with a roughly 90-day production and freight lead time, sold mostly direct. Plot cumulative net cash from the deposit forward and you get a deep V: cash drops on the deposit, drops again hard when the balance and freight clear at landing, then claws back as units sell.

In the base case the cumulative cash trough bottoms near $300K around the time the inventory lands, and the brand does not turn cash-positive on the launch until roughly month 7. That is on a launch that works. The number you fund is the depth of that trough, not the size of the order.

Stress-test the downside before you commit

The slow line on the chart is the one that matters. Hold the deposit, the freight, and the planned ad spend fixed, then cut sell-through to 60% of base. Now the curve bottoms lower and stays underwater far longer, past month 8, because the cash kept leaving on schedule while the cash coming back slowed to a trickle. This is how a profitable-on-paper launch becomes a solvency problem.

How long you stay underwater depends heavily on your category, because slow-moving categories hold inventory longer before it converts to cash. Beauty is the worst offender.

This is where over-ordering quietly compounds the problem. The pattern we see again and again is a brand carrying more inventory than the financials would ever justify, often because a price break tempted them into a bigger order. When I talk to founders sitting on 250 days of inventory, the cash conversion cycle is so long the business is starved even while it stays profitable on paper. On a launch, the order you can sell through in three to four months is almost always the right one, even if the per-unit cost is a little higher. One operator we worked with went the whole way the other direction, refusing every volume discount and placing fresh orders every two weeks for a year. It was a clusterload of work, but the on-hand inventory balance stayed tiny and the cash stayed free.

The discipline is simple: forecast the downside first, fund to its trough, and only then decide whether the launch is safe to place. A launch you can only survive if everything goes right is not a launch. It is a gamble. For context on how accurate these models can get, Eightx clients have forecast revenue to 94% accuracy over 17 months using driver-based models, which is what makes the downside case trustworthy enough to bet on.

What to do about it

1. Map the cash events on a calendar: deposit date, balance and freight at landing, ad spend by month, expected sell-through by week, and refund timing. Cash dates, not accrual dates.
2. Build the model from drivers, not a revenue guess. Units, landed cost, sell-through curve, AOV, ad spend, CAC and returns. Change one and the whole curve should move.
3. Plot cumulative net cash week by week and find the trough. That number, not the order value, is the cash the launch ties up.
4. Run a downside case at 60% of base sell-through with costs held fixed. Read the new trough and the new recovery month.
5. Fund to the downside trough plus a buffer. If you cannot, cut the order, stage deliveries, negotiate longer supplier terms, or line up a line of credit before the deposit goes out.
6. Reforecast against actuals weekly once the launch is live, and fold it into your 13-week cash flow forecast so the launch never surprises the wider business.
7. If the launch lands in a peak window, layer it onto your seasonal cash flow forecasting so two cash bets do not collide in the same month.

This launch model is one piece of a bigger discipline. If you want the full system for running an ecommerce business on cash rather than on the P&L, start with our pillar guide to cash flow mastery for ecommerce CEOs.

Sources and methodology

The cash curve is an illustrative driver-based model for a representative $20M DTC brand placing a $300K production order at a 40% deposit with a roughly 90-day production and freight lead time, sold mostly direct. The 90-day anchor reflects a production-and-ocean-freight window from Asia to the US that runs roughly 8 to 13 weeks PO-to-landed in 2026: production is about 5 to 7 weeks on repeat orders and 8 to 12 weeks on first or complex SKUs, and China-to-US ocean freight adds another 20 to 40 days door-to-door.

Inventory days are anchored to a median 168 DIO observed across 26 public consumer 10-Ks in Eightx analysis of beauty working capital, with the blended public-DTC median of 129 days and the P&G anchor of 65 days drawn from the Eightx inventory-management compile. Deposit norms of 30 to 50% and the profit-is-not-cash timing traps come from Eightx cash flow forecasting work.

The returns band of 14 to 15% blended DTC, with apparel at 22 to 30% and refunds settling as cash roughly 7 to 15 days after the customer ships, comes from Eightx returns analysis cross-checked against NRF and Richpanel figures. For the cost of any line of credit used to bridge the trough, the US bank prime loan rate was 6.75% and SOFR was 3.63% as of early June 2026, pulled from the Federal Reserve Bank of St. Louis FRED series DPRIME and SOFR.

Figures are planning estimates, not a guaranteed outcome, and your numbers will differ by category, channel mix and lead time. Triangulation for this post combined Matt's founder-call corpus for the operator-voice lines, a web-and-citations pass for the lead-time and returns benchmarks, and a deep async run that re-confirmed the two FRED rate series against their primary source.

Frequently Asked Questions