DTC Cash Flow Forecasting: A 90 Day Model That Works

Cash flow kills DTC brands that are otherwise healthy

A DTC brand with 40 percent gross margin, growing revenue, and a good retention curve can run out of cash. It happens often. The reason is almost always the cash cycle: ads and inventory get paid upfront, revenue arrives on processor terms two days later, and inventory commitments land in lumpy blocks. A forecast that does not model this pattern is not a forecast.

TL;DR ▸ DTC cash cycles mean profitable companies can still run dry ▸ A 13 week weekly forecast catches timing issues a monthly P&L hides ▸ The hardest inputs are inventory commitments and ad spend return timing ▸ Refresh weekly, compare actual vs forecast, recalibrate the model

This is the 90 day cash model we build with clients. It is a practical tool, not a finance textbook exercise. Our analytics and reporting service builds the underlying data pipeline. This post is the model structure.

Why DTC cash is different

Most finance templates were built for B2B or wholesale businesses. The cash cycle looks very different for DTC.

B2B cash cycle:

• Customer commits via contract, often with deposit
• Service or product delivered over a period
• Invoice sent, payment received 30 to 90 days later
• Marketing spend is often fixed overhead

DTC cash cycle:

• Marketing spend paid daily to platforms
• Inventory paid upfront with deposit, balance due before shipment
• Customer pays at order via Shopify Payments
• Shopify Payments deposits in 2 days, minus fees
• Returns process as refunds 30+ days later

The DTC pattern means cash moves quickly in and out. Large inventory orders create shock events. A growth spurt that requires bigger inventory buys can starve working capital.

The 90 day model structure

Build the model as a 13 week grid. Each column is a week. Rows are grouped by category.

Starting cash

The first row is starting cash, the beginning bank balance plus the beginning payment processor balance. The processor balance matters because Shopify Payments deposits have a lag.

Inflows

Group inflows into categories:

1. Subscription revenue, the most predictable, forecast from the active subscriber base and next charge dates
2. New customer revenue, forecast from ad spend and expected first-order contribution
3. Returning customer revenue, forecast from the base and expected repeat rate
4. Wholesale or B2B revenue, forecast from signed POs
5. Marketplace revenue, forecast from Amazon or retail platforms

Total inflows per week.

Outflows

Group outflows with the same discipline:

1. COGS, paid to suppliers, timed to when the PO balance is due not when inventory sells
2. Fulfillment cost, paid to 3PL, typically weekly or bi-weekly
3. Shipping cost, paid to carrier via processor deductions, typically daily
4. Payment processing fees, netted from Shopify deposits
5. Paid media spend, paid to platforms, typically weekly or monthly
6. Payroll, bi-weekly or semi-monthly
7. Rent, utilities, software, monthly
8. Returns refunds, paid back to customers, timed to the return arrival
9. Tax payments, quarterly or monthly
10. Other, a catch-all for unexpected

Total outflows per week.

Ending cash

Starting cash plus inflows minus outflows equals ending cash. That becomes starting cash for the next week.

The weekly flow example

A simplified week for a $5M revenue DTC brand:

A normal week shows net positive cash. The problem weeks are the ones where payroll and a large inventory payment overlap. Those are the weeks that kill brands who did not forecast.

The inventory commitment layer

The single most important addition to a DTC cash forecast is an inventory commitment schedule.

For every PO in flight, track:

• PO issue date
• Deposit amount and deposit paid date
• Balance amount and balance due date
• Expected delivery date
• Vendor lead time

A PO issued in January with a balance due in April shows up in the April cash forecast as a large outflow. If you are only looking at the immediate month, you miss it.

Build the inventory schedule as a separate tab that feeds into the main cash model. Our ecommerce operations service builds this schedule as a managed artifact.

The ad spend return timing

A dollar spent on ads today does not return today. It returns over the payback period.

If your payback is 45 days:

• Day 1 ad spend of $1,000
• First-order contribution day 1 through 7 returns maybe $400
• Repeat purchases days 8 through 45 return the other $600

Model ad spend as a weekly outflow, and model ad-driven revenue as a rolling inflow that follows the payback curve. This is the hardest part of a DTC forecast.

The simplification that works: calculate your blended payback period once, apply it as a smoothing curve to ad spend, and reforecast monthly as the actuals come in.

The break-even ROAS guide has the payback math.

Subscription revenue forecasting

Subscription revenue is the most forecastable category. The active subscriber base and next charge dates give you a week by week projection.

The formula:

• Start with active subscribers at the beginning of each week
• Subtract expected churn based on survival curves
• Multiply by average order value for subscription orders
• Adjust for failed charges that will not recover

Our LTV modeling service builds the survival curves that feed this calculation.

Subscription revenue should be the cleanest forecast. If it is not, the first month churn fix post and dunning recovery post are the places to start.

The three scenario approach

Build three scenarios in the same model.

Base case. The most likely path based on current performance. All inputs at their current rate.

Downside case. What happens if new customer acquisition drops 25 percent and repeat rate drops 10 percent. This reveals the floor of your cash position.

Upside case. What happens if ad spend scales 50 percent and hit payback. This shows whether the growth is cash-feasible or requires financing.

The downside case is the most important. If the downside case shows a cash trough below your minimum operating balance, you need action now: reduce ad spend, delay inventory commitments, or raise capital.

The weekly update ritual

A forecast that is not updated is fiction. The ritual:

Every Monday morning:

1. Update actuals from the prior week across all categories
2. Compare actual vs forecasted, note any variance over 10 percent
3. Investigate variances, adjust the underlying model if the variance is structural
4. Reforecast the next 12 weeks based on the updated inputs
5. Publish the updated forecast to leadership in a consistent format

The structural variances are the important ones. A random week that was 15 percent off mean-reverts. A consistent pattern of new customer revenue coming in 20 percent below forecast means the forecast model is wrong and needs adjustment.

Financing triggers

Define in advance the cash positions that trigger specific actions:

Having the triggers documented makes the decisions faster when the moment arrives. A weekly forecast that shows you crossing a trigger gives you four to six weeks of lead time.

Common model mistakes

Mistakes we see:

▸ Forecasting revenue without forecasting refunds, understates true cash need ▸ Assuming all inventory is paid on the day the PO is issued rather than on the schedule ▸ Missing the Shopify Payments deposit lag, overstating same-week cash availability ▸ Ignoring seasonal working capital needs like BFCM inventory build ▸ Not modeling the lag on ad spend returns ▸ Using monthly granularity when weekly granularity would surface issues

The last one is the most common. A brand that forecasts by month can end the month net positive and still not make payroll in week three. Weekly granularity is non-negotiable.

When to invest in a warehouse-backed model

The 13 week Google Sheets model works up to about $15M in revenue. Beyond that, manual updates become error-prone and the model drifts.

Warehouse-backed cash models read directly from Shopify, the ads platforms, accounting, and inventory management. They update automatically and support scenario modeling with parameter sliders.

Our analytics and reporting service builds these models. The infrastructure is the same as the DTC finance dashboard. Once the warehouse exists, cash forecasting becomes a set of dbt models on top.

Board-level view

The board wants to see the 12 month view, not the weekly. For board reporting, aggregate the weekly model to monthly and add:

▸ Monthly ending cash position, 12 month forward view ▸ Minimum cash position across the period ▸ Scenario comparison, base vs downside vs upside ▸ Cash runway under each scenario ▸ Financing need and timing under the downside case

This view is the basis for financing conversations. Investors and lenders want to see that you have modeled the cash cycle and understand the troughs.

What to do this week

▸ Build the 13 week skeleton model in Google Sheets if you do not have one ▸ Pull actual inflows and outflows for the prior 4 weeks to calibrate ▸ Create the inventory commitment schedule as a separate tab ▸ Establish the Monday morning update ritual with a named owner ▸ Define your cash position triggers for specific actions ▸ Run the downside scenario and check the minimum cash position ▸ Start a conversation with your bank about a line of credit before you need it

Cash forecasting is not fancy finance work. It is operational discipline. A mid-size DTC brand that runs a weekly cash model with three scenarios and a triggered action plan will outlast brands with better products and worse cash discipline. Build the model, run the ritual, and let the rest of the business operate with the benefit of cash visibility.