should you sell on amazon or on your own site? the honest arithmetic and the honest politics

business · Sep 17, 2026 · 5 min read

this question is usually asked like a sports match: amazon versus my site, pick a side. the honest answer is that they are different instruments, and most sustainable shops end up using both on purpose.

what amazon actually gives you

  • buyers, today. amazon is where the shopping search already happens. you do not build traffic; you rent a stand in the busiest mall on earth. for a new product with no audience, this is the difference between selling this month and selling someday.
  • the trust default. strangers buy from amazon at two in the morning without thinking. handing them your own domain requires that trust be built first.
  • logistics, optionally. fulfilment by amazon turns your warehouse problem into their warehouse problem, for a fee that must go into the margin arithmetic.

what amazon actually takes

  • the cut. referral fees, fulfilment fees, storage fees, advertising if you want to be seen among competitors selling similar things. depending on category and price point, 30-50 percent of the sale is gone. some products survive this. some never did and their sellers have not done the math yet.
  • the customer relationship. the buyer is amazon's. you often cannot email them, cannot market to them, cannot recognize them when they come back. you get revenue; you do not get a customer.
  • the rules. account suspensions, policy changes, sudden competition from sellers of identical goods — sometimes from amazon itself. you are a tenant with excellent foot traffic and no lease.

what your own site gives and takes

your own store gives you the opposite trade: full margin (minus payment fees), the customer's email, the data, the relationship, the brand — and it takes traffic-building, which is slow and work-shaped: search, content, social, the audience you build. it also takes trust-building, because a stranger must believe a domain they have never seen. the honest costs of building it are in quanto costa un sito web and the setup path is in how do i start an online store.

the answer, finally

  • if you have a product with mass demand and no audience: start on amazon, run the margin math with brutal honesty, and treat the fees as rent for a customer-acquisition machine.
  • if you have an audience — even a small loyal one: start on your own site; sell where your people already are, and keep the relationship.
  • the endgame for most: amazon (or etsy, or a marketplace) as the acquisition channel, your own store as the home where margins and relationships live. every package from the marketplace carries a card: love it? find us at our own address — and the email list grows from there.

the politics of it are simple when you see the shape: marketplaces rent you demand. owned channels let you keep it. businesses that only ever rent pay rent forever.

FAQ

can i start my own store with no audience at all?

you can, but expect the first months to be search-and-content work, not sales. if cash flow matters now, the marketplace first is not selling out; it is financing the owned channel.

is amazon worth the fees for cheap products?

rarely below a certain price point — the fixed fees eat percentage grows as the price falls. the math decides, not the ideology: run one product through the full fee stack before you commit the catalog.

what transfers between the two?

your brand, your photos, your copy, your reviews (you can quote them), your supplier relationships. the ranking, the buyers' accounts and the buy box do not travel.

the closing thought

the question behind the question is "where do my customers come from and where do they stay". amazon answers the first brilliantly and the second terribly. your own site is the reverse. the businesses that last use the first to feed the second — and know exactly which numbers in which margin table make that true.

if you are near Parma and want the owned half built properly: