If you already run an operating store — real orders, real ad spend, real supplier invoices — "projected COGS" is not an accounting-class abstraction. It is the number that tells you whether next month's growth is actually profitable or just bigger.
What projected cost of goods sold actually means for an operating store
Cost of goods sold is the direct cost of producing what you sold. Projecting it means estimating that cost for a future period before the orders arrive.
Classic managerial accounting builds a COGS budget by taking a projected cost per unit and multiplying it by budgeted output, as Lumen Learning's managerial accounting course lays out. That framing assumes you manufacture and hold inventory.
Print-on-demand breaks that assumption in a useful way. You hold no stock, so there is no beginning or ending inventory to reconcile — your projected COGS is simply the per-order supplier cost times the orders you expect. That makes the forecast cleaner, but it also hides a trap we will get to below.
Two ways to project COGS — and which fits POD
Method one: COGS as a percentage of revenue
The fastest method is to calculate your historical COGS as a share of revenue, then apply that same percentage to projected sales. If COGS has run at roughly half of revenue, you forecast future COGS at half of forecast revenue. This revenue-ratio approach is the standard shortcut GorillaROI describes for ecommerce forecasting.
It is quick and good enough for a rough plan. Its weakness: it assumes your cost structure stays frozen. If a supplier raises a blank price or you shift product mix, the ratio lies to you.
Method two: per-order bottom-up projection
The more accurate method for POD is to project order volume and multiply by your true per-order supplier cost — product price plus the shipping your supplier bills you. This mirrors how you actually get charged: one print, one ship, per order.
It catches cost changes the ratio method misses, and it ties directly to the per-order profit you care about. Use the ratio for a napkin estimate; use per-order for any decision involving money.
A worked projection for an operating store
Say you run 340 orders a month at a $31 average order value, with $2,800 a month in Meta spend. Your monthly revenue is 340 × $31 = $10,540.
Your Printify-fulfilled hoodie costs $11 for the product plus $5 shipping = $16 per order in supplier cost. Projected COGS for the month is 340 × $16 = $5,440.
That puts your COGS ratio at $5,440 ÷ $10,540 = 51.6% of revenue — a figure you will reuse for sanity checks.
Now project a growth month to 400 orders. Both methods should roughly agree:
- Per-order method: 400 × $16 = $6,400 projected COGS.
- Ratio method: 51.6% × (400 × $31 = $12,400) = about $6,398.
They match — until a supplier price change. Suppose Printify raises the blank by $1, so your per-order cost becomes $17. The per-order method instantly updates to 400 × $17 = $6,800. The ratio method still says $6,398, understating your real cost by roughly $400 for the month. (These are illustrative figures for one example store, not market data.)
This is why an operating POD seller should anchor projected COGS to the live per-order cost, not a stale percentage. Your supplier's price list — not last quarter's ratio — is the source of truth. If you are weighing a fulfillment switch as you scale, our guide on how 3PLs help scale ecommerce operations covers how those per-order economics shift.
What belongs in projected COGS — and what does not
Getting the boundary right is where most forecasts go wrong. Include the direct cost to produce and deliver the item: supplier product price, the supplier's shipping charge, and primary packaging if you add any.
Exclude your operating costs — ad spend, apps, your own time, platform subscriptions. As Sal Accounting's ecommerce COGS guide notes, mixing marketplace and fulfillment fees into COGS hides your true gross margin.
A quick way to see the distinction in a real income statement is to study how a large product company reports it — for instance, the line-by-line breakdown in Coca-Cola's Q4 2021 cost of goods sold. The categories scale down to a one-person POD store surprisingly well. For the mechanics of booking the number once orders close, see recording cost of goods sold.
The number every forecast ignores: COGS you can't recover
Here is the POD-specific trap. A standard retailer who refunds an order usually gets the item back and returns it to stock, so the loss is only shipping. For POD there is no restock — the item was printed for that one customer and cannot be resold.
That means when a POD order is refunded or charged back, the COGS you already paid your supplier is gone. Your clean "orders × per-order cost" projection quietly understates reality by every order that reverses.
A chargeback makes it worse. On Shopify Payments, US merchants pay a $15 chargeback fee that is deducted immediately and only refunded if you win the dispute, according to chargeback.io. The same source notes a lost dispute typically costs two to two-and-a-half times the order value once you add unrecoverable product, shipping, ad spend, and time.
So a realistic projected COGS adds a reversal allowance. On 400 orders, if even 2% reverse, that is 8 orders × $16 = $128 of supplier cost you will never recoup — before any chargeback fees. Build that line into the forecast instead of discovering it in your bank balance.
From projected COGS to projected profit
Projected COGS only matters because it feeds projected profit. Take the 340-order month: $10,540 revenue minus $5,440 COGS minus $2,800 Meta spend leaves $2,300 in contribution before Shopify fees and the reversal allowance.
Divide by 340 and you are making roughly $6.76 per order before those remaining costs. That per-order number is the one that tells you whether scaling ad spend to hit 400 orders is worth it — because if your per-order cost ticks up and your AOV does not, more orders can mean less profit.
Running that math by hand every time a supplier price or ad cost moves is exactly the grind PodVector AI is built to remove. Victor, PodVector AI's AI employee, connects to Shopify, your Printify, Printful, and Gelato fulfillment, and your Meta and Google Ads accounts, then computes your true per-order profit from live data — not a guess. Victor is not a dashboard you read; it is an employee that does the work, and every write action it takes is approval-gated, so you approve before anything executes.
Ready to stop projecting COGS in a spreadsheet? Put Victor to work on your store.
For the full picture of how these forward estimates connect to margin, pricing, and fulfillment decisions, start with our hub on ecommerce operations economics.
FAQs
Is projected COGS the same as a COGS budget?
Effectively, yes. A COGS budget is the formal planning document; projected COGS is the number inside it. Both answer the same question — what will it cost to produce what you expect to sell. The budget framing assumes manufactured inventory, while a POD projection is simpler because you carry no stock to reconcile.
Should I project COGS as a percentage of revenue or per order?
Use the percentage of revenue for a fast, directional estimate and per order for any real decision. The ratio method, as GorillaROI describes, assumes your cost structure stays fixed. For POD, where a single supplier price change moves every order's cost, the per-order method stays accurate when the ratio drifts.
Do shipping and fulfillment fees go into projected COGS?
The supplier's shipping charge to produce and send the item is part of COGS, because it is a direct cost of delivering that unit. Marketplace and platform fees are operating expenses, not COGS — Sal Accounting warns that blending them in hides your true gross margin.
How do refunds and chargebacks affect my projected COGS?
They raise it above your clean estimate, because a printed POD item cannot be restocked, so its production cost is lost on any reversal. A chargeback adds a $15 Shopify Payments fee on top for US merchants, refunded only if you win, per chargeback.io. Add a small reversal allowance to your forecast so these do not surprise you.
How often should I update my projected COGS?
Re-check it whenever a supplier changes prices, you add or retire a product, or your order mix shifts — and at minimum each month when you reconcile. A stale projection built on an old cost-to-revenue ratio is the most common reason a "profitable" growth month turns out not to be.