Ecommerce Web Design · Guide

Which Platform Is Best for Ecommerce Web Design?

There is no best platform. There is a fit. It is decided by catalogue size, order volume, who maintains the shop day to day and what has to connect to it. Almost everything written on this subject earns a commission on the answer, which is worth knowing before you read any of it.

Updated: July 2026
Written by: Andrew Odgers, Managing Director
Reading time: 12 minutes
The question is the problem

Why The Question Has No Single Answer

Asking which platform is best assumes there is one, in the way that asking which vehicle is best assumes it. The answer depends entirely on what you are carrying, how far, how often and who is driving.

That is not evasion. The proof is that the same platform genuinely is the right answer for one business and the wrong one for its neighbour. A shop selling forty handmade items and a distributor selling nine thousand parts with trade pricing are both retailers. Almost nothing about the correct answer transfers between them.

A useful answer therefore looks different from a recommendation. It looks like a short list of things about your business that determine the outcome, followed by an assessment of which platforms sit comfortably with those answers. Once the inputs are settled the choice is usually obvious. Frequently there are two acceptable answers rather than one.

One thing to hold in mind while reading anything on this subject, including comparison articles that appear independent. Most platform comparison content earns a commission when a reader signs up. That is a legitimate business model rather than a scandal. It does mean the recommendation and the writer's income point in the same direction, which is a reason to check any conclusion against what happens to a shop two years later.

The inputs, before any names

The Five Things That Actually Decide It

Answer these five and the shortlist writes itself. Answer none of them and you are choosing on aesthetics.

Catalogue size and complexity. Not only how many products. How many variants, whether products relate to each other, whether different customers see different prices and how often any of it changes. Complexity matters more than raw count.

Order volume. Because transaction fees scale with success while subscriptions do not. A charge that is trivial at ten orders a week is a salary at a thousand.

Who maintains it day to day. The decisive question for most small businesses and the one asked least. Somebody has to apply updates, watch security, fix what breaks and add products. Whether that is you, an employee or an agency changes the correct answer entirely.

What has to connect to it. Stock control, accounting, shipping, marketplaces, an ERP. Existing systems constrain the choice more than anything on a feature list, because an integration that does not exist has to be built.

Budget across three years. Not the launch figure. Subscription, transaction fees, apps, hosting and maintenance, added up over the period the shop will actually run before it needs serious attention.

Notice that only one of those five is about the platform. The other four are about you, which is why a recommendation made without asking them is a guess wearing a suit.

The split underneath everything

Hosted Against Self Managed

Before any individual platform, there is one structural decision that determines most of what follows. Are you renting a managed service or installing software you become responsible for?

Hosted platforms. The company runs the servers, applies updates, handles security and keeps the thing available. You pay a subscription and frequently a share of transactions. What you gain is that a large category of problems is no longer yours. What you give up is control at the edges, since you can only change what the platform allows. You are also exposed to decisions the platform makes about its own direction.

Self managed software. You install it, host it and own it. Anything can be changed because you have access to everything. What you gain is control and the absence of transaction fees. What you take on is updates, security patching, backups, hosting, performance and somebody to do all of it.

The framing that makes this decision simple is not technical. You are choosing who carries the risk when something goes wrong. Hosted means renting safety and accepting limits. Self managed means owning the problem and keeping the freedom.

Which is why the maintenance question in block two decides so much. A business with nobody to look after a self managed installation has not chosen freedom. It has chosen an unpatched shop, which is a different thing and considerably more expensive.

What each suits and where each strains

The Main Platforms

Written from what tends to break later rather than from feature lists, since a reader will check us against each platform's own account of itself.

Shopify

Hosted. Suits most small and medium retailers, particularly those wanting to run the shop themselves without thinking about servers. It strains where a business has genuinely unusual requirements the platform does not anticipate. It strains again where deep content publishing sits alongside selling. Costs are predictable: subscription, transaction handling and apps. Suited to an owner running it day to day, with an agency for structure.

WooCommerce

Self managed software running on WordPress. Suits content led businesses and anybody needing control over structure and output. It strains when nobody has been made responsible for it, since an unmaintained installation degrades rather than sitting still. Costs move from subscription into hosting, maintenance and developer time. Needs a named person or supplier looking after it.

Wix and Squarespace

Website builders that also sell. They suit small catalogues, simple products and shops that are secondary to the main business, particularly where the owner wants to manage everything personally. They strain at catalogue complexity, variant handling and multi channel selling. Costs are low and predictable. Run by the owner, which is the point of them.

BigCommerce

Hosted, aimed at larger catalogues, with more capability built in rather than added through apps. It suits businesses wanting hosted convenience without assembling functionality from extensions. The ecosystem around it is smaller, which matters when you need something specific.

Adobe Commerce and comparable enterprise platforms

Extremely capable and built for large, complicated operations. They require a development team and a budget to match. For most small and medium businesses this is the wrong answer regardless of how impressive the demonstration was.

Permission to stop worrying

What Changes At Scale

A great deal of platform comparison is written for businesses considerably larger than the ones reading it. Here is what only matters above a certain size, so a smaller shop can stop weighing it.

Transaction fees against subscription. At modest order volumes the difference is small. At high volumes it becomes one of the larger numbers in the business. The calculation then reverses.

Bulk catalogue management. Editing four hundred prices at once, handling supplier feeds, managing product data at volume. Irrelevant with sixty products and central with six thousand.

Control over filtered pages. How much say you have over what happens when shoppers narrow a list. This becomes serious once a catalogue is large enough to generate very many combinations. It is covered in how web design affects ecommerce SEO.

Multiple warehouses, currencies and tax regions. A different order of complexity, arriving suddenly when a business starts selling abroad.

Performance under concurrent load. Whether the shop holds up when a great many people arrive simultaneously, which matters if you run promotions or appear in the press.

If none of those describes your business, you have more acceptable options than the comparison articles suggest. The maintenance question from block two should then decide it.

Five bad reasons

What Not To Choose On

Each of these decides more platform choices than the five inputs above, which is why they are worth naming.

A friend's recommendation. They are reporting on a business with a different catalogue, different volumes, different systems and a different person maintaining it. What worked for them is evidence about them.

Your designer's preference. Frequently reasonable and rarely disinterested. A supplier proposes what they build fastest, which is a legitimate efficiency and is not the same as what suits you. Ask which platforms they have built on and why they are proposing this one for your situation specifically.

A template that looked nice. The design is the most reversible part of the whole decision and the platform is among the least. Choosing the second because of the first is exactly the wrong way round.

A comparison article. Covered in block one. Check who benefits from the conclusion.

What a large retailer uses. Their platform was chosen to solve problems of scale you do not have, by a team you do not employ, with a budget that is not comparable. Copying the choice imports the cost without the reason.

What all five share is that they answer a question about somebody else. The five inputs in block two answer questions about you, which is the only place a correct answer can come from.

Why the decision matters

Changing Platform Later

Platform choice is reversible in the sense that anything is reversible. It is worth knowing what reversing it involves before treating it lightly.

In money. A migration is a full rebuild plus data work. You are paying for the new shop and for moving everything into it, which frequently costs more than the original build did.

In time. Longer than a new build, for the reasons in how long it takes to build an ecommerce website. Mapping, redirects and running two systems in parallel all sit on top of the ordinary work.

In search visibility. The genuine risk. An established shop has visibility built over years, every address is changing and every one of them needs a destination. Done carefully the disruption is temporary. Done carelessly it is not. The recovery takes longer than anybody budgets for.

In things quietly lost. Reviews, customer accounts, order history and stored addresses. Each has to be deliberately carried across. Each is routinely forgotten until somebody asks where it went.

None of that means never move. Businesses outgrow platforms and moving is the right call when they do. It means the decision is worth an afternoon of proper thought at the start, since the alternative is paying for it twice. Migrations are their own discipline and sit in our site migrations guides.

Taken head to head

The Direct Comparisons

Three pairings account for most of the decisions we are asked about. Each is worked through properly rather than as a feature list.

Shopify against WooCommerce is convenience against control. It turns on who is going to maintain the shop.

Shopify against Wix is a comparison between things built for different jobs, where the useful answer is when a site builder is genuinely enough and how to recognise the point at which it stops being.

WordPress against Shopify is really a question about whether you are a shop that has a website or a website that has a shop, which answers itself once put that way.

Platform specific build detail sits in the dedicated guides for each platform rather than here, since this page is about making the decision rather than about working within it.

Ecommerce web design

We ask the five
before recommending.

Catalogue, volume, who maintains it, what it connects to and the three year budget. We build on more than one platform, so the recommendation is not decided by what we happen to be fastest at.

What a build covers:

Platform assessment Structure and category planning Design and build Product data preparation Basket, checkout and payments Integrations Testing and launch

Ongoing management afterwards sits inside our fixed monthly fee. Build work is quoted per project.

The full guide series

Twenty guides.
One subject.

This page frames the platform decision. The rest of the series covers the direct comparisons, what a build costs, where conversion is lost and why the search ceiling is set during the build.

Questions people ask

Choosing an Ecommerce Platform

Which ecommerce platform is best?
There is no single best one, in the way there is no best vehicle. The same platform is genuinely right for one business and wrong for its neighbour, because a shop selling forty handmade items and a distributor selling nine thousand parts with trade pricing share almost nothing that matters to the decision. A useful answer is a short list of facts about your own business, followed by which platforms sit comfortably with those answers. Frequently two options are acceptable rather than one.
What should decide my platform choice?
Five things. Catalogue size and complexity, meaning variants, relationships and whether different customers see different prices. Order volume, because transaction fees scale with success while subscriptions do not. Who maintains the shop day to day, which is the decisive question for most small businesses and the one asked least. What has to connect to it. And budget across three years rather than at launch. Only one of those five is about the platform. The other four are about you.
What is the difference between hosted and self managed platforms?
With a hosted platform the company runs the servers, applies updates and handles security, so a whole category of problems is not yours. You accept limits on what can be changed and usually pay a subscription plus a share of transactions. With self managed software you install and own it, so anything can be changed. You take on updates, patching, backups, hosting and performance. You are choosing who carries the risk when something goes wrong.
Which platform decisions only matter for larger shops?
Several. A smaller shop can stop weighing them. Transaction fees against subscription, where the calculation reverses at high volumes. Bulk catalogue management, irrelevant at sixty products and central at six thousand. Control over filtered pages, which becomes serious once a catalogue generates very many combinations. Multiple warehouses, currencies and tax regions. And performance when a great many people arrive at once. If none of those applies, you have more acceptable options than comparison articles suggest.
Why should I be careful with platform comparison articles?
Because most of them earn a commission when a reader signs up. That is a legitimate business model rather than a scandal. It does mean the recommendation and the writer's income point in the same direction. The practical response is not to dismiss them but to check any conclusion against what happens to a shop two years later, particularly around maintenance, running costs and what changing platform would involve.
How difficult is it to change platform later?
A migration is a full rebuild plus data work, so it frequently costs more than the original build. It takes longer than a new build because mapping, redirects and running two systems in parallel sit on top of the ordinary work. The genuine risk is search visibility, since every address changes and each needs a destination. Reviews, customer accounts and order history are also routinely lost. None of that means never move, only that the initial decision deserves proper thought.