'Should we focus on Shopify or Amazon?' is the wrong question. The useful one is narrower and more commercial: which of your products are still profitable on each channel once every cost is counted? Fees, fulfilment, advertising, returns and the cost of finding the customer in the first place all land differently on Amazon and on your own site. Until you've worked that through per product, any debate about channels is mostly instinct.
Plenty of brands end up on both. That works when each channel has a clear job and the numbers support it, and it doesn't when you're simply duplicating effort.
What Amazon gives you
Amazon has an enormous base of shoppers who already trust the platform. That's the pitch, and it's real. You don't have to convince someone to buy from a site they've never heard of. Amazon supplies an existing shopping audience, already searching for products like yours, though your offer still has to win visibility within it.
For a new brand with no reputation, this is a big deal. Someone who's never heard of you will buy because Amazon's returns process makes it low-risk. That borrowed trust is worth a lot when you're starting out.
FBA is very good logistics. Prime delivery, easy returns, customer service handled. You ship boxes to a warehouse and Amazon does the rest. Few small brands can match that on their own.
But Amazon takes a lot in return. There's a referral fee on every sale, a percentage of the price that varies by category. There are FBA fees, which depend on the size and weight of the product. There's advertising, because new products rarely get seen without it. And you don't get the customer relationship. Amazon's buyer messaging is for order-related contact, and you can't move those customers onto your own email list.
Amazon's a landlord. You're renting shelf space in their shop. That can be a great deal, but don't confuse it with building something you own.
What Shopify gives you
Your own store means you own the customer relationship. You get their email, with their permission. You can market to them again. You build a brand that exists independently of any platform.
You also control the experience. Bundles, subscriptions, customisation, upsells, the way products are presented. All things that are either impossible or constrained on Amazon.
The catch is obvious: you have to bring your own traffic. Your Shopify store has nobody walking past the shelf unless you send them there. That means paid acquisition (expensive, and rising), content (slow), influencers, or an existing audience. No traffic strategy, no sales.
And you carry the costs Amazon was quietly covering: picking, packing and delivery, returns, customer service, the platform subscription and the apps. Shopify is owning your own shop. You control everything, but you're responsible for everything too, including getting people through the door.
One product, both channels
The margin comparison only means something if you count the same things on both sides. Here's a hypothetical worked example. The figures are illustrative, not quotes; replace them with your own.
Say a product sells for £30 including VAT on both channels. VAT takes £5 either way, leaving £25. It costs you £8 to make and land.
On Amazon:
- Referral fee: say 15% of the £30 price, £4.50.
- FBA fulfilment fee: say £3.50.
- Advertising: say £3 per order.
- Returns allowance: say £1.
- Selling plan and tools, spread across orders: say £0.40.
That leaves £4.60 per order.
On your own Shopify store, for a customer you had to find with paid social:
- Payment processing: call it £0.60.
- Picking, packing and delivery: say £3.80.
- Advertising to win the order: say £7.
- Returns allowance: £1.
- Platform and apps, spread across orders: say £0.40.
That leaves £4.20 on a first order. On those assumptions, direct is no better than Amazon for a new customer.
The difference shows up on the next order. If that customer comes back through an email, the advertising cost might drop to £0.50, and the same order leaves £10.70.
That's also how you work out what direct acquisition you can afford. Before advertising, the first order leaves £11.20. Suppose one in three new customers orders once more within a year: that's worth about £3.57 per customer (£10.70 divided by three). So on these numbers each new direct customer is worth roughly £14.77 in contribution over a year, and that's the ceiling on what you can pay to acquire one. Pay that much and you break even after twelve months, with the cash spent up front. Both sides carry the same £0.40 share of platform fees; other overheads, such as your own time, are left out of both.
Run that calculation per product. Heavy, bulky or low-priced items often look very different from small, light, higher-priced ones, and the answer for one product won't hold for the range.
When to lean into each
Amazon makes sense when you're new and need credibility, when you're in a category where people shop Amazon first (electronics, books, commodity products), when you want to test whether a product has demand before building out a full store, or when you're willing to trade margin for volume.
Shopify makes sense when your brand is part of the value proposition, when you have an existing audience or can build one, when your products have the margins to absorb acquisition costs, or when you want to build something long-term rather than just transact.
Running both can work well with clear roles. Some brands put the core range on Amazon at a competitive price for people who search there, and keep the full range, bundles and subscriptions on their own site for people who come looking for the brand. Others keep Amazon to a few hero products and treat their own site as the main shop.
Don't plan on converting Amazon buyers into direct customers. You can't email them off Amazon, and inserts that push buyers to your own site can breach Amazon's rules. What you can do is make the product and the brand memorable, so that anyone who later searches for you by name finds your site.
The mistake is treating the two channels identically. Different platforms, different customers, different jobs.
You're renting shelf space in their shop.
When running both costs more than it earns
Two channels mean two of a lot of things, and for some businesses that's a net loss.
- Stock gets split. Stock sent into FBA sits in Amazon's network. Getting it back costs money and time, so you can end up out of stock on your own site with plenty sitting in an Amazon warehouse, or the other way round. Decide in advance which channel gets stock first when you're short, and keep a reserve you control.
- Returns behave differently. Amazon's returns process is generous to the buyer, and returned items aren't always resellable. Your returns allowance per channel should reflect what happens on that channel.
- Customer service doubles up. Two inboxes, two sets of rules, two sets of response-time expectations.
- Platform dependence grows. If Amazon becomes most of your revenue, a suspended listing or an account health problem can stop your business overnight.
- Price management gets harder. If your own site is cheaper, Amazon may stop showing your offer prominently.
If your range is small, your margins are thin, or your products are bulky, the extra operational cost of a second channel can outweigh the sales. Doing one channel well is a perfectly good strategy.
Will Amazon cannibalise your Shopify sales?
Everyone worries about this, and the honest answer is that it depends on your customers.
Some customers only buy on Amazon. They have Prime, they trust the platform, they're not interested in visiting individual brand websites. If you're not on Amazon, you don't get these customers.
Other customers actively avoid Amazon. They want to support independents, they like the brand experience, they want subscriptions or bundles that Amazon can't offer. These people aren't switching because you launched there.
The overlap is the people who would have bought from you directly but take the Amazon option when it's there. You can get an indication rather than guess. Launch some products on Amazon and hold others back, then compare how direct sales of the listed products move against the ones you didn't list, over the same weeks. Treat the result as indicative: it holds only if the two groups were selling alike beforehand, and customers may switch between them, buying a held-back product on your site in place of a listed one.
You can also limit it. Keep prices the same on both channels; don't undercut yourself on Amazon. Make some products or bundles exclusive to your own site. Offer loyalty benefits for direct customers.
Getting started on each
If you're Shopify-only and considering Amazon: don't launch your whole range. Start with five to ten bestsellers that pass the per-product calculation above. Let them prove the channel before you invest more.
Amazon listings require different skills from Shopify product pages. The photography needs to work as thumbnails. Titles are keyword-heavy in ways that would look bizarre on your own site. Bullet points follow specific patterns. If you're not willing to learn this or hire someone who knows it, your results will suffer.
Budget for advertising. New products on Amazon rarely rank organically without it, and the launch period is usually the most expensive. Set a ceiling per product based on the contribution it leaves, and a date by which it has to pay its way.
If you're Amazon-only and considering Shopify: don't half-arse it. A mediocre Shopify store won't compete with Amazon's trust and convenience. Invest in proper design, fast loading and a good mobile experience. Match the buying experience people get on Amazon, including delivery promises and returns.
And have a traffic strategy before you launch. 'Build it and they will come' doesn't work. How will people find you? Paid social? Search? Influencers? Content? Have an answer, and know from the calculation above what you can afford to pay per customer.
How we help: Amazon Consulting · Shopify Consulting
Want to talk this through for your own business? Dax reads every message personally and usually replies within a day.
Tell us what you needOr get the next article by email, every couple of weeks.
