Should you sell on Amazon Australia or your own website?

This question usually arrives in one of two forms. Either a store owner with a working website is watching competitors appear on Amazon and wondering whether they are missing out, or someone with a good product and no audience is deciding where to launch at all. They look like the same question and they are not, because the right answer depends almost entirely on which problem you currently have: a demand problem or a margin problem.

It helps to be precise about what each channel actually is. Amazon is a demand marketplace. You are renting access to a large pool of shoppers with existing purchase intent, and paying for that access per sale. Your own website is an owned channel. Nothing sits between you and the customer, but nothing brings the customer to you either. Everything else in this decision, fees, data, control, risk, follows from that single structural difference.

What is the real difference between Amazon and your own website?

The most common mistake is to compare the two on fees alone. Fees are the visible cost of the marketplace, but the more consequential differences are structural and show up months later.

Factor	Amazon Australia	Your own website
Where demand comes from	Built in — shoppers are already searching	You generate it via ads, SEO, email, social
Cost per sale	Referral fee per order, plus fulfilment fees if using FBA	Platform and payment fees, plus your acquisition cost
Customer data	Restricted — Amazon owns the relationship	Yours, including email and purchase history
Brand control	Constrained by listing format and competing ads	Complete
Pricing control	Constrained by buy-box dynamics and competitors	Yours to set
Cash flow	Settlement on the platform's cycle, roughly fortnightly	Payouts typically within a few business days
Concentration risk	High — a suspension stops revenue	Low — you control the store
Setup effort	Lower — listings, not a build	Higher — build, content, tracking, ads

Read that table as a single sentence: Amazon converts money into demand, and your website converts effort into an asset. Neither is a shortcut. The marketplace shortcut is real but it is rented, and the rent is charged in the currency you can least afford to lose if you are building a brand: margin and customer contact.

Which channel actually makes more money per order?

Work it through with your own numbers rather than a rule of thumb. Take a hypothetical $100 product with $35 of landed cost. On Amazon you deduct a referral fee, which varies by category and commonly sits somewhere between the high single digits and the mid teens as a percentage of the sale price, plus fulfilment and storage fees if you use Fulfilment by Amazon. Check the current Amazon Australia rate card for your category before you model anything, because fees change and category boundaries are not always intuitive.

On your own website you deduct payment processing, your platform subscription, shipping if you absorb it, and then the number people conveniently forget: customer acquisition cost. That last figure is what makes direct-to-consumer margin look better on a spreadsheet than it is in a bank account. If your blended acquisition cost is $30, your $100 direct sale is not meaningfully more profitable than the marketplace sale, it just fails in a different place. If your acquisition cost is $8 because you rank organically and your email list does the heavy lifting, direct wins by a wide margin.

Two things follow. First, the honest comparison is not fees versus fees, it is marketplace fee versus your true acquisition cost. Second, the direct channel only compounds if customers come back, because the acquisition cost is paid once and every later order carries full margin. That makes customer lifetime value the deciding number rather than first-order profit. A product bought once every five years struggles to justify the cost of building an audience. A consumable with a genuine reorder cycle justifies it easily.

Who owns the customer, and why does it matter?

On Amazon, the buyer is Amazon's customer. You will see what sold, but the platform restricts the contact details you receive and limits what marketing you can do off-platform. You generally cannot build an email list from marketplace orders, run a win-back flow, survey buyers about a new range, or tell them about a product launch.

For a business selling a commodity where the customer will never think about the brand again, that costs little. For a business trying to build something durable, it is the whole game. Every one of the compounding levers covered in the five levers of ecommerce growth — retention, average order value, repeat frequency — depends on being able to reach the customer a second time. Marketplace revenue is real revenue, but it does not build the asset that makes next year cheaper.

What are the risks people underestimate on each side?

On the marketplace side, the underestimated risk is concentration. A listing suppression, a policy change, a category fee revision or an account review can stop revenue on someone else's timeline. Businesses that reach a large share of revenue through a single marketplace have built a genuinely fragile structure, however healthy the profit and loss looks. There is also the competitive dynamic: a marketplace makes your product directly comparable with everything near it, which pushes towards price competition and away from the brand story that justifies a premium.

On the direct side, the underestimated risk is the cost of demand. Australian ad auctions are dense, the domestic audience is small, and a good-looking site with no traffic strategy is an expensive brochure. Plenty of stores launch on their own domain, spend months on design, and discover the hard part was never the website. If you have not yet chosen a platform, the practical trade-offs are covered in Shopify versus WooCommerce — but the platform is the smaller decision.

Should a new Australian store start on Amazon or its own website?

Use the problem you actually have as the tiebreaker.

• Start on the marketplace if you are validating whether the product sells at all, if your category has obvious existing search demand, if your margins can absorb the fee, and if you have no audience and no budget to build one yet.
• Start with your own site if the product needs explanation or a brand story to justify its price, if you have an audience or organic strength already, if the item is consumable or subscription-like, or if your margin cannot survive a marketplace fee.
• Run both once you have proven demand and can operate two channels without breaking inventory accuracy.

If you do start on Amazon, treat it as validation with a deadline rather than a permanent strategy. The failure pattern is predictable: strong marketplace sales for two years, no owned audience, thin margin, and no leverage when the platform changes terms.

How do you run both channels without cannibalising yourself?

Running both is normal, but it introduces operational problems that quietly cost money if unmanaged. Four rules cover most of it.

1. One source of truth for inventory. Two channels drawing on one pool of stock will oversell without a system syncing them, and marketplace performance metrics punish cancellations hard.
2. Consistent pricing. Undercutting your own marketplace listing damages your standing there and trains customers to wait for a discount.
3. A real reason to buy direct. Not a coupon — a broader range, genuine bundles, personalisation, a better warranty or a better post-purchase experience. Differentiate on value, not price.
4. Separate unit economics. Track contribution margin per channel, not blended. Blending hides a marketplace line that is quietly losing money on every unit.

The fulfilment question sits underneath all of this. Using the marketplace's own fulfilment network buys conversion and delivery speed but puts your stock in someone else's warehouse and complicates serving your own site from the same pool. The alternative is a third-party logistics provider serving both channels, which trades some marketplace conversion for control and simpler inventory. The trade-offs are worth working through properly before committing stock, and are covered in how to choose a fulfilment partner. Getting channel structure and fulfilment right early is exactly the kind of foundational decision that Alpha Vault's e-commerce work is built around.

How to decide

Strip it back to three questions. Do you have a demand problem or a margin problem? Does your product earn repeat purchases, or is it a one-off? Could your business survive losing its largest channel for a month? The first tells you which channel solves your current constraint. The second tells you whether investing in an owned audience will ever pay back. The third tells you how much concentration risk you can responsibly carry.

For most Australian businesses building something they intend to keep, the sequence that works is to use the marketplace for what it is genuinely good at, discovery and validation, while deliberately building the owned channel that captures the margin and the relationship. The marketplace pays this quarter's bills. The website is what you actually own. If you want help modelling the numbers for your specific product and category before committing stock to either, book a consultation and we will work through it together.