Amazon Coupons and Deals in 2026: Calculate Profit First

Quick answer: Amazon promotions include coupons, price discounts, and event deals, each with an upfront and variable fee. The true cost isn't only the customer discount, it includes promotion fees, higher advertising cost during the event, incremental returns, and the contribution lost on units that would have sold at full price anyway. Calculate break-even incremental units before submitting the campaign.

The sales graph looked fantastic. A home-goods seller ran a promotion during a major Amazon event. Units sold jumped from 18 a day to 74. Revenue doubled. The team celebrated before the settlement report arrived. The discount had removed $4.50 per unit. Advertising cost rose because clicks became more expensive. The promotion carried an upfront fee and a variable fee. Return rate also increased because the campaign reached a broader group of buyers. She'd sold four times as many units and made less contribution than during an ordinary week. Promotions create visible revenue and hidden arithmetic.

The 2026 promotion landscape

Amazon's promotion names, fees, and eligibility vary by marketplace, event, and date. For Prime Day 2026, Amazon's public seller guidance lists a $100 upfront fee plus 1.5% of sales for certain price discount or Best Deal campaigns, with the variable fee capped at $5,000. Coupon guidance has used a $5 upfront fee plus 2.5% of coupon sales in the US store. These details are time-sensitive and must be verified on the campaign submission page. The important habit isn't memorizing one fee, it's collecting the complete fee stack before committing inventory.

The complete promotion cost formula

Start with ordinary contribution per unit: normal selling price minus product cost, Amazon fees, fulfillment, expected ad cost, and expected return cost. Then calculate promotion contribution per unit the same way, but with the promotional price and the promotion's own variable fee, higher ad cost, and higher return allowance. Multiply promotion contribution per unit by promoted units sold and subtract upfront promotion fees for total promotion contribution. The last step matters most: subtract expected contribution without the promotion to get incremental promotion contribution, since some promoted units would have sold at full price anyway.

A worked coupon example

At normal economics: selling price $39.99, product and inbound cost $12.00, referral and fulfillment fees $11.20, expected ad cost $4.00, expected return allowance $1.20, normal contribution $11.59. During a 15% coupon promotion: promotional price $33.99 after a $6.00 discount, coupon variable fee about $0.85, upfront fee $5, advertising rises to $5.20 per order, return allowance rises to $1.50. Promotion contribution per unit works out to roughly $4.14, using the current fee calculator for the exact Amazon fee portion. The product still contributes, but only $4.14 instead of $11.59. If 100 units would have sold normally, those baseline sales lose $7.45 contribution each, a $745 opportunity cost the promotion must overcome.

Break-even incremental units

Required incremental units equals contribution lost on baseline units plus fixed fees, divided by promotion contribution per incremental unit. With $745 lost plus a $5 upfront fee, divided by $4.14 contribution per incremental unit, that's roughly 182 incremental units needed, beyond the expected 100 baseline units, just to replace the sacrificed contribution. That's a much harder target than "sales should double."

The cannibalization question

Promotion dashboards show promoted sales. They don't automatically tell you how many were incremental. Estimate a baseline using the same weekdays from recent weeks, seasonality-adjusted prior periods, and the organic trend before the event. No baseline is perfect, but a reasonable estimate beats treating every promoted order as new demand.

Reference pricing can make the campaign ineligible

Amazon uses reference-price rules, including Typical Price and Was Price concepts, to determine whether a discount is genuine enough to display. In May 2026, Amazon updated how certain non-advertised price discounts are treated in Typical Price calculations. This creates real risk: a recent low price makes the required event price lower than expected, or a test discount damages future promotional eligibility. Build a price calendar at least 60 to 90 days before major events, recording normal price, coupons, business price, and planned promotions.

Prime Day is not automatically profitable

Prime Day works better when the product has healthy contribution before discount, conversion is already strong, and inventory is sufficient but not excessive. It works poorly when the product barely profits at normal price, the seller uses a steep discount to fix a weak listing, or stock runs out mid-event. Both patterns are common in seller forum discussions around Prime Day, sometimes for the same event.

Inventory is part of the promotion cost

Event stock requirement equals expected baseline demand plus expected incremental demand plus safety stock, minus usable inbound arriving before the event. Also consider FBA receiving delays, storage cost if the event underperforms, and aged inventory if leftover stock remains. A promotion is a demand bet and an inventory bet at the same time.

Advertising during promotions

Sellers often increase discount and ad spend together, which makes the event impossible to evaluate. Plan campaigns in three groups: defend (brand and high-converting exact terms), expand (relevant category and competitor targets for incremental reach), and test (new targets with a defined learning budget), covered further in Amazon ads out of budget. Track contribution after both ad spend and promotion fees, since a low ACOS can still accompany weak profit when the selling price is heavily discounted.

The product that should not have joined

A seller had 900 units of a slow kitchen gadget with aging stock, so he submitted a major-event deal to "clear it out." The product had three problems: the main image didn't explain the use, return reasons showed customers expected a larger size, and normal contribution was only $3.80. The promotion increased clicks and units, but discounted contribution went negative after ads and returns. He would have lost less by lowering ad spend, improving the dimension image, and using a controlled clearance price outside the event. Promotions amplify the existing product. They don't repair it.

Post-event analysis

Review four windows: pre-event baseline, the event period, seven days after, and thirty days after. Measure units, contribution, ad spend and TACOS, organic sales, return rate, and Buy Box percentage. A promotion may look weak during the event but create repeat demand later, or look strong during the event and cannibalize the following month.

---

EcomSanity doesn't create Amazon promotions or calculate every campaign fee automatically. It helps sellers monitor the operating result, sales velocity, Buy Box, conversion, days of inventory, and return rate, before, during, and after the promotion, comparing promoted ASINs against a control group of similar non-promoted products to separate account-wide event demand from the effect of the deal itself.