Is Amazon Subscribe & Save profitable? The break-even math for sellers

A shopper subscribes to your body lotion. Two deliveries go out. Then they cancel.

Your dashboard counts one subscriber won and one lost. It reads like a wash.

It was not. You funded a discount on both orders, and that shopper would probably have bought once anyway.

They left you less than a plain buyer who never signed up. The average hides it, because the loyal subscribers carry the program.

Amazon Subscribe & Save is profitable for a seller only when subscribers place enough extra orders to pay for the discount. You give up a fixed slice of margin on every subscription order, so the program earns money only if it adds more orders than it discounts.

Most articles answer this question with a lifetime value multiple and no source. We think the better answer is a ratio you can compute from your own numbers in ten minutes. Below: how the program works for sellers, the break-even math with illustrative numbers, the traps that make subscriptions look better than they are, and a way to decide.

How Subscribe & Save works for sellers

Subscribe & Save lets a shopper sign up to get a product on a schedule, such as every six weeks. They pay a lower price on each delivery. Amazon’s terms say customers can cancel at any time or skip their next delivery before it ships (Subscribe & Save terms). They can also change how often deliveries come.

Who funds which discount

The discount comes in two layers (Subscribe & Save program page):

• The seller-funded layer. Products enroll at a 0% seller-funded discount. You can raise that to 5% or 10%. You pay it on every subscription order.
• The Amazon-funded layer. Amazon says it funds a 5% discount on subscription orders of five items or more. That is how a shopper reaches “up to 15%”: your 10% plus Amazon’s 5%.

So a shopper who subscribes to one bottle of body lotion at the 10% tier gets 10% off, and you fund all of it. On a subscription order of five or more items the shopper gets 15%, and Amazon says it funds the extra 5%.

Amazon says there are no extra selling costs for the program (Subscribe & Save program page). You pay the usual referral and FBA (Fulfillment by Amazon) fees. The referral fee is a percentage of the total price the buyer pays (Amazon selling fees). That matters for the math: a lower price also means a slightly lower referral fee.

Who can join

Amazon’s seller pages list these requirements (Amazon seller blog):

• You are a Brand Representative for a brand in Amazon Brand Registry.
• A strong fulfillment history and an in-stock rate above 90%.
• An average selling price without exaggerated increases, and a buyable offer.
• A category Amazon allows, such as Beauty, Grocery, Kitchen or Pet Supplies.

Automatic enrollment is only for products sold through FBA. Merchant-fulfilled products need a request per product and must meet shipping and defect targets for three months in a row.

Where the numbers show up

Seller Central has a Subscribe & Save page for managing discounts, reviewing metrics and planning inventory. For software, Amazon’s Replenishment API (a data feed for subscription metrics) returns totals by period and by product (Replenishment API model). The useful ones for profit questions:

• Active subscriptions at the end of a period, shipped subscription units and subscription revenue.
• Subscriber retention: the share of subscriptions still active 30 and 90 days after they started.
• Revenue split by subscriptions with more than two deliveries, active ones with one delivery, and ones cancelled after one delivery.
• Revenue share and sign-up conversion by seller-funded tier.
• The share of new subscriptions that came from coupons.
• Revenue lost because the product was out of stock.

The API covers only the trailing two years. It also reports totals, not individual buyers.

What one order actually leaves you

Start with what one order leaves you before advertising. This is contribution margin: revenue minus the costs that grow with each sale. If you have not worked yours out, start with contribution margin on Amazon and Shopify.

Illustrative numbers only. A bottle of body lotion lists at $24 and ships through FBA.

A 10% discount on a $24 item takes $2.40 off the price. It costs you $2.04, because the referral fee shrinks with the price. That $2.04 is 19% of what a full-price order leaves you. A 10% price cut is a much bigger cut to margin.

How many extra orders a subscriber must place

Divide full-price margin by subscription margin:

• 10% tier: $10.50 ÷ $8.46 = 1.24. Subscribers must place 24% more orders than they would have placed anyway.
• 5% tier: $10.50 ÷ $9.48 = 1.11. They must place 11% more.

Now make the product thinner. Say cost of goods and fees leave only $6.00 on a full-price order. At the 10% tier the same $2.04 discount leaves $3.96, and the ratio jumps to 1.52. Subscribers now need 52% more orders. The thinner your margin, the harder the same discount is to earn back.

One-time buyer vs subscriber over 12 months

The ratio tells you the bar. This table shows what clearing it looks like. Illustrative numbers again, at the 10% tier with the lotion above.

The one-time buyer is the average first-time buyer who never subscribes: the first order plus some repeat orders, 2.2 orders in a year. The subscriber is the average of everyone who signed up on a six-week schedule. That is up to 9 deliveries in a year. In this example, 30% cancel after the first delivery, 20% after two or three, and half stay but skip some, averaging 7.4 orders. The average comes to 4.5 orders.

On this table the subscriber is worth $14.97 more. That gap is the headline most pages stop at. It hides two things.

First, the average hides a split. A subscriber who cancels after two deliveries leaves $16.92. That is less than the one-time buyer who never signed up. Break-even is 2.2 × 1.24 = 2.73 orders. Every subscriber below that line cost you money compared with a normal buyer.

Second, the comparison assumes the subscriber would have behaved like an average one-time buyer. They probably would not. People who subscribe tend to be the ones who were going to buy again.

Where it goes wrong

These are the traps we see when brands read their own subscription data.

Counting subscribers who never ship a second order

A sign-up is not a subscriber until the second delivery ships. The first delivery would often have been a normal order anyway, now at a discount. Active subscription counts include people who signed up last week and will cancel before the next box. Amazon’s revenue-by-deliveries metric splits out one-delivery and cancelled-after-one subscriptions, so use it. Count subscribers by deliveries shipped, not by sign-ups.

Coupons make this worse. Amazon says Subscribe & Save coupons apply to the first delivery of a new subscription. A shopper can sign up for the coupon and cancel right after it ships. Check the share of new subscriptions that came from coupons, and track those cohorts on their own.

Subscription figures that restate after the fact

In the data we read, recent Subscribe & Save periods keep changing after they close. A week pulled the day after it closes can come back short. The same week pulled a week later stops moving. We have also seen a period requested on its last day come back incomplete, with nothing saying so.

That creates a quiet double count. Say you pull weekly and add each pull to a history table. The same week now sits in the table twice, once short and once complete. Add them up and subscription revenue looks like it is growing. Keep one copy per period: the newest pull. Wait at least a week after a period closes before you trust it.

Judging a young cohort’s retention too early

A cohort is a group of subscribers who signed up in the same month. On a six-week schedule, a cohort that is five weeks old has not had its second delivery yet. It will show almost no repeat orders, and that means nothing. Amazon’s own retention metric looks at 30 and 90 days after sign-up. Compare cohorts only at an age every one of them has reached, and leave younger ages blank.

Discounts stacking with coupons

Amazon says seller promotions and discounts can be added on top of a Subscribe & Save discount (Amazon seller blog). So a coupon on a subscription product cuts into the same order your tier already discounts. If coupons run on your subscription products most of the time, the 1.24 ratio above is too low. Work out the ratio again using the price the buyer really pays with the coupon, and any cost of running it. Amazon reports the share of revenue from products with coupons. Look at it next to your tier.

Ignoring subscriptions that would have repurchased anyway

This is the biggest one. Every subscription order carries the discount, including orders the shopper would have placed at full price. Loyal buyers are the most likely to subscribe. So a large share of subscription orders may be orders you already had, now $2.04 cheaper.

Run the table again with that in mind. Suppose subscribers would have placed 3.0 orders a year at full price anyway. That is $31.50 of contribution. Subscribing raised it to $38.07, a gain of $6.57, not $14.97. At 3.63 orders a year without subscribing, the gain is zero.

How to settle it with your own numbers

These steps use numbers most Amazon brands already have.

1. Work out contribution per full-price order for each subscription product. Include referral fee, FBA fee, landed cost of goods and storage.
2. Compute the break-even ratio for each tier. Discount cost per order is price × seller discount × (1 − referral fee rate). Ratio = full-price margin ÷ (full-price margin − discount cost).
3. Build subscriber cohorts by sign-up month. For each cohort, count deliveries shipped per subscriber at 90, 180 and 365 days. Fill in an age only once the whole cohort has reached it.
4. Build two comparison groups for the same months and product. One is all first-time buyers who never subscribed. The other is only those who came back for a second order. Count their orders at the same ages. The Amazon customer lifetime value guide covers how to group buyers by first order.
5. Compare. Divide subscriber orders by each group’s orders at the same age. Against all one-time buyers the result flatters the program. Against repeat buyers it is strict. The truth sits between them. If even the flattering number is below your break-even ratio, the discount is not paying.
6. Check what sits on top. List the coupons and deals running on subscription products, and the share of sign-ups that came from coupons.
7. Record each period only after it settles. Pull Subscribe & Save data at least a week after a period closes, and keep only the newest copy of each period.
8. Revisit every quarter. Re-read each cohort at its next fixed age. If the ratio on a thin-margin product stays under break-even, test the 5% tier. Amazon’s pages do not spell out how a seller’s tier change reaches current subscribers, so check in Seller Central before you make it.

One more thing works in the program’s favor. A subscription reorder needs no ad click, while some one-time repeat orders come back through your own ads. If your repeat buyers often return through sponsored ads, count that ad cost in step 5. The first order’s ad cost is the same either way, so it does not change the comparison. For that side of the math, see first-order profitability and CAC payback.

Amazon’s own claim is that a 10% to 15% discount “can drive up to a 1.8x increase in conversion” (Amazon seller blog). Conversion is not profit. The break-even ratio is how you find out whether the extra sign-ups are worth what they cost.

So do not decide the program as a whole. Decide it one product at a time. Start with your thinnest-margin subscription product. That is where the same discount is hardest to earn back, and where a tier change is worth testing this quarter.

Synthesis joins Subscribe & Save, Amazon orders and cost data into one warehouse per brand. Every number carries its caveats, such as “this cohort is too young to judge repeat rate”, so a five-week-old cohort is not read as a failure.