Four things, and what each one actually covers

• ✓ 26 checks across five categories: crawler access, structured data, content structure, freshness, brand authority
• ✓ A 0–100 score and an A–F grade, with the category weights published on the methodology page
• ✓ The evidence behind every failed and partial check — the actual robots.txt line, the actual missing field
• ✓ A fix list ordered by priority, effort ascending, so the cheap wins surface first
• ✓ Hard gates called out separately: blocked crawlers, noindex and JS-only content cap the score, and the report shows both the capped and uncapped number
• ✓ Any URL, in about a minute, with no signup and no site access

Scope: Runs on any public URL. Benchmarking a score against an industry average is available for twelve industries; the audit itself is industry-agnostic.

• ✓ Run against one exact product URL, independently of the site-wide readiness scan
• ✓ Per-platform evidence, and the report names which platforms it actually queried on that run
• ✓ Commerce platform detection, with integration notes specific to the platform found
• ✓ Two scored categories — merchant ecosystem presence and commerce platform integration

Scope: Product pages only, one URL per run. Which AI shopping engines are queried is a server-side setting, so the report states the engines it checked rather than promising a fixed list.

• ✓ Three engines — ChatGPT, Gemini and Claude — with multiple samples per run, on commercial-intent queries only
• ✓ Two share-of-voice numbers: head-to-head against the competitor set you declare, and full-field against every brand the answers named
• ✓ The full field of brands the models mentioned, split into your brand, competitors you declared, and brands you never declared — any of which you can promote into the tracked set
• ✓ Brand sentiment scored per dimension on the samples that actually discussed the brand, with the distribution, not just an average
• ✓ Per-engine breakdowns kept on separate charts, because the two share numbers are not on the same scale
• ✓ Sample size and a 95% confidence range on every rate — and an explicit “no range yet” below eight queries

Scope: Query banks exist for Jewellery, Fashion & Apparel, B2B CRM. Other verticals aren't supported yet. Runs are on a weekly cadence and are triggered from the dashboard today — see below.

• ✓ A portfolio table across the whole roster: brand presence, full-field share of voice, citation rate, readiness grade, last run and alerts
• ✓ Sample size under every rate in the table, and a confidence band wherever the sample supports one
• ✓ Each client keeps its own tracked brand, its own competitor set and its own history — switching client switches the whole dashboard
• ✓ Pause and swap clients as the roster changes, without losing what came before
• ✓ One subscription for the roster: 1, 5, 15 or 30 clients, priced per agency rather than per domain

Scope: Client allowances are per plan, from one on Solo to thirty on Agency Scale. Reports are viewed in the dashboard or shared as a link; there is no exported or branded deliverable yet.