Premium SaaS-style workspace showing WooCommerce analytics dashboard with failed AI checkout metrics and ecommerce performance reporting.

Meanwhile Shopify is running the opposite playbook. Their agentic commerce marketing has reached a fever pitch. “Millions of merchants can sell in AI chats.” “One click and your products are everywhere AI conversations happen.” The implicit message to everyone on WooCommerce: you’re being left behind, and you need to come to Shopify to fix it.

I want to talk about what’s actually true here, because the gap between the pitch and the reality is wider than most people realise.

The TL;DR:

  • OpenAI’s flagship in-chat buy button converted below 1% — against an industry average of 3-4%. Of millions of eligible Shopify merchants, about a dozen went live.
  • Adobe and BrightEdge both measure current LLM referral traffic at under 1% of total ecommerce sessions. Real growth, but from almost nothing.
  • WooCommerce is already a launch partner for Stripe’s Agentic Commerce Protocol. Native MCP support shipped in WooCommerce 10.3. The capability is here.
  • Shopify’s advantage is packaging, not infrastructure. Their version is turnkey; WooCommerce requires a plugin and an hour of configuration.
  • Replatforming to chase agentic commerce solves a problem you don’t have with a solution that doesn’t fully work yet.

Three small things this quarter — GA4 setup, llms.txt, product schema — will position you for whatever shakes out, without spending six figures on a migration to chase a pitch.

What just happened with OpenAI’s Instant Checkout

Of the millions of Shopify merchants who were theoretically eligible, about a dozen actually went live. And the infrastructure was half-built: no sales tax collection, no fraud prevention, no real-time inventory sync, single-item carts only.

The reason the flagship agentic commerce feature of the world’s most-used chatbot didn’t work isn’t some deep philosophical problem with AI. It’s that the thing was shipped before it was ready, and users could tell. OpenAI may relaunch a v2 of Instant Checkout — they probably will. But the pattern is the one that matters: agentic commerce pitches keep outrunning agentic commerce reality. Anyone replatforming today to catch the wave is replatforming for v1 of a v3 problem.

What Shopify is actually selling — and what they leave out

Modern ecommerce analytics dashboard comparing AI shopping hype with real WooCommerce performance metrics.

Credit where it’s due. Shopify Catalog is a real piece of engineering. The admin toggle genuinely does push your products into ChatGPT, Microsoft Copilot, and Google AI Mode. If you run a Shopify store in the US, it works. I’m not going to pretend otherwise.

But the pitch glosses over several things that would matter to anyone making a platform decision.

  • Perplexity requires a separate application and routes checkout through PayPal, not Shopify.
  • Claude isn’t integrated at all.
  • In January, Shopify auto-enrolled US merchants into agentic shopping without explicit consent. If a WooCommerce plugin did that we’d call it a dark pattern.

So when you see “millions of merchants can sell in AI chats,” what that actually means is: some merchants, in some chats, on some surfaces, in one country, with one payment processor, sometimes by default whether they wanted it or not.

What the data actually shows about AI referral traffic

WooCommerce dashboard showing AI commerce integrations, structured schema validation, and ecommerce infrastructure metrics.

A brief caveat, because I’d be misleading you if I didn’t mention it: GA4 is known to undercount LLM traffic by a factor of three to five. Users research in ChatGPT, then type your brand into Google, then buy. The sale gets attributed to organic search. So the real influence of LLMs is larger than the referrer data suggests. But even generous adjustments don’t get you anywhere near “this is a material channel today.”

Get the real numbers for your store

Have us audit your AI-readiness directly.

You can spend weeks figuring out which LLMs are sending you traffic, which products are getting picked up, and whether your schema is firing. Or we can tell you in 7 days — with the exact gaps to fix.

  • GA4 + LLM referral audit
  • Product schema verification
  • No contracts, ever
Get My Free Performance Analysis →

No contracts. See real data from your actual store before any commitment. We’ll show you exactly where your AI visibility is — and what it would take to improve it.

What Shopify reps won’t tell you about WooCommerce and AI commerce

Dark-mode ecommerce operations dashboard showing GA4 AI referral tracking, llms.txt setup, and product schema optimization.

WooCommerce is on the list.

This isn’t theoretical work on some future roadmap. It’s shipped. It’s just not pre-installed, because WooCommerce has never been a walled garden and never will be.

Three things to do this quarter instead of replatforming

Not a replatform. Three small things that cost you an hour or two and position you for whatever shakes out.

1. Set up a custom channel group in GA4

So you can see your AI referral traffic in standard reports going forward. Admin → Data Display → Channel Groups → create a new channel above “Referral” that matches sources like:

chatgpt.com|openai.com|perplexity.ai|gemini.google.com|copilot.microsoft.com|claude.ai

From that point on you’ll have a baseline you can actually track, and you won’t have to go fishing through acquisition reports every time someone asks about AI traffic.

2. Install an llms.txt plugin

3. Audit your product schema

This is the one that actually matters. ChatGPT and Perplexity don’t maintain their own product databases — they pull from Google Shopping and from structured data on your product pages.

A clean Google Shopping feed and complete schema.org Product markup (with GTIN, brand, price, availability, and aggregate rating populated) is the actual currency of AI product discovery. If you’re already running Rank Math or Yoast WooCommerce SEO, most of the plumbing is there; you just need to verify it’s firing correctly on every product page. If you’re not, now’s the time.

That’s the whole job for this quarter. Set up the measurement, publish the llms.txt, fix the schema. When the protocol wars settle and one of the hyperscaler plays actually starts sending meaningful traffic, you’ll be ready. And you won’t have spent six figures replatforming to chase a pitch that OpenAI just publicly admitted didn’t work.

Skip the DIY route

Want us to set up all three for you?

GA4 channel group, llms.txt configuration, full product schema audit and fix. 7 days. Done properly the first time. You get the measurement baseline and the AI visibility positioning without spending a weekend on it yourself.

Get My Free Performance Analysis →

No contracts. We’ll audit what you have, fix what’s broken, and you’ll have a working baseline by week two.

Frequently asked questions

Can WooCommerce stores sell in ChatGPT?

Yes — through structured data and product schema rather than a native checkout integration. ChatGPT pulls product information from Google Shopping feeds and schema.org Product markup on your pages. Complete schema and a clean feed put your products into ChatGPT’s discovery surface today. A native one-click checkout flow comparable to Shopify Catalog requires the Stripe Agentic Commerce Protocol, which has free plugins available in the WordPress directory.

Is Shopify really ahead of WooCommerce on AI commerce?

On packaging, yes. On capability, less than the marketing suggests. Shopify Catalog is a single admin toggle. WooCommerce requires installing the Stripe ACP plugin and configuring it. The underlying protocol support, MCP integration in WooCommerce 10.3, and Stripe partnership are all live. The difference is roughly an hour of configuration work, not a platform gap.

How much traffic is actually coming from LLMs today?

Under 1% of total ecommerce sessions, according to Adobe and BrightEdge. Growth rates are steep — Adobe measured 693% year-on-year over the holidays — but the base is still very small. GA4 also undercounts LLM traffic by a factor of three to five because users research in ChatGPT and then buy through branded search, which attributes to organic.

What is the Agentic Commerce Protocol?

An open standard launched by Stripe in December 2025 for letting AI agents complete purchases on behalf of users. WooCommerce, Wix, BigCommerce, Squarespace, and commercetools were named launch partners. It’s the infrastructure layer underneath features like ChatGPT’s Instant Checkout — but designed to work across any platform, not lock merchants to one.

Should I replatform to Shopify to get AI commerce features?

What is llms.txt and do I need one?

llms.txt is a file at the root of your domain that tells AI crawlers what your site is about, how to describe it, and what content to prioritize. It’s the AI-era equivalent of robots.txt for search engines. A WordPress plugin generates it automatically based on your site content. Five minutes of work; 95% of stores haven’t done it.

Why did OpenAI’s Instant Checkout fail?

Multiple reasons. Conversion was below 1% versus a 3-4% industry average. Walmart’s EVP told Wired that ChatGPT conversions were a third of their own site. Only about a dozen Shopify merchants out of millions of eligible ones went live. The feature was missing sales tax collection, fraud prevention, real-time inventory sync, and multi-item carts. It was shipped before it was ready, and shoppers could tell.

Will WooCommerce get the same in-chat checkout features Shopify has?

The capability is already live via Stripe’s Agentic Commerce Protocol. The difference is packaging — Shopify offers it as a single toggle, WooCommerce offers it via free plugins. Native one-click ACP integration is on WooCommerce’s public roadmap and partial functionality has shipped in WooCommerce 10.3.

What’s the difference between MCP and ACP?

MCP (Model Context Protocol) is how AI assistants like Claude and ChatGPT access information about your store and products. WooCommerce 10.3 shipped native MCP support. ACP (Agentic Commerce Protocol) is how AI assistants actually complete a purchase on the user’s behalf. WooCommerce supports ACP through Stripe’s plugin. You need both for a full agentic commerce experience.

How do I track AI traffic in GA4?

Create a custom channel group in GA4 Admin → Data Display → Channel Groups. Add a new channel above “Referral” that matches sources like chatgpt.com, openai.com, perplexity.ai, gemini.google.com, copilot.microsoft.com, and claude.ai. Going forward, all AI referral traffic appears in standard acquisition reports instead of being lost in “Other Referral.”

The AI commerce pitch is louder than the AI commerce reality. Position for both.

The agentic commerce future will arrive. It’s just not arriving on the timeline the marketing suggests, and it’s not arriving as a Shopify-exclusive event. WooCommerce is already in the room. Stripe partnered with us at the same tier they partnered with Wix and BigCommerce. The protocol work has shipped. The plugins exist. The capability is real.

What’s also real: spending six figures to replatform today to capture a channel that’s running at under 1% of sessions is the most expensive way to solve a problem that doesn’t yet need solving.

Set up the GA4 tracking. Publish the llms.txt. Audit your schema. Reassess in twelve months when the actual numbers are visible. That’s the playbook.

Get my free AI-readiness audit

We’ll audit your GA4, llms.txt status, and product schema and send specific findings within 7 days — what’s working, what’s missing, and what to fix first.

Get My Free Performance Analysis →

No contracts. See real data from your actual store before any commitment.

CB

Talk directly with Campbell

Founder · 10 years WooCommerce

Book my strategy call

30 minutes with Campbell. We’ll review your store’s AI visibility and give you the honest read on what matters this quarter — whether or not you work with us.

Book My Strategy Call →

30 minutes. No pitch — just an honest assessment.