Skip to content
StoreCited
Methodology

How we measure storefront readiness

A reproducible audit of observable public inputs—not a claim about proprietary ranking signals or live answer visibility.

01

Crawl

StoreCited fetches your homepage, a product page, a collection, and FAQ/about pages from the public web. Other crawlers can receive different responses, so this is a StoreCited fetch—not a proxy for every search or AI crawler.

02

Audit

We run deterministic checks across six categories and score observable public outputs available for independent review. The score does not measure another system's retrieval, interpretation, or placement.

03

Model questions

We generate category-specific buyer-question hypotheses and check whether the submitted public pages appear to answer them. We do not claim observed prompt demand.

04

List peers to verify

We infer category-peer candidates and provide a public-page checklist for the merchant to verify. StoreCited does not crawl those peers or claim that a feature is present.

The six categories

What goes into your score

Structured data

Product, Offer, Review, FAQ, and Organization JSON-LD label public facts. We check which types are detected in StoreCited's fetched output; field validity and visible-fact consistency require separate validation.

Product readiness

Product attributes—material, size, use case, dimensions, and ingredients—are public facts a shopper or system may evaluate. We grade only whether the fetched PDPs expose that detail; we do not observe matching or selection outcomes.

Buyer-question content

Useful FAQ, comparison, and buying-guide pages can supply direct evidence for high-intent buyer questions. We check whether that content exists and is accessible; we do not claim it was cited live.

Trust & reviews

Visible reviews and accurate machine-readable rating markup are separate checks. We inspect both without claiming a particular AI system will use the markup or treat it as a ranking factor.

Crawlability

Retrieval requires accessible pages. We check robots.txt, sitemaps, canonical tags, noindex, and the optional emerging llms.txt convention, while keeping crawler access separate from indexing or citation.

Brand entity

An about page, Organization schema, and accurate sameAs links make brand identity more consistent across public sources. They support entity clarity but do not create automatic trust or placement.

What we promise — and what we don't

We will never promise that ChatGPT or Google will recommend you—no outside tool controls that outcome. We report what the submitted public pages expose, which checks passed, and which observable gaps you can address. Indexing, citation, ranking, traffic, and revenue remain separate outcomes that require their own evidence.

Run your free scan

Frequently asked questions

Is this just AEO/GEO buzzwords?
No. Google says its AI features use the same foundational SEO practices as Search. We grade concrete public readiness checks and tie each finding to an observable fix; we do not convert those inputs into a promise of indexing, citation, recommendation, traffic, or sales.
Do you actually query ChatGPT and Perplexity?
No. The score is built on a deterministic crawl-and-audit of submitted public pages plus category modeling. It does not query a live prompt panel or claim to measure current mention share.
Why a score out of 100?
The score is an editorial readiness model that weights deterministic checks across six categories. It gives you a repeatable baseline for the submitted pages; it is not a calibrated probability of a mention, citation, ranking, click, or sale.
How accurate is the competitor list?
We infer category-peer candidates from the modeled category and label them as hypotheses. StoreCited does not crawl their pages or claim that they have a particular feature, ranking, citation, or recommendation. Use the supplied checklist to verify each public page independently before making a comparison.