Cash Flow Forecasting for Seasonal Business: A Weekly Playbook

Cash Flow Forecasting for Seasonal Business: A Weekly Playbook

The forecast on the founder's screen showed a green Q4. Bank balance projected: positive. Revenue plan: on track. Inventory cover: thirteen weeks. Then week 37 hit, and the bank account read $11,400. Payroll was due Friday. The Q4 inventory invoice had cleared on a Wednesday, three weeks before the marketplace payout cycle would catch up.

This is the most expensive bug in seasonal eCommerce finance, and it lives inside the monthly cash-flow forecast every founder runs in Xero or QuickBooks.

The Week 37 Bank Account That Killed a Profitable Brand

The brand was real, the numbers slightly anonymised. Outdoor goods, $4M revenue, a four-year-old DTC business with retail wholesale doing about thirty percent of shipped units. Margins held up. Marketing spend was disciplined. The founder had built a twelve-month cash-flow forecast in Xero with monthly granularity and reviewed it the first Tuesday of every month.

September hit. The pre-Q4 inventory order was the largest of the year: $480,000 in containerised goods, with a forty percent deposit due on the booking date and the balance on bill-of-lading. The deposit cleared the bank in week 35. The balance cleared in week 38. Across the same window, marketplace payouts were running on a fourteen-day rolling cycle, ad spend was pacing up to support the late-September retargeting flight, and a returns spike from the August clearance promotion was draining the contra account.

By the close of week 37, the founder was three days from missing payroll. The monthly forecast still showed September ending at $145,000 cash positive.

The math was not wrong. The granularity was wrong. A monthly forecast averages an entire calendar month into one number, which is fine when cash flows are smooth and predictable. Seasonal eCommerce cash flows are neither. They lump and tear: deposits land on specific Wednesdays, payouts arrive seven to fourteen days after the sale, returns clear in the week following the refund decision, and ad spend bills the day the platform charges your card. The week that kills you is always a week your monthly view has already rounded away.

The most frequent error in eCommerce cash forecasting is projecting sales rather than cash. A $50,000 sales week typically lands as roughly $32,000 of bank cash after marketplace fees of fifteen to thirty percent, returns, and seven-day-plus payout delays, yet most operators still run a monthly forecast that treats top-line revenue as cash. The 13-week ecommerce forecast playbook is the operator-grade fix for this and the foundation of what follows.

Most monthly forecasts also fail in the other direction. They assume disbursements are smooth across a month when in reality the bank sees lumpy weeks: a payroll Friday, a supplier payment Wednesday, an ad-spend charge on the first business day of the month. Two clean months on a monthly view can hide a Tuesday in the middle that looks ugly enough to miss a payment.

The outdoor goods founder did not lack discipline. The forecast was reviewed monthly, the bank balance was checked daily, the inventory order was approved by both the founder and the operations lead. None of that mattered. The control loop was running on the wrong cadence. By the time the September close was visible in the monthly report, the cash floor inside that month had already been breached and patched with personal credit. The control loop has to operate at the same frequency as the cash flows it controls. For seasonal eCommerce, that frequency is weekly.

Why the Math Doesn't Work: Sales Are Not Cash

The first job of a cash-flow forecast is to translate revenue into bank balance. Most monthly forecasts get this wrong because they pull the number straight from the Shopify or Amazon report and assume it lands the same week. It doesn't. Three frictions sit between a sale and a cleared deposit, and seasonal volume amplifies all three.

The first friction is platform fees. Amazon FBA pulls fifteen to twenty-five percent in seller fees, fulfilment fees, and storage charges before the payout hits. Shopify Payments takes about 1.7 percent plus thirty cents per transaction. eBay, Etsy, and the marketplace channels take their cut at different points in the settlement cycle. The Wayflyer 13-week plan walks through how the gap between gross sales and net deposits widens during peak weeks, when fee tiers and storage charges step up alongside volume.

The second friction is the payout cycle. Shopify Payments runs daily but settles on a two- to three-business-day delay. Amazon FBA pays every fourteen days. PayPal holds new-seller balances for up to twenty-one days. Stripe runs on a rolling two-day cycle. When you stack these channels, the sale on Tuesday week 1 becomes cash on Friday week 1, the sale on Tuesday week 2 becomes cash on the second Friday, and so on. A monthly forecast collapses all of this into one number and presents it as the closing balance.

The third friction is returns and chargebacks. Returns clear on the week the refund is processed, not the week the original sale settled. For seasonal apparel, outdoor goods, and consumer electronics, the returns curve typically lags the peak by eight to twelve weeks, which means your December peak generates a January refund cliff that the monthly forecast already booked as revenue. Chargebacks are even later: Visa and Mastercard rules give the cardholder up to 120 days to dispute a transaction, so the back end of the seasonal returns curve runs deep into Q1.

You also need to budget cash for the things you can predict and the things you cannot. Inventory deposits, supplier balance payments, freight invoices, port fees, and customs duty land in lumpy weeks. Ad spend bills in real time. Payroll runs fortnightly or monthly. The 8fig cash flow guide describes the peak-season inventory buy-in as the single largest cash drag for any growing physical-product brand, and a monthly forecast cannot model the three-week gap between a deposit clearing and the corresponding revenue arriving.

The result is the gap I keep seeing in the brands I work with: the month closes positive, the founder feels safe, and a specific Friday inside that month was a heart attack the founder never saw coming. The forecast lied by averaging.

The Seasonal Liquidity Blueprint

I call the fix the Seasonal Liquidity Blueprint. It is a three-block weekly model that replaces the annual Xero forecast with a rolling thirteen-week view of net cash. The structure is borrowed from the treasury world, where corporate finance teams have run thirteen-week forecasts for decades, and adapted for the realities of marketplace fees, returns, and payout delays that an Xero export cannot see. The 13-week cash flow model is the original treasury-side reference and a clean explanation of why thirteen weeks is the right horizon: long enough to see a peak coming, short enough to update weekly without losing accuracy.

The Seasonal Liquidity Blueprint has three blocks.

Block 1: Net-cash-by-week revenue. This is gross sales by sales channel, by week, with a per-channel deduction for marketplace fees, payment-processing fees, and a returns reserve. The output is the cash you actually expect to bank in the week the payout settles, not the week the sale was made. For most brands this means a Shopify column, an Amazon column, an eBay column, and a wholesale column, with separate fee assumptions on each.

Block 2: Weekly disbursements. This is every cash outflow with the week it clears the bank, not the week it is incurred. Inventory deposits, supplier balance payments, freight, customs, ad spend, payroll, software subscriptions, rent, tax, and any debt service. Keep these line items granular. A $480,000 container split into a forty percent deposit in week 35 and a sixty percent balance in week 38 is two separate line items.

Block 3: Rolling thirteen-week cash position. This is the running balance, week by week, starting from current bank balance plus expected receivables. The output is a single chart that shows the cash floor for the next ninety days. The lowest point of that chart is the number you actually need to defend. The Dryrun rolling forecast walks through the rolling discipline in detail.

I have run the Seasonal Liquidity Blueprint inside a number of brands between $1.5M and $11M revenue across the last few years. The pattern is consistent: every founder discovers a week inside the next ninety days where their cash floor is materially lower than they thought, and roughly half discover that week is closer than thirty days out. The Blueprint is not a forecasting tool. It is a survival tool.

The model replaces the monthly Xero export, but it does not replace the annual budget. The two run side by side. The monthly view answers "are we profitable" and the Blueprint answers "do we survive Friday".

Execution: Day 0 to Day 90

The Seasonal Liquidity Blueprint is buildable inside a Google Sheet over a single weekend. The discipline of running it weekly is harder than the build, which is why most brands stop at month two. The execution sequence is three phases, ninety days end to end.

Phase 1: Build the Skeleton (Day 0-30). Pull last ninety days of bank statements and cross-reference them against your Shopify, Amazon, and marketplace payout reports. The goal of week one is one number per channel: the average gap between sale week and cash week, and the average percentage drop between gross sales and net deposit. Most founders are surprised by both. Then build the disbursement side. Pull every supplier invoice, every container booking, every recurring software charge, every payroll run. Map each one to the week it clears the bank, not the week it is dated. This is week two work and takes about six hours if your accountant has clean books, double that if you do it solo. The Growth Lab 10 steps build guide is the cleanest non-finance-founder reference for this phase. By the end of week four, you should have a working thirteen-week model running off historical data with no seasonality overlay yet.

Phase 2: Add the Seasonal Overlay (Month 2-3). Take last year's same-week revenue figures by channel and apply your year-on-year growth rate. Then layer on the current marketing calendar: any flash sale, paid-traffic push, or wholesale shipment that is dated. The seasonal overlay turns the model from a rolling average into a forecast. The Abacum 13-week guide covers seasonality handling and the two adjustments that matter most for retail-DTC: a returns curve that lags the peak by eight to twelve weeks, and an inventory deposit cycle that leads the peak by six to ten weeks. If you sell apparel, outdoor, or consumer electronics, the returns curve alone is worth a separate sub-row.

Phase 3: Friday Cash Review and Inventory Go/No-Go (Month 3+). Every Friday afternoon, the founder, the finance lead, and the operations lead spend thirty minutes inside the model. Update the prior week's actual revenue and disbursements. Re-roll the thirteen-week view forward by one week. Review the lowest cash floor inside the next ninety days. If the floor drops below an agreed threshold (I use four weeks of disbursements as a default), the team triggers an inventory go/no-go decision: the next non-essential PO either gets pushed back by a fortnight or held entirely until the cash floor recovers. The Intuit 13-week guide describes the rolling weekly cadence as the single most important habit for SMB cash management, and the Blueprint stands or falls on whether this Friday review actually happens.

Two pitfalls to avoid. First, the model gets stale fast if you only refresh it monthly. Weekly is the cadence and there is no shortcut around it. Second, do not try to build the model directly in Xero or QuickBooks. Both tools are accounting systems and forecast in monthly buckets by design. Use the Xero export as input, not as the model. Build the Blueprint in a Google Sheet, an Excel workbook, or a purpose-built tool like Wayflyer. The Wayflyer cashflow planning free template is the closest off-the-shelf starting point if you would rather not build from a blank sheet.

Two roles matter on Friday. The finance lead owns the actuals refresh: closing the prior week, reconciling payouts to deposits, updating supplier invoices that arrived. The founder owns the forward view: pacing decisions, supplier negotiations, inventory go/no-go calls. Splitting these roles keeps the review tight. Thirty minutes is enough when the actuals are clean. It blows out to ninety minutes when the actuals are stale, which is the second-strongest argument for keeping the cadence weekly.

From Monthly Comfort to Weekly Discipline

The before-and-after on the outdoor goods brand was unspectacular. They did not raise capital. They did not switch suppliers. They did not refinance anything. They moved from a monthly Xero forecast that lied to a weekly thirteen-week sheet that told them the truth, and the truth was that their cash floor inside the next ninety days was three weeks of disbursements lower than they had assumed.

That single number changed three decisions inside the first quarter of running the model. The Q4 inventory order was split into two smaller containers four weeks apart instead of one large container in week 35. The next paid-traffic flight was paced to track payout cycles rather than calendar months. The supplier negotiation that the founder had been avoiding got prioritised, because the model showed that a fourteen-day extension on creditor terms moved the cash floor up by $62,000.

Year two on the model produced a quieter set of changes, but the cumulative effect was larger. The wholesale customer list was re-priced to reflect the true cash cost of net-60 terms. A second freight forwarder was added so port congestion at one hub did not flow straight into a missed payroll. The board pack was rewritten to lead with the thirteen-week cash floor rather than the trailing-twelve-month revenue line. None of these moves was dramatic. All of them were impossible to spot inside a monthly forecast. The pattern I see across the brands I work with is the same: the first ninety days of the model surface one big problem, and the next ninety days surface ten small ones, and the small ones in aggregate matter more than the first.

The new north star metric is not revenue. It is not gross margin. It is the thirteen-week cash floor, the lowest projected bank balance inside the next ninety days, refreshed every Friday and defended like the survival number it is. Above the floor is operational room. Below the floor is a heart attack you can see coming. A monthly forecast cannot show you this number. A thirteen-week weekly forecast cannot hide it.

The brands that scale through seasonal peaks do not have better revenue. They have better cash visibility. The Blueprint is the cheapest insurance policy you will buy this year. Build the skeleton this weekend. Run the Friday review next week. Defend the floor.