There is no best ecommerce platform for a small business

No two "best platform" lists agree because they are answering a question with no general answer. A platform is not better or worse in the abstract; it is better or worse at absorbing your particular constraints — your staffing, your product rules, your marketing plan, your margin, your appetite for risk. A star rating hides every one of those. Many lists are also affiliate-funded, so the order reflects commission rates rather than fit, and the same three names rotate through first place depending on who is paying. The useful exercise is subtraction rather than selection. Nobody available to maintain a server disqualifies self-hosting. Wholesale tiers and configurable products disqualify the simplest hosted plans. Thin margins at high volume disqualify percentage-based fees. By the end you are not choosing a winner; you are noticing what is left.

This decides most cases, and it is the question owners answer aspirationally. Be literal: name the person who will apply updates, renew certificates, restore a backup at ten on a Friday night, and read the error log when payments start failing. If that name is "nobody" or "me, eventually", a hosted platform is doing real work on your behalf and paying for it is rational. Self-hosted WordPress with WooCommerce is the opposite trade: more control, more surface to look after. Every plugin is a dependency with its own release schedule and its own way of breaking, and the failure mode is gradual rather than dramatic — a store that quietly accretes twenty extensions until nobody can say what half of them do. Plugin bloat is the predictable end state of an install that has no owner.

• Name the actual person, not a role or an intention
• Decide who gets called when checkout fails on a Sunday
• Budget maintenance as a recurring cost, not a one-off
• Count the plugins nobody can currently justify

Ordinary rules — one price, one currency, flat or weight-based shipping, a coupon at checkout — are handled well everywhere, and this question will not constrain you. Unusual rules are where platforms diverge sharply. Subscriptions with pauses and pro-rated changes. Wholesale tiers where the price depends on who is logged in. Configurable products priced by dimensions or material. Delivery windows tied to a van route or a local zone. Deposits, quotes, minimum order quantities per customer group. On hosted platforms these usually live in third-party apps, and each app is a monthly fee plus a compatibility risk at every platform update. On WooCommerce they live in extensions or in code you commission: more up front, less per month, and shaped to how you actually sell. Write your three strangest rules down before you evaluate anything, because they are the real specification.

If your acquisition plan is paid ads and a marketplace presence, publishing barely matters and any platform will hold a handful of pages. If the plan is to earn search traffic — buying guides, comparisons, sizing explainers, service pages for the towns you deliver to — then you are running a publication attached to a shop, and editorial tooling stops being a detail. Ask how each option handles many URLs: templated pages, internal linking, redirects when things move, structured data, translations if you sell across borders. WordPress grew up as a publishing system and this is where it is genuinely strong; hosted platforms have improved but still assume the catalogue is the site. The real question is not which wins at SEO in the abstract, but how many pages you will actually write and who will write them. Ecommerce SEO work is a staffing decision before it is a platform one.

Platform cost comes in two shapes, and they invert with scale. Fixed cost — hosting, licences, a maintenance retainer — hurts most when you are small and becomes trivial as orders grow. Variable cost, a percentage of every sale, is painless at low volume and becomes the largest line on the page precisely when things go well. Marketplaces sit at the far end of that spectrum: Amazon's published referral fee schedule puts most categories between 8% and 15% of the item price, before fulfilment costs. Do the arithmetic on your own numbers rather than the industry's. Take last month's order count, your average order value and your gross margin, then model each option at today's volume and at three times today's volume. An option that stays healthy at both ends is safe. One that only works while you stay small is a decision you will be making again.

• Model fees at today's volume and at three times it
• Separate card processing from platform percentage fees
• Count app subscriptions as part of platform cost
• Check the fee against your worst-margin category

Ownership sounds ideological until something forces the issue — a policy change, a pricing change, an account suspension, a customer list you cannot export in usable form, a checkout you are not permitted to modify. Hosted platforms trade control for reliability, and for many small businesses that is a fair trade honestly made. But some businesses have a specific reason to hold the checkout and the customer data themselves: regulated categories, unusual payment methods, purchase flows that need a step nobody's standard checkout offers, or a first-party data strategy that cannot depend on somebody else's roadmap. Be concrete about which of those applies to you rather than answering on principle. And if the answer turns out to be "the checkout is fine, the storefront is the problem", that is a different project — a hybrid arrangement leaves the payment flow where it already works and rebuilds only what the customer sees.

For a small US business the field is narrower than the lists suggest. A hosted platform gives you a store that stays up without you, a large app ecosystem and someone to call; you pay in monthly fees, per-order percentages, limits on what the checkout may do, and the standing risk that an app you depend on is deprecated. WordPress with WooCommerce gives you control, strong publishing and no per-order tax; you pay in maintenance and in the requirement that somebody owns it — that is its genuine downside and it is not a small one. A marketplace-only presence gives you demand on day one and no site to run; you pay in margin, in a customer relationship you do not own, and in a business your landlord can reprice. Weigh the source too: this site is run by people who build on WooCommerce, which is exactly the bias to discount for.

• Hosted: fastest to launch, hardest to bend
• WooCommerce: most control, needs a named owner
• Marketplace: instant demand, borrowed customers
• Any of them beats a store you never finish

Having spent five questions on the choice, the honest conclusion is that the platform is usually not the constraint. Stores fail because too few people were looking for what they sell, because the assortment is too thin to justify a visit, because the price is beaten by a marketplace that delivers tomorrow, or because there was never a plan to bring anyone to the site at all. Those problems follow you across every migration, and switching platforms is an expensive way to postpone facing them. So the cheapest good advice is often to stay where you are and spend the money on demand and assortment instead. Move when the platform is measurably in the way: a rule you need cannot be expressed, fees have outgrown the alternative, or the site is slow and cannot be fixed where it stands. What this kind of work costs is published so you can judge that before talking to anyone.