Free tool · Flagship grader

Shopify SKU Visibility Grader.

Your best-seller can rank #1 on Google and still be invisible to ChatGPT. Paste your store URL: the grader samples five PDPs from your live catalog, scores each 0–100 on the signals AI engines reward, and checks your robots.txt against 15 AI crawlers.

Free No signup ~90-second result
Report manifest
Brand-level average AI-visibility score across 5 sampled PDPs (0–100) and tierfree
robots.txt verdict for 15 AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and morefree
Top 3 brand-wide findings, ranked by severityfree
Per-PDP score for every sampled productemail
Every remaining finding with the concrete fixemail

Rate limit: 1 run per 30 seconds per IP. The grader reads only public data.


How it works

1 · Sample.  The grader reads your public /products.json feed and selects five PDPs from the live catalog. No app install, no admin permissions, no theme edit.~10s
2 · Parse.  Each PDP is fetched and parsed: Product JSON-LD, description depth, canonical tags, plus store-level robots.txt and llms.txt.~60s
3 · Score.  Four weighted buckets per PDP, averaged into a brand score with a tier. Findings ranked by severity, each with a named fix.~15s

What each check maps to

Every finding the grader raises has a fix documented on this site. The table below is the map from check to fix.

CheckWhat a failure costsFix guide
Product JSON-LD completenessEngines can't resolve price, availability, or identity; the PDP loses citation slots to retailers that publish clean schema.Schema for AI search
Content depth & answer coverageThin descriptions can't answer buyer questions, so engines quote review blogs instead of your PDP.Product AI visibility
robots.txt AI-crawler accessBlocked bots mean the engine never reads the page. The most common silent failure on Shopify stores with bot-blocking apps.Crawl surface for Shopify
llms.txt presenceWithout a curated map, AI crawlers see every URL in the sitemap with no priority signal for flagship SKUs.llms.txt generator
Canonical hygieneVariant and collection-path duplicates split the citation surface across URLs engines treat as different products.SKU-level AEO

How scoring works

Each sampled PDP gets a 0–100 score from four weighted buckets: structured data (30%) checks Product structured data for name, brand, SKU, GTIN, offers, availability, and aggregateRating; content depth (30%) measures description length and whether the copy answers the questions buyers ask — material, fit, compatibility, care; crawler access (30%) reads robots.txt admittance for AI bots and canonical hygiene; llms.txt (10%) checks for a curated crawl map at the domain root. The brand score is the average across the sample.

Tiers band the composite: 80–100 is ready, 60–79 partial, 40–59 work to do, below 40 foundational gaps. The score measures readiness to be cited — the input you control. Whether engines actually cite the product on a given query also depends on review signal and competitive density; the ChatGPT checker tests that live. The full methodology is in the AI visibility score glossary entry.

Why SKU-level rather than brand-level: a brand can have healthy aggregate AI mentions while its best-seller is absent from every shopping answer. Perplexity cites 3–7 sources per shopping answer; the question is whether your PDP is one of them, product by product. That mismatch is invisible to brand trackers — see SKU-level tracking for the longer argument.

Beyond the sample

Five PDPs is a diagnosis. The catalog is the treatment.

eCommerce Insights runs these checks against every SKU weekly (daily on Growth), tracks citation movement across six engines, and ships the title, schema, and metafield fixes as reviewable diffs. A Monday-morning digest reports the delta.

Ask AI about the SKU Visibility Grader

Have your favorite AI engine summarize this tool for your store.

Frequently asked questions

Do I need to install anything to grade my Shopify store?
No. The grader reads your public storefront — /products.json, sampled PDPs, robots.txt, and llms.txt if one exists. No Shopify app, no admin permissions, no theme edit. The paid eCommerce Insights product uses the admin API to push approved fixes; the free grader is strictly read-only.
Which products does the grader check?
It samples five PDPs from your store's public /products.json feed and grades each one individually. The sampled URLs are listed in the result so there is no guessing. To audit one specific product in depth, run the AEO Grader or the ChatGPT checker on that URL instead.
What does the 0–100 score measure?
Four weighted buckets per PDP: structured data, content depth, crawler access, and llms.txt. The store score is the sample average. It measures readiness to be cited — the controllable input. Live citation testing is the ChatGPT checker's job.
Why does robots.txt matter for AI visibility?
If GPTBot, ClaudeBot, PerplexityBot, or Google-Extended are disallowed, those engines cannot read your PDPs, and your products cannot be cited from your own pages — the engine falls back to retailers, review blogs, and Reddit threads. The grader checks 15 AI crawlers and flags every blocked one.
How does the free grader differ from the paid product?
The free grader checks five sampled PDPs once. The paid product checks every SKU on a schedule, adds citation tracking across all six engines, and ships diff-style PDP fixes that push to Shopify through the admin API. Same scoring model; catalog-wide coverage and cadence. Pricing.

Grade five products free. Track every product weekly.

The grader answers today's question. The product answers it every Monday.