Why "We Do AI" Doesn't Mean Anything Anymore
Every agency rewrote their website in 2024. Now they all specialize in AI marketing. All of them.
The problem is that AI commerce became a label anyone could attach to an existing service deck. Most agencies renamed their old work, added some screenshots of ChatGPT, and started pitching it as transformation. The actual expertise? Rarely present.
Choosing the wrong agency is expensive. Not because you'll burn money on bad ads. Because you'll waste 12 months while your competitors build the AI visibility that should have been yours.
What "AI Commerce Specialization" Actually Requires
Real expertise here is narrow and technical. It means understanding how platforms like ChatGPT Shopping, Google AI Overviews, and Perplexity pull product data, and knowing exactly what your Shopify store needs to show up in those answers.
According to Google's Merchant Center product data quality guidelines, AI-powered Shopping results depend on feed completeness: accurate titles, detailed descriptions, GTIN values, and structured attributes. That's not a vague recommendation. It's a checklist with clear pass/fail outcomes.
An agency with real expertise can tell you specifically which attributes your feed is missing and why that matters for AI recommendations. Generic talk about "using AI to improve your marketing" is a different thing entirely.
Red Flags That Tell You the Expertise Isn't Real
These patterns show up again and again across agency pitches. None of them mean the agency is dishonest. They do mean the AI commerce knowledge isn't there.
They Lead With Their Tech Stack, Not Your Store
If the first 20 minutes of a call is about their AI tools, their automation platform, their reporting dashboard, that's a problem. The conversation should open with your product data, your current AI visibility, and where the gaps are. If they haven't asked to see your product feed before the pitch, they don't know enough to audit it.
They Can't Explain How ChatGPT Decides What to Recommend
Ask this question directly. If the answer involves content marketing, domain authority, or ad spend, they don't have it. AI recommendation engines work from product feeds, structured data, and entity signals. An agency that can't walk you through that process isn't operating at that level.
Their Case Studies Are From 2022 or Earlier
The rules changed fast in 2024 and 2025. ChatGPT Shopping launched. Google's AI Overviews expanded to cover product queries. Perplexity rolled out shopping features. If their proof of expertise predates all of this, it doesn't tell you much about what they can do today.
They Treat AI Advertising and AI Commerce as the Same Thing
Running campaigns through Meta's Advantage+ or Google Performance Max is AI-assisted advertising. Worth doing. But it's a separate discipline from making your products visible to AI shopping assistants. Many agencies conflate these, sometimes intentionally.
What to Ask Before You Sign Anything
These questions will surface real expertise, or the lack of it, in about 15 minutes.
"Can you audit our product feed for AI visibility right now?"
A specialist will say yes and immediately describe what they're looking for: missing GTINs, thin attribute data, unstructured product descriptions. Shopify's structured data documentation outlines the product schema fields that directly affect how AI systems read your catalog. A real agency knows this document.
"Which AI platforms currently index our products, and how?"
They should name ChatGPT Shopping (powered by Bing's product index), Google's Shopping Graph (which feeds AI Overviews), and Perplexity's commerce integrations. If they name only Google and stop there, they're behind.
"Show me a before/after on AI citation rates for a client."
Vague answers about "improving visibility" won't cut it. Ask for specific metrics: feed completeness scores, AI appearance rates by product category, or prompt-level tracking showing their client's store appearing in AI answers where it didn't before.
"What's your process for tracking AI platform changes?"
The rules here shift regularly. Google's AI Mode expansion, Shopify's OpenAI partnership, Perplexity's new commerce features, these are all moving targets. According to Salesforce's 2024 Connected Shoppers Report, 53% of consumers now expect AI to help them during a shopping journey. That number will keep climbing. The agency you hire should know where this is going, not just where it's been.
"Is AI commerce a separate deliverable from your SEO work?"
If everything folds into one "digital marketing package," the AI-specific work probably isn't happening. Structured data audits, feed completeness scoring, and AI citation tracking are distinct from traditional SEO. They need to show up as distinct line items.
What a Real AI Commerce Agency Actually Delivers
The output looks different from a standard marketing retainer.
Product feed audits with specific scores. A detailed review of every feed attribute against what each AI platform requires before it makes a product recommendation, not just general notes about "opportunities."
Structured data builds. Schema markup for products, reviews, pricing, and availability. The markup that Google's AI Overviews and ChatGPT Shopping actually parse when deciding what to surface.
AI citation tracking. Ongoing monitoring of whether your products appear in ChatGPT, Perplexity, or Google AI answers for the queries your customers are actually using in 2026.
I've run product feed audits on dozens of Shopify stores over the past year. The stores that have done this work are showing up in places their competitors aren't. The gap keeps growing.
Not great for the stores that waited.
The Bottom Line
Most agencies claiming AI expertise are running traditional digital marketing with a new name on the invoice.
The ones doing real AI commerce work can walk you through your product feed gaps in 15 minutes. They know which platforms matter and why. Their case studies are from 2025, not 2022. They measure results against AI citation rates, not just traffic numbers.
Those agencies exist. They're just rarer than their websites suggest.
If you want to know where your Shopify store actually stands with AI shopping platforms before you spend a dollar on an agency, start with a free AI Commerce audit from WRKNG Digital.
Frequently Asked Questions
How do I know if a marketing agency actually understands AI commerce?
Ask them to explain how ChatGPT Shopping decides what to recommend. If the answer involves content marketing, SEO authority, or ad spend, they don't know. Real expertise means understanding product feed requirements, structured data standards, and how platforms like Google's Shopping Graph and Bing's product index actually work.
What's the difference between AI advertising and AI commerce?
AI advertising means running campaigns through platforms like Meta Advantage+ or Google Performance Max. AI commerce means making your products visible to AI shopping assistants like ChatGPT, Perplexity, and Google AI Overviews. Both matter, but they require different skills. Many agencies conflate them, sometimes on purpose.
How much should an AI commerce marketing agency cost?
There's no fixed rate, but be skeptical of any agency charging a premium for AI services without separate deliverables that prove the work is happening. Product feed audits, structured data builds, and AI citation tracking should appear as visible line items, not rolled into a general retainer.
What results should I expect from an AI commerce agency?
Measurable improvements in product feed completeness scores, structured data coverage, and AI appearance rates for your most important product categories. Any agency that can't show you those specific numbers after 60 days probably isn't doing the work.
Can I improve my Shopify store's AI visibility without an agency?
Yes, for the basics. Shopify's built-in structured data handles some product schema automatically. You can also improve feed quality through your product descriptions and attribute fields. But gaps in GTIN data, custom schema, and ongoing AI platform monitoring are harder to manage without a specialist.
