Nobody sent you a memo. There was no press release addressed to CMOs. But sometime in March 2026, Shopify flipped a switch that quietly changed how your products get discovered and most marketing leaders I talk to have no idea it happened.
Here is what changed: Shopify activated agentic storefronts by default across its entire merchant base. Every store. No opt-in required. What that means in plain terms is that AI agents (the kind embedded in ChatGPT, Perplexity, Claude, and the shopping features being built into every major platform) can now query your product catalog directly, without ever loading your website.
The browse experience you spent years and millions of dollars optimizing? It is being bypassed. Not eventually. Now.
What an AI Agent Actually Sees
When someone asks an AI assistant to find them a product — “find me a natural vitamin C serum under $60” or “what are the best standing desks for small apartments” — the agent does not visit your homepage. It does not experience your brand story. It does not see your hero image or your lifestyle photography or the email capture popup you spent three weeks A/B testing.
It queries a catalog. It gets back a result set. It picks from that result set based on data quality, relevance signals, and how cleanly your product information maps to what was asked.
That is it. The entire front-end brand experience your team built: invisible to the machine making the recommendation.
I built a simple tool that demonstrates this directly. It is a search interface that queries Shopify’s Global Catalog API the same way an AI shopping agent does. You type in a product category. You see exactly what the agent sees: product titles, descriptions, pricing, variant structure, seller information. No storefronts. No brand design. Just raw catalog data, ranked by how well it answers the query.
When I show this to brand teams and they search their own category, the reaction is always the same. Some brands surface cleanly. Others are absent entirely. And the ones that surface are not always the brands with the biggest budgets or the strongest consumer recognition. They are the brands with the cleanest data.
The Ranking Signal Has Inverted
This is the part that should matter most to you as a CMO.
For the past decade, winning in ecommerce search meant brand equity plus ad spend plus SEO. You built awareness, you bought traffic, you optimized landing pages. Brand size was a meaningful advantage because it translated into domain authority, review volume, and budget.
Agentic search does not work that way.
When an AI agent queries a product catalog, it is evaluating what is actually in the data fields. Product title clarity. Description depth and relevance. Variant completeness. Pricing signals. Contextual metadata. A brand with $50 million in revenue but thin product descriptions and inconsistent variant naming will lose to a $2 million brand whose catalog is tight, complete, and well-structured.
That is a fundamental inversion. Brand equity built on the consumer side does not automatically transfer to the machine side. You have to earn visibility in both places now, and they require different things.
The consumer wants to feel something. The agent needs to understand something. Most brands have optimized exclusively for the first and ignored the second entirely.
What Poor Catalog Data Actually Costs You
I want to be concrete here because this can feel abstract until it hits a revenue number.
If an AI shopping agent cannot clearly understand what your product is, who it is for, what problem it solves, and how it compares to alternatives — it will not recommend it. It will recommend the product it can understand. That recommendation happens upstream of your website, upstream of your ads, upstream of everything your team controls.
You do not get a second chance at that moment. The agent made a decision and moved on.
Now multiply that by the volume of AI-assisted shopping queries happening today, which grew 15x in 2025 according to Shopify’s own data, and are accelerating. Every one of those queries is a decision point where catalog data quality determines whether your brand is in the conversation or not.
This is not a future problem. The queries are happening right now, across every category, at scale.
What Brands Need to Do About It
The good news is that catalog data is fixable. Unlike brand equity, which takes years to build, product data quality is an operational problem with operational solutions.
The work falls into a few categories.
Product title structure matters more than it ever has. Titles need to lead with what the product actually is, in the language a person would use to search for it, followed by the attributes that differentiate it. Generic titles that were written for shelf placement or internal SKU logic will not surface in agent-returned results.
Descriptions need to do real work. A two-sentence description that reads like marketing copy tells an AI agent almost nothing. Descriptions need to communicate use case, materials, fit, comparisons, and context. They need to answer the question someone might be asking, not just sell the product.
Variant structure and naming needs to be consistent and complete. An agent trying to evaluate whether a product matches a buyer’s criteria — size, color, availability, shipping time — needs that information to be structured and queryable. Gaps in variant data create gaps in recommendations.
Pricing and availability signals need to be current and accurate. Agents filter on these in real time.
None of this is technically complex. It is operationally complex — the kind of work that requires someone to own it, prioritize it, and execute it systematically across a catalog that may have thousands of SKUs.
The Window Is Narrow
Here is the strategic reality: the brands that move on this now will build a compounding advantage. Agentic search, like any search, will develop its own version of domain authority over time: patterns of recommendation that reinforce as agents learn which products consistently satisfy queries. Getting into that recommendation pattern early matters.
The brands that wait will spend the next 18 months watching a competitor show up in recommendations they should have owned, trying to reverse-engineer why their catalog is invisible.
I have spent the last several months building toward this at Avenue Z precisely because I believe it is the most important channel shift in ecommerce since the rise of paid social. The playbook is not written yet. The brands and agencies that write it will have an advantage that is very hard to close later.
If you want to see what your catalog looks like through the eyes of an AI agent, I can show you. The results are usually clarifying.
The AI-First Agency
Win AI search, grow revenue and build reputation through PR and digital marketing.
