WooCommerce, Shopify, and the AI Commerce Coup. The WordPress E-Commerce Autopsy.

There is a special kind of hubris that comes from watching a piece of blogging software absorb an entire industry. In 2011 a plugin called WooCommerce turned WordPress — a platform designed to publish diary entries — into an e-commerce engine capable of processing payments, managing inventory, calculating shipping, and sending tax invoices to the European Union. By 2015 Automattic had bought it for thirty million dollars in stock. By 2020 WooCommerce was powering roughly twenty-five percent of all online stores. The achievement is staggering. The architecture is indefensible.

What WooCommerce proved is not that WordPress is a good e-commerce platform. It proved that merchants will tolerate almost any technical experience if the software is free and the alternatives are expensive. For a decade the equation held: WooCommerce plus cheap hosting plus seventeen plugins equalled a functioning shop for the price of a monthly Netflix subscription. The merchants were happy. The developers were employed. The platform was, by the only metric that matters, successful. But success is not the same as fitness. And WooCommerce was never fit. It was just cheap.

The miracle. How a blog tool became a checkout.

WooCommerce began life as Jigoshop, a British e-commerce plugin that forked in 2011 when the founders disagreed about direction. The WooThemes team — who sold WordPress themes for a living — saw an opportunity to bundle commerce with design and create a vertically integrated ecosystem. They were right. The plugin shipped with a product post type, a cart, a checkout, and enough hooks to let third-party developers build an industry on top of it. Within two years the extension marketplace was generating millions in annual revenue. Within five it was the dominant force in open-source e-commerce.

The genius of WooCommerce was not technical. It was structural. By building on WordPress, WooCommerce inherited the world's largest plugin economy, the world's most familiar admin interface, and a global community of developers who already knew the database schema. A merchant did not need to learn a new platform. They installed a plugin, ran a wizard, and started uploading products. The barrier to entry was so low it was practically underground.

“WooCommerce did not win because it was good. It won because it was there.”

The numbers are worth stating. At its peak, WooCommerce was estimated to power over four million active stores. It was available in fifty languages. It integrated with every major payment processor on the planet. It had extensions for subscriptions, memberships, bookings, multi-vendor marketplaces, dropshipping, point-of-sale, and enough niche verticals to fill a trade show. The ecosystem was valued in the tens of billions. Automattic, which had started as a blog hosting company, was now primarily a commerce company with a blogging side hustle. The tail had wagged the dog so vigorously the dog needed a chiropractor.

The add-on tax. Death by a thousand subscriptions.

Here is the experience of building a serious WooCommerce store in 2024. You start with the free plugin. You discover that the free plugin does not handle subscriptions. You buy WooCommerce Subscriptions for two hundred dollars a year. You discover that subscriptions do not integrate with your membership logic. You buy WooCommerce Memberships for another two hundred. You need bookings for your service product. That is three hundred. You need a multi-currency switcher because you sell to Europe. That is a hundred and fifty. You need advanced shipping rules because flat rate does not cover your product dimensions. That is another hundred and twenty. You need a product configurator because your variants are too complex for the default interface. That is two hundred. You need tax automation because you sell in seventeen jurisdictions and you would rather not go to prison. That is four hundred.

Before you have launched, you have spent over a thousand dollars a year on plugins that all promise to 'seamlessly integrate' with WooCommerce and all require separate update cycles, separate licence keys, separate support portals, and separate compatibility headaches every time WooCommerce releases a major version. The free platform has become a fifteen-hundred-dollar annual subscription with a PHP fatal error habit.

• →Subscription billing: not included. Buy it.
• →Membership tiers: not included. Buy it.
• →Bookable appointments: not included. Buy it.
• →Multi-currency: not included. Buy it.
• →Advanced shipping: not included. Buy it.
• →Tax automation: not included. Buy it.
• →Product bundles: not included. Buy it.
• →Abandoned cart recovery: not included. Buy it.
• →Analytics that actually work: not included. Buy it.

This is not a complaint about commercial software. Commercial software should cost money. The complaint is about the architecture that makes every essential feature a separate commercial decision. Shopify includes subscriptions, memberships, multi-currency, shipping rules, tax automation, abandoned cart recovery, and analytics in its base subscription. You pay more upfront, but you do not discover, six weeks into a build, that your checkout cannot handle a feature your business model requires. WooCommerce's à-la-carte model turns project scoping into a guessing game and budget planning into a séance.

The performance lie. A shop that cannot shop.

WooCommerce is slow. Not inconveniently slow. Not 'add a caching plugin' slow. Structurally, foundationally, database-architecture slow. Every product is a post. Every product variation is a post. Every order is a post. Every order item is a post meta row. Your shop with five thousand products is not a database with five thousand rows. It is a database with fifty thousand rows, most of them connected by meta queries so expensive they should come with a carbon offset.

“Running WooCommerce on shared hosting is not e-commerce. It is performance art. The art is called 'Waiting for a Page Load.'”

The checkout flow is a masterclass in inefficiency. Add to cart: a database write. Update quantity: a database write. Apply coupon: a database query, a calculation, another write. Calculate shipping: a query to the products table, a query to the zones table, a query to the shipping classes, a query to the customer address. Proceed to payment: the entire cart serialised, passed through a filter chain, rendered into a form, validated, then sent to a payment gateway that waits for a response before the page can complete. On a quiet day this takes three seconds. On Black Friday it takes until the customer opens Amazon.

Agencies have built entire practices around WooCommerce performance optimisation. Object caching. Database indexing. Query rewriting. CDN configuration. Image lazy-loading. Cart fragment AJAX disabling. These are not optimisations. These are workarounds for an architecture that was never designed to handle commerce at scale. You are not tuning a race car. You are reinforcing a horse cart so it can survive the motorway.

Shopify. The platform that ate the bottom.

Shopify launched in 2006, the same year WordPress was still arguing about whether tagging should be in core. It was not built on a CMS. It was built as a shop. Every decision — the database schema, the checkout flow, the inventory model, the tax engine, the payment processing — was designed for selling things. It did not inherit a blogging platform's assumptions. It did not carry twenty years of backward compatibility. It was commerce-native from day one.

For a decade Shopify and WooCommerce occupied different markets. Shopify was for merchants who wanted simplicity and could afford thirty dollars a month. WooCommerce was for merchants who wanted control and could afford a developer. The boundary was porous — plenty of Shopify stores outgrew the platform and migrated to WooCommerce for customisation, and plenty of WooCommerce stores gave up on maintenance and migrated to Shopify for sanity. But the general trend, especially after 2020, was in one direction.

The pandemic accelerated everything. Merchants who had spent a decade tolerating WooCommerce's update roulette discovered that Shopify's hosted model meant they never thought about security patches, PHP versions, or plugin compatibility again. The migration pipeline became an industry. Agencies that had built their practice on WooCommerce customisation pivoted to Shopify setup and theme development. The Shopify Partner programme grew so large it became a career path in its own right. Meanwhile WooCommerce's market share, while still substantial in absolute terms, stopped growing. The line on the graph went flat. Then it bent.

• →Shopify handles hosting, security, scaling, and CDN. WooCommerce makes you buy them separately and blame yourself when they fail.
• →Shopify updates its own platform silently. WooCommerce sends you an email about a critical security patch and a list of twelve plugins that might break.
• →Shopify's checkout is optimised by a team of engineers whose only job is checkout optimisation. WooCommerce's checkout is optimised by whatever plugin you installed last Tuesday.
• →Shopify's support will look at your store. WooCommerce's support will ask you to deactivate all plugins and switch to a default theme, then blame your hosting.

The honest truth is that most WooCommerce stores should have been on Shopify all along. The only reason they were not was cost — Shopify's monthly fee felt like an expense, while WooCommerce's plugin stack felt like an investment. It is a powerful psychological trick. Paying two hundred dollars for a plugin feels like buying a tool. Paying thirty dollars a month forever feels like a subscription. But the plugin stack costs more, breaks more, and requires a developer on retainer. The subscription is cheaper, faster, and comes with someone else's engineering team.

The AI platforms. Commerce without the assembly.

And now the landscape is shifting again. Not from WooCommerce to Shopify — that shift is well underway and largely complete for new builds. The next shift is from Shopify to AI-native commerce. Platforms like Lovable are not just building websites. They are building the entire stack — storefront, checkout, inventory, payments, analytics — from a prompt, deployed to a global edge network, with no server to patch, no plugin to update, and no PHP version to worry about.

The implications for WooCommerce are catastrophic. The platform's last remaining advantage was control. 'You own your data,' the advocates cry, as if owning a MySQL dump of fifty thousand product variations is a retirement plan. 'You can customise anything,' they add, as if the ability to rewrite a shipping calculation hook is worth the three hundred hours it consumes annually. AI platforms do not offer control. They offer outcomes. The merchant describes the store they want. The platform builds it. The merchant sells things. The end.

“The future of e-commerce is not a dashboard. It is a conversation.”

Shopify, to its credit, saw this coming. Its 2025 product roadmap leaned heavily into AI — auto-generated product descriptions, AI-powered customer support, predictive inventory management, and a 'commerce assistant' that can rebuild a store's theme from a text prompt. The integration between Shopify and AI generation tools is now so tight that a merchant can describe their brand in a sentence and have a functioning storefront an hour later. WooCommerce cannot compete with this. WooCommerce requires a developer, a staging environment, a plugin compatibility audit, and a prayer.

For agencies, the pivot is obvious and painful. The WooCommerce specialist — once a prestigious, well-paid role — is becoming as niche as the osCommerce specialist before them. The new premium skill is not 'can you customise a checkout template?' It is 'can you describe a commerce experience in natural language and validate that the AI built it correctly?' The technical layer is being abstracted away. The value is moving up the stack, from implementation to strategy, from code to composition.

The honest takeaway. The cart is empty.

WooCommerce is not dying tomorrow. It powers millions of stores, generates billions in revenue, and has a community of developers who have spent a decade learning its internals. Platforms do not vanish. They fade. They become the thing you maintain because migrating is expensive, not because it is good. WordPress itself is fading in exactly this way — still enormous, still present, still forty percent of the web, but no longer the default choice for new projects.

If you are running a WooCommerce store today, the platform will continue to work. Updates will ship. Plugins will sell. The ecosystem will trundle on. But if you are planning a new e-commerce project in 2026, you should be asking hard questions. Why would you build on a platform that makes every feature a separate purchase? Why would you accept performance that requires a DevOps team to survive Black Friday? Why would you choose architectural control over commercial outcomes? The honest answer, for most merchants, is nostalgia. They know WordPress. They trust it. They have invested in it. And investment, as every economist notes, is the enemy of rational decision-making.

The future of e-commerce belongs to platforms that understand the merchant does not want to be a systems administrator. Shopify understood this a decade ago. AI platforms understand it now. WooCommerce — and WordPress beneath it — still thinks the merchant's problem is a lack of customisation options. The merchant's actual problem is that they want to sell things without thinking about database indexes. The platform that solves that problem wins. The platform that offers them a hundred plugins and a server to maintain loses. It really is that simple.

“WooCommerce turned a blog into a shop. The next generation will turn a sentence into a shop. The difference is not technology. It is the absence of everything WooCommerce made you carry.”

The cart is empty. The checkout is closed. The receipt is printed. Thank you for shopping with WordPress. Please come again, though honestly, nobody would blame you if you didn't.