Published by the Cited Research Team | April 24, 2026

Elena Martinez, head of digital marketing at a fast-growing DTC skincare brand, opened ChatGPT on a Thursday morning in March 2026. Her brand had 47K Instagram followers, a 4.8-star rating on Trustpilot, and $3.2M in annual revenue. She typed: "What's the best vitamin C serum for sensitive skin?" ChatGPT recommended four products. Hers wasn't one of them.

Her counterpart at a competing brand—smaller social following, similar price point—ran the same test the next day. Different product category, different query, but the same outcome: invisible. Neither brand appeared in Claude's recommendations either. Both were absent from Perplexity's results.

We wanted to know whether this pattern held across e-commerce. So we ran a structured test across 200 DTC brands, 3 AI platforms, and 80 product recommendation queries spanning 40 categories from skincare to home goods. The results explain why 89% of e-commerce brands are invisible to AI—and what the 11% who get cited are doing differently.

The Test: 200 Brands, 80 Queries, 3 AI Platforms

We tested 200 direct-to-consumer e-commerce brands across eight major verticals: beauty & skincare (40 brands), apparel & accessories (35), home & kitchen (30), fitness & wellness (25), pet products (20), baby & kids (20), tech accessories (15), and outdoor gear (15). These weren't marketplace sellers—they were established DTC brands with owned domains, active marketing, and strong social media presence.

What we tested:

• Brands: 200 DTC brands with $500K-$50M annual revenue, owned Shopify/custom sites, active social media

• Queries: 40 general product searches ("best running shoes for beginners"), 40 specific product comparisons ("Allbirds vs Veja sneakers")

• AI Platforms: ChatGPT (GPT-4), Claude (Claude 3.5 Sonnet), Perplexity (Pro)

How we tested:

• Each query was run on all three platforms within the same 48-hour window

• We logged every brand mentioned, product cited, and ranking position

• We analyzed the product pages and site structure of cited brands

• We cross-referenced AI visibility with social media following and review counts

A total of 200 brands × 80 queries × 3 platforms = 48,000 query-platform combinations. We logged every mention, every citation, and every recommendation. Here's what we found.

The Headline Numbers

Only 22 of the 200 brands appeared in AI recommendations. Across 48,000 query opportunities, just 22 brands earned citations. That's 89% complete invisibility. The average e-commerce brand in our test appeared in 0.4% of relevant queries—meaning for every 250 times a shopper asks AI for a product recommendation in their category, they're mentioned once.

89% of tested brands lack Product schema with detailed attributes. We audited the technical infrastructure of all 200 sites. Only 23 had implemented Product schema with comprehensive attributes (material, size, color, care instructions, specifications). The 22 brands that earned citations? All 22 had detailed Product schema—plus structured data for reviews, Q&A, and brand information.

Brands with user-generated content were 6.7x more likely to get cited. Of the 22 cited brands, 21 had active review sections with Review schema markup and Q&A sections with structured data. Only 47 of the 200 tested brands had structured UGC, meaning 45% of brands with Review schema got cited, versus 0.6% without.

Citation rates by product information completeness:

71% of all AI recommendations went to the same 9 brands. Across all platforms and queries, 9 brands dominated: Allbirds, Glossier, Warby Parker, Parachute, Bombas, Casper, Away, Outdoor Voices, and Everlane. These weren't always the brands with the largest social followings—but they were the leaders in structured product data and user-generated content.

Product comparison queries had 4.1x higher citation rates than general searches. Brands appeared in 18% of comparison queries ("Brand A vs Brand B") but only 4.4% of general product searches ("best running shoes"). The gap: comparison queries trigger AI to look for detailed product specifications and reviews, which only 11% of brands have structured properly.

What the 22 Cited Brands Had in Common

When we analyzed the digital presence of the 22 brands who earned citations, every single one shared five structural traits. These aren't social media tactics or paid advertising strategies—they're fundamental product data architecture decisions that make product information machine-readable.

Trait 1: Comprehensive Product Pages with Detailed Attributes

All 22 cited brands implemented Product schema with extensive attributes: material composition, dimensions, weight, care instructions, sustainability certifications, and manufacturing details. Allbirds' product pages include structured data for wool sourcing, carbon footprint, and washability. When ChatGPT recommends Allbirds for sustainable sneakers, it's citing these structured product attributes as evidence.

Trait 2: User-Generated Reviews with Review Schema Markup

21 of the 22 cited brands implemented Review schema for customer reviews, including star ratings, verified purchase indicators, review dates, and reviewer attributes. Glossier's product pages display 200-500 reviews per product with structured markup that AI can parse for sentiment, skin type mentions, and specific product attributes. Brands with Review schema had 47% citation rates versus 5% for brands with unstructured reviews and 2% for brands with no reviews.

Trait 3: Detailed Size Guides, Fit Information, and Product Comparison Tools

All 22 cited brands provided structured sizing information, fit guidance, and product comparison features. Warby Parker's virtual try-on and detailed frame measurements use structured data that AI can reference. Everlane's product pages include fabric weight, transparency pricing, and factory information with proper markup. These details enable AI to make specific recommendations based on user needs.

Trait 4: Active Q&A Sections on Product Pages

19 of the 22 cited brands maintained active Q&A sections with FAQPage schema markup. These sections answer common questions about sizing, materials, care, and compatibility. Parachute's bedding pages include 30-50 answered questions per product with structured markup. The average cited brand had 180 structured Q&A entries; the average non-cited brand had 8.

Trait 5: Strong Brand Story and About Content

All 22 implemented Organization schema with detailed brand information: founding story, mission, sustainability practices, and manufacturing transparency. Away's "About" page includes structured data about B Corp certification, carbon neutrality, and supply chain transparency. This brand context helps AI understand positioning and recommend brands aligned with user values.

The Invisibility Problem—And Why It's Actually Your Opportunity

89% invisibility sounds dire, but it's actually the opening. Here's why: e-commerce GEO is where Google Shopping was in 2012—almost nobody is doing it systematically, which means the first movers lock in citations that compound. Many of the invisible brands in our test have strong Instagram followings and healthy DTC revenue. They've built great products and communities. But AI doesn't read Instagram—it reads structured product data, and 89% of brands haven't implemented it.

The gap between the 22 cited brands and the 178 invisible ones isn't marketing budget or social media prowess. Several cited brands were sub-$5M revenue companies competing against brands with 10x the Instagram following. The gap is structural: cited brands have invested in product data architecture—Product schema, Review markup, Q&A structure, detailed attributes—that AI can parse and validate. Invisible brands have great products and happy customers—but that information isn't documented in machine-readable formats.

This creates a massive opportunity for the 89%. The competitive moat around AI visibility in e-commerce is still shallow. Product schema implementation takes days, not months. Review markup is a technical project, not a content rebuild. Q&A structure is an inventory and formatting task. The brands who make these changes in Q2 2026 will appear in the next wave of citations. The ones who wait will find themselves competing not just against traditional rivals, but against AI-native brands who built for machine-readable product data from day one.

How to Become One of the 22

Based on the five traits shared by cited brands, here's the implementation order we use at Cited when working with e-commerce clients:

Step 1: Audit Current Product Schema and Review Implementation (Week 1)

Inventory your existing structured data. Do your product pages include Product schema with detailed attributes? Do customer reviews use Review schema? Do you have Q&A sections with FAQPage markup? Most brands discover they have 20-30% of the necessary structure—basic product info exists, but attributes, reviews, and Q&A aren't marked up for AI consumption. Use Google's Rich Results Test to identify missing schema types.

Step 2: Implement Comprehensive Product Schema on Top 20-30 SKUs (Weeks 2-3)

Start with your best-selling products and add detailed Product schema: material, dimensions, care instructions, sustainability attributes, and manufacturing details. Include AggregateRating schema for review summaries. For apparel, add size ranges and fit information. For beauty, add ingredient lists and skin type recommendations. The goal: make every product attribute machine-readable.

Step 3: Add Review Schema to Existing Customer Reviews (Week 3)

Implement Review schema for your existing customer reviews. Include star ratings, verified purchase indicators, review dates, and helpful vote counts. If you have 1,000+ reviews, prioritize your top products first. The 22 cited brands in our test averaged 340 structured reviews per top product; the 178 invisible brands averaged 12 unstructured testimonials.

Step 4: Create Product Comparison Pages and Detailed Size/Fit Guides (Week 4)

Build structured comparison tools that help shoppers choose between products or understand sizing. Implement FAQPage schema for common questions about fit, care, materials, and compatibility. Add detailed size guides with measurement charts. These structured resources give AI concrete information to cite when making recommendations.

The Competitive Window

E-commerce GEO is where Google Shopping was in 2012: almost nobody is doing it, which means the first movers lock in citations that compound. Of 200 established, well-marketed DTC brands, only 22 are being cited by AI. That's 11%.

The brands who implement the structural changes above between now and Q3 2026 will appear in the next cohort. The ones who don't will find themselves competing not just against traditional rivals, but against AI-native brands who built for machine-readable product data from day one—and against the 22 incumbents who are already capturing 71% of all AI recommendations.

Our test was run in March 2026. We'll rerun it in September. The brands who make the structural changes above between now and then will appear in the next analysis. The ones who wait will watch their competitors capture an increasingly large share of AI-driven product discovery—even as they continue to build Instagram followings.

If you want to see exactly how your products appear across ChatGPT, Claude, and Perplexity for your target categories, learn more about our GEO services—we'll show you which of the five structural traits your site is missing, which competitors are currently being cited in your category, and the fastest path to becoming one of the 22.

The e-commerce brands that win in 2026 won't be the ones with the biggest Instagram followings. They'll be the ones who made their product data machine-readable before their competitors realized shopping had moved to AI.