§ 01 The shape of the problem
Why generic Shopify SEO advice fails by niche
A generic Shopify SEO checklist tells you to write a title under 60 characters, fill the meta description under 160, populate alt text, set canonicals, and submit a sitemap. That advice is correct for every vertical and decisive for none. What decides whether a skincare brand gets cited by ChatGPT or a furniture catalog ranks for 'modular sectional under $2,000' is the layer above the checklist — which Catalog fields the AI engines weigh in that vertical, what compliance language is allowed in product copy, and whether the category has already been captured by a brand or a regulator that no independent merchant can outrank with copy work.
Take three Shopify stores doing the same generic install. A skincare DTC brand fixes meta titles, ships product schema, gets eligible on Catalog, and waits — but every ChatGPT query for "best vitamin C serum" returns La Roche-Posay, CeraVe, Vanicream, and The Ordinary; the brand the merchant built does not appear at all, and 81% of facial-skincare recommendations4 are already concentrated in a single brand. A supplement merchant in the GLP-1 space does the same install and finds that two pharma companies own ~100% of citations across ChatGPT, Claude, and Perplexity5. A furniture brand ships pristine product schema and gets traffic — but the wrong traffic, because the AI agents weigh dimensions, shipping classes, and configurator state, none of which the generic playbook touches.
The generic install is not wrong. It's foundational, and every niche page on this site assumes it's in place. What goes on top is what actually moves the needle for a specific vertical. The Shopify Catalog optimization doc1 publishes the field list once for every merchant — Title, Description, Images, Product organization (Type, Vendor, Collections, Tags), Barcode, Variants, External product URL — but the relative weight of each field varies wildly by category. GTIN matters more for jewelry (when it exists) than for digital products (where it doesn't exist at all). Sizing schema matters more for apparel than for furniture. Local-pickup schema matters more for food-and-beverage than for skincare.
§ 02 The axes
Three axes that change per vertical
Three things change across verticals. The Catalog field weight — which of the seven AI-readable product fields the engines emphasize most for that category. The policy and compliance gate — what claims you can put in product copy without tripping Catalog's 'sensitive content' exclusion or industry-specific regulators. And the category-capture state — whether one or two brands already own the AI engine recommendations in your space, and what the independent merchant playbook is when that is true.
The field-weight shift is the easiest to underestimate. Shopify's AI optimization doc2 lists the recommended fields once, but the engines that consume Catalog data weigh them differently per category. Apparel queries lean heavily on Variants (size, color, material) and on the sizing/material/care instructions sub-fields. Skincare queries lean on Description and Tags (ingredient lists, dermatologist credentials). Furniture queries lean on Description (dimensions, materials) plus Images (lifestyle vs studio). Digital-product queries lean on Description and Type (because Barcode is empty and Images may be screenshots). The generic playbook treats all fields as equal and lets you choose what to optimize first. The niche playbook orders them.
The compliance gate is the second axis. Shopify Catalog requirements3 exclude products with "sensitive content, such as mature content," and the practical effect for supplements, skincare, and food-and-beverage is that any claim language that could be read as therapeutic — "cures," "treats," "reverses," "FDA-approved alternative to" — risks pulling the product out of Catalog entirely. The niche playbook separates marketing claims (homepage, blog, About) from structured-data claims (PDP copy, metafields, schema), which lets the merchant carry the truth the engines need without putting the catalog at risk.
81%
facial-skincare ChatGPT recommendation share held by La Roche-Posay in Q1 2026 (eMarketer / 5W AI Visibility Index, 5,200+ responses analyzed).
→ eMarketer · 2026-Q1 ~100%
of GLP-1 supplement citations across ChatGPT, Claude, and Perplexity concentrated in two pharma brands (5W Index, May 2026).
→ PR Newswire · 2026-05 7
AI-readable Shopify Catalog product fields — Title, Description, Images, Product organization, Barcode, Variants, External URL — weighted differently per vertical.
→ Shopify · 2026-05-22 § 03 Category capture
Category capture is real, measured, and asymmetric
Two of the seven niches on this pillar have published category-capture evidence severe enough to change the install. Skincare's facial category is 81% concentrated in La Roche-Posay alone across ChatGPT recommendations (5,200+ responses analyzed by eMarketer and 5W in Q1 2026). GLP-1 supplements are ~100% concentrated in two pharma companies across ChatGPT, Claude, and Perplexity (5W Index, May 2026). The other five niches have less measured concentration today but show the same pattern emerging — editorial-coverage incumbents capturing the recommendation surface before independent merchants build owned authority.
Category capture means a different thing than market share. A brand can hold 5% retail share and 80% AI-recommendation share if the AI engines have trained on disproportionate editorial coverage of that brand. La Roche-Posay's 81% facial-skincare share4 reflects the volume of dermatologist-endorsement content and clinical-credential signals in training data, not the brand's retail position. The implication for independent skincare brands is that the install cannot win on copy quality alone — it has to import credential signals the AI engines already weight (named formulator, board-certified dermatologist consultant, INCI-level ingredient transparency) or it cannot enter the citation set at all.
The GLP-1 capture is more extreme because it is regulatory plus pharma. The two pharma companies own the safety, efficacy, and clinical-trial coverage that AI engines treat as authoritative for any GLP-1 query, and supplement merchants in the "GLP-1 alternative" space are functionally invisible regardless of product quality5. The independent supplement playbook is therefore not "compete in the GLP-1 category"; it is "build owned authority in adjacent categories where capture has not yet happened" — sleep, magnesium, creatine, electrolytes — and let those compound while the regulated category settles.
§ 04 The install
How the $499 install differs from niche to niche
The mechanical shape of the install is the same across every vertical: the seven-day, $499, refund-backed Catalog Readiness Setup that ships the 12 deliverables (catalog audit, PDP rewrites, robots.txt.liquid audit, Knowledge Base setup, schema, policies, AI prompt tests, 30-day visibility report). What changes is which Catalog fields are weighted in the audit, which schema types are written, which compliance reviews run on the copy, and which AI engines and queries the 30-day report tracks against.
The audit half changes most. A skincare audit weighs Description, Tags (ingredient lists), and credential signals heavily; a furniture audit weighs Description (dimensions, shipping classes) and Images (lifestyle imagery); a digital-product audit weighs Description and Type because Barcode is empty by design. The compliance review changes too — supplements and skincare get a claims-language scan, food-and-beverage gets a health-claim scan, jewelry gets a hallmark/metal-purity verification, apparel gets a sizing/materials/care audit, furniture gets a dimensions/shipping-class verification, digital products get a delivery-method and refund-policy verification.
The build half is largely shared. Catalog eligibility, robots.txt.liquid, Knowledge Base FAQs, Product schema, Offer schema, and the policy pages (return, shipping, privacy) are populated the same way every install. The niche-specific layer adds the metafields and schema sub-types each vertical needs — INCI fields and dermatologist credentials for skincare, sizing schema and variant titles for apparel, GTIN with hallmark for jewelry, claims-compliance metafields for supplements, dimensions and shipping-class fields for furniture, local-pickup schema for food, no-shipping schema for digital. The pricing is the same for every vertical: one-time fee, no subscription, seven business days, 14-day Walk-Away Guarantee. The differences are surgical, not structural.
§ 05 Priority niches
Seven priority niches — built or building
Seven niches carry the priority cluster: skincare, apparel, jewelry, supplements, furniture, food-and-beverage, and digital products. Each has its own hub plus two or three leaves addressing the field, schema, or compliance mechanism that matters most in that vertical. Skincare and apparel get three leaves each (category capture, policy/claims, before-after handling for skincare; sizing schema, variant titles, seasonal collections for apparel) because the per-vertical surface area is larger. Jewelry, supplements, furniture, food, and digital products each carry two leaves.
The selection is not random. Each priority niche has (a) measurable Shopify Catalog and AI-engine surface area, (b) at least one published 2026 study documenting current AI-citation behavior in that space, (c) at least one niche-specific schema or compliance constraint that distinguishes the install from the generic playbook, and (d) enough volume of merchants in that space for the playbook to compound. Edge cases without all four (cannabis, firearms, alcohol-only, B2B-only) sit in the backlog until they meet the criteria.
The hub for skincare routes into three leaves: AI shopping category capture (the 81% problem), product claims and policies, and before-after images and the Catalog "sensitive content" line. Apparel routes into sizing schema, variant titles for AI parsing, and seasonal collections that decay. The other five niches carry parallel structures, accessible via the sub-page grid below.
§ 06 The transitions
Service add-ons — three transition states the niche pages don't cover
Three service add-ons sit alongside the seven niche playbooks. Not because they are themselves verticals, but because the install changes shape (rather than the niche) in three specific transition states: launching a new Shopify DTC brand from zero, migrating to Shopify from a legacy platform (BigCommerce, WooCommerce, Magento), and selling on Shopify Agentic without an online store at all. Each add-on inherits a niche playbook where the merchant already has one — a skincare brand launching its first DTC store reads the skincare hub plus the DTC-launches add-on — and stands alone where the niche is generic or yet to be added.
The DTC-launches add-on covers the first 90 days for a new Shopify DTC brand: when to enable Catalog (immediately), when to launch the Knowledge Base FAQ pipeline (week 2), when to invest in editorial coverage to feed AI citation density (week 4 onward), and which 30-day visibility report metrics to expect from a zero-history store. The legacy-migrations add-on covers the redirect-mapping work, the PDP-copy port (and rewrite where the source copy was manufacturer-sourced), and the schema re-emission from the new theme — the things that decide whether 12 months of acquired SEO equity survives the platform change. The agentic-only add-on covers the Shopify Agentic plan6 setup for merchants who want AI-channel reach without standing up a full Shopify online store.
Each add-on is a complement to the niche hubs, not a substitute. A skincare DTC launching from zero reads the skincare hub for vertical-specific schema and compliance plus the DTC-launches add-on for the launch sequencing. A furniture brand migrating from WooCommerce reads the furniture hub for the dimensions/shipping schema plus the legacy-migrations add-on for the redirect topology. A digital-products merchant selling only through ChatGPT and Copilot reads the digital-products hub for no-shipping schema plus the agentic-only add-on for the Agentic plan setup.
