By Steve Merrill, Founder of WRKNG Digital — September 27, 2026
An AI agent can already find your product, read your price, and decide whether to recommend it. Soon it will click buy on behalf of a shopper. If your Shopify store isn't readable to that agent, you don't exist in that transaction.
I've had a hard time watching this one land. Because I've seen this movie before.
What is agentic commerce, in plain terms?
Agentic commerce is when an AI agent browses, compares, and buys products for a person instead of the person doing it by hand. The agent reads structured product data, checks price and availability, and completes checkout through a supported path. Shopify has been shipping the plumbing for this since 2024, and OpenAI's Instant Checkout work with the Agentic Commerce Protocol pushed it further.
Here's the thing. The shopper never sees your homepage. The agent does the shopping. So the question stops being "does my site convert" and becomes "can an agent read and trust my data."
Why does the window matter?
I ran an $10M/year clothing brand for 15 years. In 2013 Facebook changed its algorithm and my organic sales went from $50K a month to $2K almost overnight. I refused to run ads for two years. Competitors who adapted early hit $80M while I clawed back to $10M. Same product. Same market. Two-year head start. I never caught up.
Agentic commerce has the same shape. The stores that get machine-readable now compound. The ones that wait get sorted to the bottom of the agent's list. There's no catching up once the agent has already learned who to trust.
Step 1: Clean your product feed first
Agents don't guess. They match. A missing GTIN, a blank brand field, or a vague color name means the agent can't confirm what you're selling, so it skips you. Fill every required field in your Shopify product feed: GTIN, brand, condition, price, availability, and specific attributes like material and size. Google's Merchant Center product data specification is still the cleanest checklist for this, and agents lean on the same fields.
Most stores fail this. In our audits, a large share of products are missing at least one field an agent needs to recommend them.
Step 2: Make your buy conditions machine-readable
An agent won't recommend a product it can't price or confirm in stock. Add Offer schema to your product pages with price, priceCurrency, availability, and shipping details. Schema.org's Offer type is the standard, and it maps directly to what assistants read. When those signals are missing, the agent treats your listing as unverified and moves on.
Step 3: Answer the questions agents ask
Before an agent recommends, it looks for direct answers. Does it fit true to size? What's the return window? Is it in stock in blue? Add FAQPage schema and Product schema that answer these in plain sentences. Question-style headings help too, because they match how people phrase requests to an assistant.
Not optional anymore.
Step 4: Turn on a supported checkout path
Reading your product is half the job. The agent also needs a way to complete the purchase. Shopify has been rolling out agent-facing checkout support, and OpenAI's Instant Checkout uses the Agentic Commerce Protocol to let assistants finish a buy. Enable the supported path for your plan so an agent can actually transact, not just window-shop.
Step 5: Test with a real assistant
Open ChatGPT and Perplexity. Ask them to shop your exact category the way a customer would. See whether your products show up, and whether the details are right. If you're invisible or the data is wrong, that's your fix list. We ran this for a client last week and the store simply wasn't in a single answer. That's the starting line, not the finish.
FAQ
Do I need to be a developer to get ready for agentic commerce?
No. Most of the work is data hygiene and schema, which many Shopify apps and a good agency can handle. The technical intimidation is the biggest thing holding stores back, and it's mostly fields and structured data, not custom code.
How is agentic commerce different from AI search?
AI search gets your product recommended in an answer. Agentic commerce lets the agent complete the purchase. You need both, but the buy step raises the bar on clean data and supported checkout.
What breaks agent recommendations most often?
Missing product feed fields and no Offer schema. If an agent can't confirm what you sell, the price, and availability, it skips you. That's the pattern in nearly every audit I run.
How long before this matters for my store?
It already matters for discovery. Checkout is arriving in waves through 2026. The stores that clean their data now are the ones agents will trust first.
Want to know if an AI agent can actually read and buy from your Shopify store? Get a readiness check at WRKNG Digital's agentic commerce readiness page.
