Sell on Amazon if shoppers already look for your brand there, if a specific set of SKUs still clears your contribution target after Amazon’s fees and ads, and if you can keep pricing consistent across channels. If none of those hold, staying off is a legitimate choice, but make it an active decision, because resellers may list your products anyway.
Why the question matters more as you scale
At a few hundred orders a month, Amazon is usually a distraction from the site, the offer and your first paid channel. As the brand grows, three things change:
- Paid social starts sending shoppers who check Amazon before they buy, because their account, payment and fast shipping are already there.
- Resellers and wholesale accounts start listing your products, often with weak images and inconsistent prices.
- DTC growth slows while acquisition costs rise, and a marketplace full of purchase intent starts to look attractive.
The mistake is treating this as one yes-or-no vote. It’s four smaller decisions: how much demand is already there, which SKUs make money there, which selling model you use, and how you’ll know whether it helped or hurt the store you already own.
Demand you’re already losing to Amazon
Before any margin math, find out how much of your demand ends up on Amazon without you. Check:
- Search Console queries. Filter for your brand name plus “amazon.” Those searchers are telling you where they’d rather buy.
- Support tickets and chat logs. Count “Are you on Amazon?” questions over recent months.
- Existing listings. Search Amazon for your brand and products. Note who sells them, at what price, with what content and reviews.
- Amazon search demand. With Brand Registry, Brand Analytics shows how often your brand terms are searched on Amazon. Without it, third-party research tools give rough estimates.
- Post-purchase survey. Add “Where else did you consider buying?” with Amazon as an option.
If these signals are faint, there’s little demand to capture. Listing then means buying category searches with ads, and it has to be judged like any other paid acquisition channel.
If the signals are strong and someone else already sells your products, you aren’t deciding whether to be on Amazon. You’re deciding whether to control how you appear there.
Margin math after fees and ads
Amazon’s cost stack looks nothing like your DTC cost stack. Selling as a third party, you typically pay:
- A referral fee on each sale, a percentage of the sale price that is around 15% in many categories
- Fulfillment fees if you use Fulfillment by Amazon (FBA), set by size and weight tier
- Storage and inbound fees for inventory in Amazon’s warehouses
- A monthly selling plan fee
- Advertising, rarely optional, since Sponsored Products ads sit above and among organic results, even on your brand searches
Compare channels on contribution per order after marketing, calculated the same way as for the store (the contribution margin guide covers the build). A hypothetical $60 product, with made-up round numbers:
| Per order (hypothetical) | DTC store | Amazon 3P with FBA | Amazon 1P |
|---|---|---|---|
| Revenue to you | $60.00 | $60.00 | $30.00 (wholesale) |
| COGS | –$15.00 | –$15.00 | –$15.00 |
| Shipping and fulfillment | –$9.00 | –$7.00 (FBA fee) | –$1.00 (freight in) |
| Fees and allowances | –$1.80 (payments) | –$9.00 (referral) | –$2.40 (allowances) |
| Storage, returns, other | –$2.00 | –$2.50 | –$1.00 |
| Contribution before marketing | $32.20 | $26.50 | $10.60 |
| Marketing per order | –$25.00 (new customer) | –$7.20 (ads at 12% of sales) | –$3.00 (ads) |
| Contribution after marketing | $7.20 | $19.30 | $7.60 |
On a first order, Amazon 3P often looks better, because the marketplace supplies intent you’d otherwise pay Meta or Google to create. DTC wins from the second order on: a repeat buyer reached by email or SMS costs very little, so that order keeps most of its $32.20. On Amazon, every repeat order pays the referral and FBA fees again, plus ads to defend your brand term. If your business depends on repeat orders, compare a customer’s first year, not a single order. The guide to increasing repeat purchase rate shows why that second order carries the economics.
Three rules keep the model honest:
- Use each SKU’s real dimensions. FBA fees step up by size tier, so a package just over a boundary costs noticeably more.
- Model ads as TACoS, not ACoS. ACoS only counts ad-attributed sales. Total advertising cost of sale (ad spend ÷ all Amazon sales) shows what the channel costs.
- Watch low-priced items. Per-unit fulfillment fees can consume most of the margin on an item under $20.
Brand control, pricing and MAP
Margin is the part founders model. Control is the part they underestimate.
Price parity. Amazon can remove the featured offer (the Buy Box) when it finds your product priced noticeably lower elsewhere, including on your own site. A deep sitewide DTC sale can cost you the Buy Box while it runs. Decide in advance whether promotions run on both channels, or whether DTC offers use bundles, gifts with purchase and exclusive SKUs instead of price cuts.
Resellers. Distributors and retail accounts that sell online can undercut you on Amazon. A minimum advertised price (MAP) policy written into reseller agreements, and enforced, sets a floor for authorized sellers. It won’t stop unauthorized ones, so pair it with tighter distribution and regular listing checks, and have a lawyer review the wording.
Content. Brand Registry requires a registered or pending trademark. It gives you more control over product detail pages, A+ Content, a Brand Store, Sponsored Brands ads and tools to report counterfeits. Without it, your brand’s page may be written by whoever listed it first.
The customer. Amazon masks buyer email addresses and restricts marketing messages, and its policies bar package inserts that steer buyers off Amazon. You get the order, not the customer. That cost never shows up on the per-order table.
1P vs 3P vs not at all
| 3P (Seller Central) | 1P (Vendor Central) | Stay off Amazon | |
|---|---|---|---|
| Who sets retail price | You | Amazon | Resellers, if they list you |
| Inventory | Yours, in FBA or your own warehouse | Amazon buys it at wholesale | Yours |
| Listing control | High with Brand Registry | Shared; Amazon has the final say | Low if others list you |
| Best fit | Most DTC brands | High-volume brands whose wholesale economics work | Premium or exclusive positioning, subscription-led models |
| Main risk | Operational load and fee creep | Price erosion and loss of pricing control | Unmanaged third-party listings |
For most DTC brands I’d start with 3P: you keep price, content and inventory decisions, and you can pull back if it doesn’t work. 1P is usually by invitation and trades control for simpler operations, which suits brands already built around wholesale.
Staying off is reasonable when exclusivity is part of the product’s value, when the model depends on subscriptions, or when distribution is tightly controlled. Even then, enroll your brand in Brand Registry and check listings regularly, so “not on Amazon” doesn’t become “represented badly on Amazon.”
Selecting which SKUs to list
You don’t need the whole catalog on Amazon. Pick SKUs that survive the fees, match existing demand and don’t pull away your best repeat buyers:
- Contribution after referral fee, FBA and a realistic ad budget stays above your target
- The SKU already shows Amazon demand: brand searches, reseller listings, customer questions
- Size and weight sit comfortably inside an efficient FBA size tier
- It isn’t your core replenishment or subscription SKU, the one that drives DTC repeat orders
- It isn’t a new launch you want to build reviews and data for on your own site first
- It isn’t fragile, hazmat or prone to returns that would drag down ratings
- You can keep it in stock, because stockouts hurt organic rank on Amazon
Two assortment moves help. Amazon-specific packs, multipacks or bundles with their own identifiers, make price comparison with your site harder and spread fees across a larger order. And entry products on Amazon, depth at home: list what introduces the brand, and keep the full range, refills and subscriptions on your site.
Measuring the effect on your DTC store
The question: does Amazon add net new customers and contribution, or move existing demand to a channel where you pay more and own less? Amazon’s dashboard won’t tell you, so set up measurement before launch.
- Run a SKU holdout. List about half of your eligible SKUs and hold back comparable ones for 8-12 weeks. Compare the DTC sales trend of listed versus held-back SKUs against the pre-launch baseline.
- Watch repeat behavior by cohort. If returning customers start reordering on Amazon, DTC repeat rate falls for the cohorts that bought listed SKUs.
- Read new-to-brand metrics carefully. Amazon Ads reports new-to-brand orders, but they’re based on Amazon purchase history only, so a loyal DTC customer buying on Amazon for the first time still counts as new.
- Tag off-Amazon traffic. Amazon Attribution lets you tag social, search and email links that send people to Amazon and see the resulting sales. Eligible brands can also earn a bonus on those sales through the Brand Referral Bonus program.
- Ask. Add “Have you ever bought from us on Amazon?” to your DTC post-purchase survey.
| Signal | Points to halo | Points to cannibalization |
|---|---|---|
| DTC sales of listed vs held-back SKUs | Move together or listed SKUs rise | Listed SKUs fall behind |
| DTC repeat purchase rate | Steady | Falls in cohorts buying listed SKUs |
| Branded Google searches and direct traffic | Rise after launch | Flat while Amazon brand searches climb |
| Total contribution after marketing, both channels | Grows | Flat while combined revenue grows |
The last row is the scoreboard: channel revenue can rise while the business earns less. Building this model and the launch measurement is part of my ecommerce growth work, alongside paid, retention and CRO on the DTC side.
Get it built
If you’re weighing Amazon, or already on it without knowing whether it helps or hurts, the Growth Audit models channel margin by SKU and sets up the measurement plan. It’s $1,500 fixed and credited if we continue. See pricing or get in touch.