Amazon Sellers' CFO Guide: ACOS, Inventory & Cash Flow

Key Takeaways

ACOS, inventory and cash are a single operating loop, not three dashboards. The ad you run today consumes inventory you paid for 90 days ago and gets settled by Amazon 14 days from now. Reading them as separate reports is why profitable-on-paper Amazon brands still miss reorders.
ACOS is a constraint, TACOS is the truth, break-even ACOS is the lever. ACOS only covers ad-attributed sales. Total advertising cost of sale (TACOS) measures ad spend against all revenue, and your break-even ACOS, set by contribution margin after FBA fees, tells you how hard you can actually push.
FBA fees and landed cost belong in CM1 before you trust any margin. Referral fees, fulfilment fees, inbound and storage fees and returns processing can swallow 30-40% of an Amazon order. A margin that ignores them is fiction, and every ACOS decision built on it is wrong.
The cash conversion cycle decides how fast you can grow. Money goes out for a PO, sits as inventory for months, then comes back after Amazon's settlement lag. A brand can be profitable and still run out of cash, because growth pulls cash forward into inventory faster than profit replaces it.
These numbers are downstream of decisions, so the work is upstream. Which SKU to reorder, how hard to push ACOS this month, and how to finance the next inventory cycle are operating calls. A 13-week cash model and SKU profit autopsy turn the metrics into next week's decision, not last quarter's report.

If you sell on Amazon, your three hardest numbers are ACOS, inventory and cash flow, and the trap is treating them as three separate problems with three separate dashboards. They are one loop. The ad you run today is spending against inventory you bought 90 days ago, and the revenue gets settled by Amazon two weeks from now. This guide teaches what each number actually means for an FBA seller, walks a worked example tying ad spend to margin to cash, and shows why the real work is upstream, in the weekly decision, not in the report you read after the month closes.

Why ACOS, inventory and cash flow are one system

Most Amazon finance advice treats advertising, inventory and cash as separate disciplines. For a real FBA business they are a single operating loop, and the loop is what trips brands up.

Here is the loop. You commit cash to a purchase order, often a deposit months before the stock arrives. That cash becomes inventory sitting in FBA, accruing storage fees, for anywhere from 60 to 180 days. You run ads to sell it through, and your ACOS decides how much margin each advertised sale keeps. Amazon then settles roughly every 14 days, net of fees and reserves, and the cash finally comes back, at which point you reorder and the loop starts again.

Read as separate reports, each number can look fine while the business quietly breaks. ACOS can be on target while the SKU it is selling has negative margin after FBA fees. Inventory turns can look healthy while a long cash conversion cycle starves the next reorder. The point of looking at them together is that the decision in one always lands in another: push ACOS harder and you sell through faster but pull cash forward into the next PO sooner. Cash, profit and revenue are downstream of these decisions, so the work is connecting them, not reporting them in isolation.

ACOS, TACOS and break-even ACOS: define them so they are usable

These three advertising metrics get quoted constantly and defined loosely, which is how brands end up chasing the wrong target.

ACOS (advertising cost of sale) is ad spend divided by the revenue those ads directly generated. A 25% ACOS means you spent $25 in ads for every $100 of advertised sales. It only measures the advertised slice, so it tells you how a campaign is performing, not how the business is performing.

TACOS (total advertising cost of sale) is ad spend divided by total revenue, organic and advertised together. This is the more honest health metric. If TACOS falls while sales grow, your ads are earning organic rank and the whole listing is getting more efficient. If TACOS climbs, you are buying sales you used to get for free. Watch TACOS for the trend of the business.

Break-even ACOS is the one that actually drives the decision. It is the ACOS at which an advertised sale makes zero profit, and it is set entirely by your contribution margin after FBA fees and landed COGS. If a SKU keeps 35% margin after all Amazon fees and product cost, its break-even ACOS is roughly 35%. Any ACOS below that earns money; above it, you are paying to lose money. There is no universal "good ACOS," only your break-even ACOS per SKU, which is why the fee and margin work below has to come first.

FBA fees and landed cost: build CM1 before you trust any margin

Break-even ACOS is only as good as the margin underneath it, and on Amazon that margin is heavily eaten by fees most brands underweight.

A single FBA order can carry a referral fee (commonly around 15% of sale price), a per-unit fulfilment fee, monthly storage fees that spike in Q4, long-term storage penalties on slow stock, inbound placement fees, and returns processing. Stacked together, Amazon fees can consume 30 to 40% of an order before you have paid for the product at all. Add landed COGS, the true cost in the warehouse including product, inbound freight, duties and per-unit handling, and the picture changes sharply.

The fix is to build the contribution-margin ladder properly:

CM1 = sale price minus landed COGS minus all Amazon fees. Referral, fulfilment, storage, inbound and returns belong here, not in a vague overhead bucket. This is your true post-fee product margin.
CM2 = CM1 minus variable advertising (your ACOS spend per order). CM2 is what an advertised order actually contributes after you paid to win it.

Your break-even ACOS is simply CM1 expressed as a percentage of price. Skip the fees and CM1 is overstated, break-even ACOS looks generous, and you happily scale ads on a SKU that bleeds on every unit. This is the single most common way Amazon brands convince themselves a losing product is a winner.

The cash conversion cycle: why profit does not equal cash

You can do all the ACOS and margin math correctly and still run out of money, because profit and cash are not the same thing on Amazon. The gap is the cash conversion cycle.

The cash conversion cycle is the time between paying for inventory and getting the cash back from selling it. For an FBA seller it is long and lumpy: a supplier deposit goes out, then production and freight, then 60 to 180 days of inventory sitting in FBA, then sell-through, then Amazon's roughly 14-day settlement with reserves held back. Cash leaves the business months before it returns.

Growth makes this worse, not better. Every time you grow, you reorder more, which pulls more cash forward into inventory before the previous batch has fully converted. A brand growing 40% can be more profitable and more cash-starved at the same time, because profit accrues slowly while the next PO demands cash now. That is why a P&L showing healthy margins is not enough. You need a rolling 13-week cash model that maps when cash actually leaves for deposits and freight against when settlements actually land, so you can see the reorder you cannot afford before you place it, not after the payment bounces.

A worked example: tying ACOS, inventory and cash together

Take one Amazon SKU selling at a $50 price.

Sale price: $50
Landed COGS: $12 (product, freight, duties, inbound)
Amazon fees: referral $7.50 (15%), fulfilment $5.50, storage and returns reserve $2.50 = $15.50
CM1 = $50 minus $12 minus $15.50 = $22.50, a 45% post-fee margin

So this SKU's break-even ACOS is 45%. Any advertised sale under 45% ACOS makes money. The brand targets a 25% ACOS to keep healthy contribution:

Metric	Value	What it means
Break-even ACOS	45%	The ceiling, set by CM1
Target ACOS	25%	Spend $12.50/order on ads
CM2 per order	$10.00	$22.50 CM1 minus $12.50 ad cost
TACOS (ads vs all revenue)	16%	Falling as organic rank builds

At a 25% ACOS the SKU keeps $10 of contribution per advertised order, and TACOS at 16% and falling says the ads are also lifting organic sales. That looks like a clear "scale it" call. Now add the cash loop. Say this SKU sells 1,000 units a month, so it consumes $12,000 of landed COGS in inventory every month, paid as a deposit roughly 90 days before those units sell. Pushing ACOS down to 18% to grow faster sells through quicker, but it also pulls the next $12,000 PO forward by weeks. The decision is not "scale ads because ACOS is good." It is: scale ads only as fast as the 13-week cash model says you can fund the reorder the faster sell-through triggers. The ACOS decision and the cash decision are the same decision.

Why this is an operating job, not a reporting job

Every number above can be reconciled by a competent bookkeeper after the month closes. That is reporting, and it is genuinely valuable, because clean Seller Central books are the raw material everything else needs. But the value is not in recording that a SKU lost money in March. It is in seeing the cash conversion cycle tighten in week two and changing the reorder or the ad pace before the cash is gone.

That is the operator lens, and it is the line between a scorekeeper and a partner:

Most CFOs keep score. We help you win. An operational CFO, not an accounting one: we tell you what to do next, not just what happened.

Eightx (eightx.co)

On Amazon specifically, the decisions that produce the numbers are which SKU to reorder or kill from the profit autopsy, how hard to push ACOS this month against break-even, and how to finance the next inventory cycle before the PO is due. The same discipline runs the whole brand as one system, because the growth lever and the risk lever are the same lever:

Contribution margin dollars and your maximum acceptable CAC are what actually grow a business faster.

Matt Putra (eightx.co)

Done well, the SKU-level profit autopsy, break-even ACOS and the 13-week cash model sit inside the weekly operating rhythm and change next week's decision. That is what a fractional CFO is for on Amazon: not a prettier dashboard, but the call on what to do next while there is still time to do it.

How to build this for your Amazon brand, in order

If you are starting from a Seller Central report and a gut feel for ACOS, here is the build order that gets you to decisions fastest.

Pull every Amazon fee per SKU. Referral, fulfilment, storage, inbound and returns. Most brands underweight storage and returns badly.
Build CM1 per SKU by subtracting landed COGS and all Amazon fees from price. This is your true post-fee margin.
Set break-even ACOS per SKU as CM1 over price. Now you know the real ceiling for each product, not a blended guess.
Track TACOS, not just ACOS, so you can see whether ads are building organic rank or just renting sales.
Run a SKU profit autopsy to separate winners, bleeders and zombies, and decide what to reorder, reprice or kill.
Map the cash conversion cycle for each SKU: deposit timing, freight, FBA hold, sell-through and settlement lag.
Tie it to a rolling 13-week cash model so the ACOS and reorder decisions are checked against the cash you will actually have when the next PO is due.

The last step is the one brands learn the hard way. Your unit economics tell you whether a SKU is worth scaling. The cash model tells you whether you can afford to scale it right now. On Amazon, where the cash cycle is long and growth pulls cash forward, you need both in the same room, every week.

Frequently asked questions

what is a good acos for amazon sellers?

There is no universal good ACOS, because the only ACOS that matters is your break-even ACOS, which is set by your contribution margin after FBA fees and landed COGS. If your margin after all Amazon fees is 35%, your break-even ACOS is roughly 35%, and any ACOS below that on profitable products earns money. A 20% ACOS can be too high on a thin-margin SKU and a 40% ACOS can be fine on a fat-margin one. Calculate break-even per SKU before judging any ACOS number.

what is the difference between acos and tacos?

ACOS (advertising cost of sale) is ad spend divided by the revenue those ads directly generated, so it only measures the advertised slice. TACOS (total advertising cost of sale) is ad spend divided by total revenue, including organic and repeat orders. TACOS is the more honest health metric: a falling TACOS while sales grow means ads are building organic rank and the brand is getting more efficient overall. ACOS tells you how a campaign is doing; TACOS tells you how the business is doing.

how do fba fees affect amazon profitability?

FBA fees stack up fast: a referral fee (commonly around 15%), a per-unit fulfilment fee, monthly and long-term storage fees, inbound placement fees, and returns processing. Together they can consume 30-40% of an order before product cost. For accurate unit economics, fold every Amazon fee plus landed COGS (product, freight, duties, inbound) into CM1, then layer ad spend for CM2. Skipping the fees overstates margin and leads to overspending on ads and reordering SKUs that actually lose money.

why is cash flow so hard for amazon sellers?

Because the cash conversion cycle is long and lumpy. You pay a supplier deposit, wait weeks for production and freight, hold inventory in FBA for 60-to-180 days, and only then sell through, after which Amazon settles roughly every 14 days and holds reserves. Cash goes out long before it comes back, and growth makes it worse, because every reorder pulls more cash forward into inventory. That is why profitable Amazon brands still hit cash crunches and need a rolling cash model, not just a P&L.

does an amazon seller need a fractional cfo or just a bookkeeper?

A bookkeeper reconciles Seller Central settlements and keeps the books clean, which is necessary raw material. A fractional CFO uses those books to make decisions: which SKU to reorder or kill, how hard to push ACOS this month, how to finance the next inventory cycle before the PO is due. If your pain is accurate reporting, a strong A2X-native bookkeeper is enough. If your pain is which decision to make next, that is an operator-CFO job, which is where a firm like Eightx fits.