Most cash flow projection guides hand you the same template: list cash in, list cash out, subtract. That math is right, and it is also the easy part. The hard part — the part that actually sinks growing stores — is timing, and nearly every generic guide waves at it and moves on.
This guide fixes that. It is written for a store with real order history and real ad spend, not a hypothetical first sale.
Projection vs. profit: they are not the same number
Your profit and loss statement tells you whether the business made money. Your cash flow projection tells you whether you can cover the bills this week. These two numbers routinely disagree, and the gap is where owners get blindsided.
Profit is booked on the sale date. Cash moves on its own schedule — payouts land days after the sale, while ad spend leaves today. You can post a profitable month and still be short on cash on a Wednesday. If that distinction is new to you, our ecommerce P&L guide walks the full statement, and the piece on managing cash flow for a small business covers the day-to-day discipline.
This matters because cash is the constraint that actually ends businesses. In one widely cited figure, about 30% of businesses fail because the owner runs out of money, and a QuickBooks survey found 60% of US small business owners say cash flow has been a problem. A projection is how you see the shortfall coming instead of discovering it at the ad-card decline.
The five inputs a real projection needs
A usable projection has five moving parts. Skip any one and the forecast lies to you.
- Opening cash — what is actually in the bank today, not your Shopify balance.
- Cash in — the net payout for orders already placed, dated to when it settles, not when the order came in.
- Variable cash out — ad spend and supplier production charges, which scale with volume.
- Fixed cash out — platform plan, apps, tools, contractor pay, and money set aside for quarterly estimated taxes.
- Timing lags — the delay between each inflow and outflow, which is the whole ballgame.
Notice that "cash in" is net, not gross. The deposit Shopify sends you is a settlement: sales minus fees minus refunds. Treat the payout as your cash consequence, and keep gross sales on your P&L — confusing the two is the single most common bookkeeping error, and our balance sheet and P&L breakdown shows where each belongs.
Build it week by week — a worked example
Monthly projections hide the problem; weekly ones expose it. Say you run a t-shirt store doing 340 orders a month at a $31 average order value, spending $2,800/month on Meta — roughly 85 orders and $700 of ad spend per week.
Here is a normal week, with the arithmetic visible:
- Weekly sales: 85 × $31 = $2,635
- Supplier (POD) production: 85 × $13 = $1,105
- Processing fees (Shopify Payments runs about 2.9% + 30¢ per order on lower-tier plans): (0.029 × $2,635) + (85 × $0.30) = $76 + $26 = $102
- Net payout that eventually lands: $2,635 − $102 = $2,533
- Fixed costs this week (plan, apps, tools): $120
Now the key move: the $2,533 payout for this week's orders does not arrive this week. Shopify Payments settles on a rolling delay — commonly a couple of business days in the US, longer across weekends. So cash in each week reflects last week's sales, while cash out — ads and supplier charges — hits today.
The float trap most templates ignore
Watch what happens when you decide to scale. You double the ad budget to $1,400/week in week two, which roughly doubles orders to 170 that same week. The supplier bills you immediately for production, but the bigger payout still lands a week later.
Starting with $4,000 in the bank:
| Week | Cash in (prior week's payout) | Ad spend | Supplier + fees | Fixed | Net change | Closing cash |
|---|---|---|---|---|---|---|
| 1 (normal) | $2,533 | $700 | $1,207 | $120 | +$506 | $4,506 |
| 2 (scale) | $2,533 | $1,400 | $2,414 | $120 | −$1,401 | $3,105 |
| 3 (scale) | $5,066 | $1,400 | $2,414 | $120 | +$1,132 | $4,237 |
| 4 (scale) | $5,066 | $1,400 | $2,414 | $120 | +$1,132 | $5,369 |
These figures are an illustration, not a market claim. Read week two: the store is more profitable than ever, yet cash drops by $1,401 because you paid for double the ads and double the production up front while still collecting last week's smaller payout. Closing cash recovers in week three once the bigger payouts catch up.
The trap is obvious on the schedule and invisible on the P&L. Had you opened week two with $1,000 instead of $4,000, that −$1,401 swing would have put you underwater — a profitable store with a declined card. The faster you scale, the wider the gap you must pre-fund.
Sizing your cash buffer
Your projection's job is to name your lowest point and tell you how much cushion covers it. A simple rule: hold a buffer at least equal to your variable cash out across the full payout gap.
During the scale-up above, you spend roughly $200/day on ads plus about $320/day on production — call it $520/day. Across a five-day settlement-and-weekend gap, that is about $2,600 you need sitting in the account before the matching payouts arrive. Build the weekly table out a full quarter and the required buffer stops being a guess.
Two habits make the projection trustworthy: update it every week with actual balances, and never raise ad spend faster than your buffer and payouts can refill the tank. When you are ready to automate the modeling, compare tools in our rundown of the best cash flow forecasting software for small business and the dedicated cash flow software for small business.
This section touches taxes only in passing. It is general information, not tax advice — rules change and vary by situation, so consult a licensed CPA before acting.
Where the numbers come from
A projection is only as good as its inputs, and the inputs live across your store, your ad accounts, and your supplier. Pulling net payouts, per-order supplier cost, and fees into one honest weekly view by hand is slow, which is why most owners project monthly and miss the squeeze.
PodVector AI's Victor is an AI employee that connects your live data — Shopify, Meta Ads, Google Ads, Printify, Printful, Gelato, and Klaviyo — and computes your true per-order profit from it, then delivers reports straight to your Google Drive. Victor is not a dashboard you log into; it is an employee that works your numbers. Every write action it takes is approval-gated, so nothing executes until you say go. Put Victor to work on your store.
FAQs
How far out should a small business cash flow projection go?
Project thirteen weeks at a minimum, updated weekly. A quarter is far enough to catch a scale-up squeeze or a seasonal dip, and short enough that your estimates stay grounded in real order history. Keep a rougher twelve-month view for annual planning, but run your decisions off the weekly schedule.
What is the difference between a cash flow projection and a cash flow forecast?
In practice the terms are used interchangeably — both look forward at cash in and cash out. Some accountants reserve "forecast" for the longer, higher-level annual view and "projection" for the detailed near-term schedule. What matters is that the number is forward-looking and dated to when cash actually moves, not when sales are booked.
Why does my projection say I am fine but my bank account is tight?
Almost always a timing problem. If you project on gross sales or assume payouts arrive the day of the order, your "cash in" shows up too early and too large. Re-date every inflow to its real settlement day and date supplier charges to production day — the tight weeks will appear.
Should ad spend go in my cash flow projection as it is spent or as it drives sales?
As it is spent. Ad platforms charge your card continuously, so the cash leaves daily regardless of when the resulting orders pay out. That immediate, daily outflow against a delayed inflow is exactly the gap your projection exists to surface.
Do I include refunds in the projection?
Yes. Refunds pull cash back out, and the original processing fee usually is not returned to you, so a refunded order costs you the fee even though you kept none of the sale. Model a realistic refund rate as a recurring outflow rather than assuming every order sticks.