How to Calculate Your Real Breakeven ACoS on Amazon (CPG LTV Edition)

Most Amazon PPC playbooks open with the same formula: breakeven ACoS equals profit per unit divided by sale price. For a $25 item with $9 contribution margin, that is 36%. Set campaigns below 36%, scale above with caution, move on.

For CPG brands on Amazon, that math is wrong – not slightly off, structurally wrong. It assumes every customer buys exactly once. Consumable categories rarely work that way. Subscribe & Save, repurchase cycles, and pantry-loading behaviour mean the first click is often the start of a relationship, not the whole transaction.

The Math Most Brands Use (And Why It Breaks for CPG)

Standard formula:

Breakeven ACoS = Profit per Unit ÷ Sale Price

Example: $9 profit ÷ $25 price = 36% breakeven ACoS.

That 36% is a valid ceiling only if the customer never comes back. For coffee, supplements, pet food, household consumables – categories where Amazon actively pushes replenishment – single-purchase breakeven systematically understates how much you can afford to pay for acquisition.

The practical consequence: your team throttles campaigns that are profitable on LTV while competitors who model repeat purchase capture NTB share you walked away from.

The LTV-Adjusted Breakeven ACoS Formula

Replace single-unit profit with expected profit across the customer relationship:

Breakeven ACoS = (Profit per Unit × LTV Units) ÷ Sale Price

Same example with 2 expected units per customer: $9 × 2 ÷ $25 = 72% breakeven ACoS. At 72% ACoS you break even accounting for the full relationship – not just the first box.

Your competitor, calculating at 36%, thinks you are overspending. You are not – you are just using different time horizons.

This connects directly to the CLTV metric in 5 Ultimate Metrics Every CPG Brand Must Track. LTV units are the input that makes PPC math honest for consumables.

Setting a Target ACoS Using Cohort Data

Breakeven is a ceiling, not a target. Mature CPG operators pick the month by which they want cumulative profit to cross zero – then back into the ACoS that gets them there.

Cohort targeting workflow:

1. Cohort data shows most customers reach 1.5 units by Month 4
2. Target ACoS = ($9 × 1.5) ÷ $25 = 54%
3. Result: break even at Month 4; profit from remaining LTV (units 1.5 → 2.0+) is upside

Aggressive growth teams accept higher ACoS early with longer payback windows. Cash-constrained teams tighten the window. The formula stays the same – only the cohort horizon changes.

Worked Example: Month 4 Break-Even

At 54% target ACoS with 2.0 LTV units and $9 per-unit profit on a $25 item:

• Month 1: losing money per order – acquisition spend exceeds first-order margin
• Month 4: cumulative units reach 1.5 – you break even on the customer
• Month 12: total profit = $9 × (2.0 − 1.5) = $4.50 per customer beyond break-even; ~18% margin on the relationship

This timeline is why NTB investment and S&S penetration belong in the same conversation as PPC targets. If repeat behaviour weakens, your LTV-adjusted breakeven collapses – and campaigns that looked smart become margin leaks.

The Three Inputs You Need

LTV-adjusted breakeven ACoS is only as good as its inputs. Lock these three before changing campaign structure:

1. Profit per unit – after COGS, FBA fees, referral, inbound, storage, and an returns allowance. Reconcile against settlement reports monthly; calculators lie when fees shift.
2. LTV units – from cohort analysis at 3/6/12-month horizons. Use units, not revenue, when promos distort ASP.
3. Sale price – the realised average selling price, not the list price on the listing. Coupons and subscribe discounts move this number.

Operationalizing Target ACoS Across Campaigns

One blended target ACoS across all campaigns hides bad decisions. A practical structure:

• NTB acquisition campaigns: target toward cohort break-even horizon – often the highest allowable ACoS
• Category conquest: cap below LTV breakeven unless query intelligence shows high repeat propensity
• Branded defence: lower ACoS acceptable – margin is protected and LTV is already partially realised

Encode thresholds in your optimisation SOPs – the same discipline described in Claude prompts for Amazon optimization, but applied continuously rather than once a quarter in a spreadsheet.