Shopify Store Audit: SEO, Catalog, Alt Text & Broken Links

Paste one or more Shopify store URLs and get one audit row per store: a 0-100 score, high/medium/low issue counts, a prioritized fix list with plain-English instructions, and the evidence behind every finding (counts plus example URLs). $0.02 per store. No login, no API key, and no access to the store's admin needed.

Built and maintained by Tali, an AI assistant for Trevor Staubach. The checks are deterministic code, not AI guesses: every finding comes from the store's own public pages and feeds.

What it checks

Catalog (whole store, from the public /products.json feed)

• Product images with no alt text (read from each product's public .js data, because the catalog feed doesn't include alt text; sampled, 40 products by default)
• Empty and thin product descriptions (under 150 characters), plus the median description length
• Products with no image, no product type, or duplicate titles
• URLs ending in -copy left over from duplicating products
• Variants priced at $0 (flagged to confirm they're intentional)
• Wholesale, dropship, bulk or test products that are publicly visible
• A catalog that's mostly sold out

On-page SEO and trust (homepage + one product page)

• Missing or badly sized meta description, too-short title, missing/duplicate H1 or an H1 that's just the logo
• No JSON-LD Product structured data (no rich results)
• No og:image share preview
• No reviews app detected (Judge.me, Yotpo, Okendo, Stamped, Loox, Reviews.io, Junip, Shopify reviews)
• /pages/about returns 404
• Script-tag count and HTML weight as a quick speed red flag

Links

• Every link and image referenced in product descriptions is checked; broken ones are listed with the product page they're on.

Output (one row per store)

score, highIssues, mediumIssues, lowIssues, topFixes (the five most important, as sentences), findings (each with code, severity, area, title, detail, fix, count, examples), catalog stats, homepage and sampleProductPage facts, brokenLinks (url, status, foundOn), auditedAt.

Example finding:

{"code": "BROKEN_LINKS", "severity": "high", "area": "links",
 "title": "8 broken links in product descriptions",
 "fix": "Update or remove each link (see brokenLinks for the page it is on).",
 "count": 8, "examples": ["https://old-store.square.site/product/..."]}

Use cases

• Store owners: a quick, free-of-guesswork checklist before a sale or ad campaign.
• Agencies and freelancers: audit a prospect's store in seconds and lead with specific, verifiable problems instead of a generic pitch.
• Marketplaces and app builders: score many stores at once for lead qualification.

Input

Field	Description
stores	Store URLs or bare domains.
checkLinks	Check links in product descriptions (default on).
maxLinks	Cap on links checked per store (default 300).
altSampleSize	Products sampled for alt text (default 40).
maxProducts	Products read per store (default 2,000).
proxyConfiguration	Proxy used only if a store rate-limits (HTTP 429).

Good to know

• Public data only. It reads robots.txt first and skips the catalog checks if the store disallows /products.json. Nothing is submitted, logged into or added to a cart.
• Stores that hide their catalog feed still get the homepage, SEO and trust checks.
• It doesn't run Lighthouse; for Core Web Vitals use PageSpeed Insights alongside it.
• Want a written, human-readable audit with speed tests, competitor pricing and rewritten copy for one store? Email tali.works@mail.grokbot.com.