Suppose Amazon suggests a $1.35 bid for a keyword. Should you take it?
Not until you know what a click is worth to your product. In the fictional example below, a $32 order leaves $12.50 before advertising. At a 12% click-to-order conversion rate, the break-even average CPC is $1.50. But if the seller wants to keep $4 of profit per order, the target maximum average CPC is only $1.02. If conversion rate falls to 8%, that same target drops to $0.68.
That is the central calculation:
Target maximum average CPC = (pre-ad contribution margin per order − desired profit per order) × click-to-order conversion rate
Amazon describes CPC as the amount paid when a shopper clicks an ad and says the final CPC is determined through an auction using the adjusted bid plus other factors. It also distinguishes the base bid from a maximum CPC that can include dynamic or other adjustments. That makes two ceilings necessary: an economic ceiling based on your margin and a campaign-setting guardrail that accounts for bid controls. See Amazon Ads’ CPC explanation.
Use the accompanying interactive Amazon PPC maximum CPC calculator to change referral fees, fulfillment costs, return reserves, desired profit, conversion rate, placement adjustments, and dynamic uplift. The calculator is a planning model, not a profit forecast or a guarantee of campaign results.
The original calculator opens as a separate self-contained HTML file and works without external libraries.
Start with contribution margin, not a market CPC average
A category benchmark can tell you what other advertisers may be paying. It cannot tell you what your business can afford. Two sellers can compete for the same query and have completely different CPC limits because their prices, fee categories, fulfillment costs, return rates, average units per order, and profit requirements differ.
The useful starting point is contribution margin before advertising:
Pre-ad contribution margin per order = net sales per order − all variable non-ad costs per order
Then reserve the profit you do not want advertising to consume:
Allowable ad spend per order = pre-ad contribution margin − desired profit per order
Finally, convert the allowable spend per order into allowable spend per click:
Target maximum average CPC = allowable ad spend per order × conversion rate
Use conversion rate as a decimal. A 12% conversion rate is 0.12, not 12.
The formula follows from a second relationship:
Ad cost per order = average CPC ÷ conversion rate
At a 12% conversion rate, the expected number of clicks per order is 1 ÷ 0.12 = 8.33. If you can spend $8.50 to acquire an order, dividing $8.50 across 8.33 expected clicks gives $1.02 per click.
This article uses maximum average CPC for the economic result. It is a portfolio expectation across enough clicks, not a promise that every individual auction will clear at that price or that every block of 8.33 clicks will produce exactly one order.
A filled maximum CPC calculation
The following is an original fictional teaching example for a one-unit order. The figures are not Amazon benchmarks.
| Line item | Calculation | Amount |
|---|---|---|
| Ad-attributed sales per order | Selling price after seller-funded discount | $32.00 |
| Referral fee | 15% × $32.00 | −$4.80 |
| Fulfillment fee | Per order | −$4.50 |
| Cost of goods | Per order | −$8.20 |
| Inbound freight and prep | Per order | −$0.90 |
| Return/refund loss reserve | Expected per order | −$0.70 |
| Storage and other variable costs | Expected per order | −$0.40 |
| Pre-ad contribution margin | $32.00 − $19.50 | $12.50 |
| Desired profit after ads | Seller’s target | −$4.00 |
| Allowable ad spend per order | $12.50 − $4.00 | $8.50 |
| Click-to-order conversion rate | Orders ÷ ad clicks | 12% |
| Target maximum average CPC | $8.50 × 0.12 | $1.02 |
| Break-even maximum average CPC | $12.50 × 0.12 | $1.50 |
The difference between $1.02 and $1.50 matters. At $1.50, the model expects advertising to consume the entire $12.50 contribution margin. The campaign may show sales and a positive ROAS while leaving no contribution profit. At $1.02, the model preserves the planned $4 per order.
The 15% referral rate is an input for this fictional case, not a universal Amazon rate. Amazon’s US selling-fee page says referral fees vary by product category and may use a percentage or a minimum fee; it also defines the relevant total price more broadly than a product’s list price alone. Pull the actual fee category and settlement values for the ASIN instead of copying this example. See Amazon’s current selling fees.
Break-even CPC and target CPC answer different questions
A seller should normally calculate both.
Break-even maximum average CPC answers: “How much could I pay per click before the advertised order contributes zero profit?”
Break-even CPC = pre-ad contribution margin × conversion rate
Target maximum average CPC answers: “How much could I pay while preserving the profit or contribution target I chose?”
Target CPC = (pre-ad contribution margin − desired profit) × conversion rate
If desired profit is zero, the two figures are identical. If desired profit exceeds pre-ad contribution margin, target CPC is zero: the product cannot meet that profit target even with free traffic. That is a product-economics problem, not a bidding problem.
The target can be expressed as a percentage or a fixed amount. For the example, a $4 profit reserve is 12.5% of $32 in sales. A team that manages to a post-ad contribution-rate target could enter sales × desired contribution rate instead.
Reconcile maximum CPC with ACoS
Amazon defines advertising cost of sales, or ACoS, as ad spend divided by ad-attributed revenue. Its own guide says there is no single definitive “good” ACoS and connects break-even ACoS to profit margin. See Amazon Ads’ ACoS calculation guide.
For a single-order model:
Break-even ACoS = pre-ad contribution margin ÷ ad-attributed sales
Target ACoS = allowable ad spend per order ÷ ad-attributed sales
In the example:
- Break-even ACoS =
$12.50 ÷ $32.00 = 39.1% - Target ACoS =
$8.50 ÷ $32.00 = 26.6%
That produces an equivalent CPC formula:
Target maximum average CPC = average order value × target ACoS × conversion rate
For the example, $32.00 × 26.6% × 12% ≈ $1.02.
The margin-first version is safer because it forces you to build the target ACoS from actual costs. Starting with an arbitrary 25% ACoS and working backward can look precise while ignoring a high return rate, a category-specific fee, or an expensive fulfillment tier.
Keep the denominators aligned. If “sales” reflects a multi-unit basket, the cost side must reflect those units too. If shoppers buy a different SKU after the ad click, use the weighted contribution from what was actually purchased rather than assigning the advertised SKU’s margin to every attributed dollar.
Build the cost stack from settlement reality
The CPC model is only as useful as the costs beneath it. Include costs that change with a sale or can be reasonably reserved per order.
| Input | Practical treatment | Common error |
|---|---|---|
| Net sales per order | Use the sales value that matches your advertising denominator, after seller-funded discounts and with taxes handled consistently | Using list price when orders commonly sell with a coupon |
| Referral fee | Use the actual category rate, minimum, and fee base shown for the product | Assuming every category is 15% |
| Fulfillment or merchant shipping | Use the current per-order FBA fee or expected FBM shipping and handling | Using an old size tier or ignoring packaging labor |
| Cost of goods | Include all units in the average order | Entering unit COGS when units per order exceed one |
| Inbound freight and prep | Allocate freight, labeling, prep, and placement costs per sellable unit | Treating inbound logistics as “already paid” and therefore free |
| Returns and refunds | Use an expected loss reserve based on your own return rate and recovery value | Subtracting nothing, or subtracting the full refund even when inventory is recovered |
| Storage and inventory costs | Reserve expected storage, aged inventory, removal, or disposal costs where material | Looking only at the pick-pack-ship fee |
| Other variable costs | Add closing fees, promotion costs, payment-related costs, or variable software/service fees that apply | Hiding material costs in fixed overhead |
Amazon says FBA costs depend on the product and services used, and it separately identifies fulfillment, storage, returns processing, aged inventory, removal or disposal, and inbound placement costs. See the official FBA cost overview. Amazon’s guide to its Revenue Calculator also notes that common fees are included while some expenses—such as long-term storage, removal orders, and return processing—may need to be added through a miscellaneous-cost field. See How to estimate profits with the FBA Revenue Calculator.
Fixed overhead deserves a deliberate choice. The pure contribution model normally excludes costs that do not change when one more order is placed. You can reserve part of the contribution margin for payroll, software, or rent through the desired-profit input, or allocate fixed costs per order if that is how your finance model operates. Do not do both.
Conversion rate is the fragile input
An $8.50 target acquisition cost produces very different CPC ceilings at different conversion rates.
| Click-to-order conversion rate | Expected clicks per order | Target maximum average CPC |
|---|---|---|
| 6% | 16.67 | $0.51 |
| 8% | 12.50 | $0.68 |
| 10% | 10.00 | $0.85 |
| 12% | 8.33 | $1.02 |
| 15% | 6.67 | $1.28 |
| 20% | 5.00 | $1.70 |
This is why a platform-wide CPC average is weak bidding guidance. The affordable CPC doubles when conversion rate moves from 6% to 12%, even though price, fees, costs, and profit target remain unchanged.
Use a conversion rate that matches the decision level:
- For a keyword or product target, use its ad-attributed purchases divided by its ad clicks when the sample is usable.
- For a placement adjustment, use that placement’s conversion rate and economics rather than the campaign-wide average.
- For a new target, use conservative low/base/high scenarios from genuinely comparable traffic. Do not blend a branded exact-match term with a broad generic discovery target.
- After a price, listing, review, inventory, or offer change, treat the old conversion rate as provisional.
Amazon’s current Sponsored Products targeting-report documentation defines purchase rate over clicks as purchases from clicks divided by clicks. Its performance-metrics documentation also warns that order metrics can take up to 12 hours to update. Match the numerator, denominator, and attribution definition in your report, and let delayed order data mature, rather than dividing whatever two totals are easiest to find. See Targeting report for Sponsored Products and Performance metrics.
A small sample can create false confidence. One order from ten clicks displays as a 10% conversion rate, but it does not establish that the next hundred clicks will behave the same way. The calculator therefore gives a deterministic scenario, not a statistical confidence interval. Run at least a downside case and make the early bid reflect the amount of uncertainty you are willing to fund.
Translate the economic ceiling into an Amazon base bid
The number calculated from margin is not automatically the number to type into the campaign.
Amazon distinguishes the final CPC from the adjusted bid and says maximum CPC can include a base bid plus settings such as dynamic bidding or cost controls. A separate Sponsored Products control can also raise bids for eligible placements. By arithmetic, a 50% placement adjustment turns a $1 base bid into a $1.50 adjusted bid before any additional eligible uplift. See Amazon Ads’ current documentation for adjusting Sponsored Products bids.
Dynamic bids—up and down require special attention. Amazon’s current guide says it may increase or decrease bids by up to 100% for all placements and gives the example that a $1 bid could become a maximum $2 bid. It explicitly tells advertisers to be prepared for the entered bid to double. See the dynamic bidding guide.
The accompanying calculator uses this conservative planning guardrail:
Base-bid guardrail = target maximum average CPC ÷ ((1 + placement adjustment) × (1 + possible dynamic uplift))
With a $1.02 economic target, a 50% placement adjustment, and a possible 100% dynamic increase:
$1.02 ÷ (1.50 × 2.00) = $0.34
That $0.34 is not an Amazon recommendation or a prediction of the CPC you will pay. It is an original risk-control calculation that assumes both entered controls can affect the same opportunity and keeps the modeled configured bid from exceeding the economic target. Confirm the controls and limits visible in your account because products, eligibility, and platform behavior can change.
It is also deliberately conservative. A top-of-search click may be worth more than a product-page click if top of search has a demonstrably higher conversion rate. In that case, calculate a separate ceiling:
Placement-specific maximum CPC = allowable ad spend per order × placement conversion rate
Then size the base bid and placement adjustment against that placement-specific ceiling. A higher click-through rate alone is not enough; the placement must convert well enough to support its CPC.
Suggested bid, actual CPC, and daily budget are different controls
Treat a suggested bid and bid range as auction context rather than a calculation of product profitability. That makes the recommendation useful auction context, not a statement about your SKU’s margin. See Understand bidding on Sponsored ads.
A useful operating distinction is:
| Number | What it tells you | What it does not tell you |
|---|---|---|
| Suggested bid or range | Auction context supplied by the platform | Whether the click is profitable for your SKU |
| Base bid | Your starting auction input before eligible adjustments | The exact CPC every click will cost |
| Actual average CPC | Spend divided by clicks over a report period | Whether each target has the same economics |
| Daily budget | How much campaign spend can be allocated or paced | How much one click is worth |
Amazon says Sponsored Products daily budgets should be sustainable and are averaged over a calendar month; its current guide notes that spend on a given day can be up to 25% above the average daily budget while monthly spend is constrained by the average. See Sponsored Products budget basics.
Budget controls exposure, not unit economics. A $1.50 click does not become profitable because the daily budget is only $10. Likewise, raising a profitable campaign’s budget can create more opportunity, but it does not justify raising bids past the margin-based ceiling.
Use the relationship between suggested bid and your target as a decision aid:
| Auction signal | Practical response |
|---|---|
| Suggested bid is below your target maximum CPC | Start below the ceiling and increase only if profitable volume is constrained |
| Suggested bid is near your target | Test with controlled settings and watch conversion rate, actual CPC, and contribution profit |
| Suggested bid is above your target | Do not chase it automatically; improve conversion or margin, narrow targeting, accept less volume, or define a separate investment budget |
What to change when the affordable CPC is too low
There are only a few durable levers because the formula is mechanical.
Increase contribution per order. Improve price realization, lower COGS, reduce fulfillment or packaging cost, fix an incorrect fee category, lower avoidable return loss, or increase profitable units per order. Price changes can also change conversion rate, so model both sides.
Increase conversion rate. Tighten search-term relevance, improve the main image and offer, keep inventory and delivery promises competitive, separate branded from generic traffic, and route proven queries into structures where their bids can be controlled. A better conversion rate raises the CPC ceiling without changing target CPA.
Lower the profit reserve temporarily and explicitly. A launch, rank-building test, inventory liquidation, or defensive campaign may be allowed to spend above the normal profit target. Record the maximum loss, dates, and success metric. Do not label the campaign “profitable” simply because the overspend was intentional.
Use credible downstream value. Repeat purchases can justify a higher acquisition cost only when the incremental contribution from those future orders is supported by cohort data. Gross lifetime revenue is not spendable margin, and an attributed first order is not proof that every acquired customer will repeat.
If none of those levers moves enough, the rational result may be a low-volume or no-bid decision. The auction does not owe every product a profitable placement.
A margin-first bidding routine
- Refresh the SKU economics. Use current price realization, referral fee, fulfillment tier, COGS, inbound cost, return reserve, and other variable costs.
- Choose the profit objective. Calculate both break-even and target allowable ad spend per order.
- Segment conversion rate. Use target- and placement-level data where available; run low/base/high cases for uncertain traffic.
- Calculate economic CPC ceilings. Keep a campaign-wide fallback, then replace it with more specific ceilings as data improves.
- Audit the bid controls. Account for placement adjustments, dynamic bidding, rules, and any other eligible modifiers shown in the campaign.
- Compare modeled and observed results. Review actual CPC, purchases, ACoS, and contribution profit after the relevant attribution data has matured. Change one major variable at a time where practical.
The answer to “How much can I afford to pay for an Amazon click?” is not the marketplace average and not the highest bid the interface accepts. It is the contribution you can spend to acquire an order, multiplied by the probability that a click becomes that order. Everything else—suggested bids, placement multipliers, dynamic bidding, and budgets—must fit inside that economic decision or be labeled as a deliberate investment outside it.
Sources
- CPC (Cost per Click) explained — Amazon Ads. Accessed September 9, 2026.
- What is advertising cost of sales (ACOS)? Calculation and tips — Amazon Ads. Accessed September 9, 2026.
- Guide to dynamic bidding—up and down with Sponsored Products — Amazon Ads. Accessed September 9, 2026.
- Adjust Sponsored Products bids — Amazon Ads Support Center. Updated July 14, 2026; accessed September 9, 2026.
- Understand bidding on Sponsored ads — Amazon Ads Support Center. Accessed September 9, 2026.
- Budgets—the basics and best practices for Sponsored Products — Amazon Ads. Accessed September 9, 2026.
- Targeting report for Sponsored Products — Amazon Ads Support Center. Updated February 24, 2026; accessed September 9, 2026.
- Performance metrics — Amazon Ads Support Center. Updated June 30, 2026; accessed September 9, 2026.
- How much does it cost to sell on Amazon? — Sell on Amazon. Accessed September 9, 2026.
- Amazon FBA (Fulfillment by Amazon) — Sell on Amazon. Accessed September 9, 2026.
- How to estimate profits with the FBA Revenue Calculator — Sell on Amazon. Published August 27, 2025; accessed September 9, 2026.
