How to read your SRPS score and fix it with technical briefs

What the Site audit looks at

The Analytics engine measures what selected AI systems say about the brand. The Site audit checks a different layer: whether retrieval paths can access and parse the company’s own site. The domain audit covers site-wide signals such as HTTPS, robots.txt, bot access, sitemaps, hreflang consistency and Organization markup. The page audit checks status code, canonical and noindex, H1 use, titles and descriptions, structured data, freshness, image alt coverage and link health.

Together they answer one question an external visibility tracker cannot: what technical condition may be stopping a retrieval path from finding or reading this page? Passing the checks establishes readiness; it does not predict citation.

How to read your SRPS score

Every page gets an SRPS — a 0–100 readiness score. It is not a simple checklist percentage: a missing title counts more than a missing image dimension. A check that cannot be evaluated is left out rather than counted against the page.

Read the tiers like a traffic light:

• 80–100 — good readiness. Most applicable checks pass.
• 60–79 — needs work. One or more applicable checks do not pass.
• below 60 — poor readiness. Review high-priority checks and blockers first.

Two details change how you read a low number:

• A failed blocker makes the score 0. If the page errors, blocks AI crawlers, or is set to noindex, the score is 0 — not “a bit lower” — until that gate is fixed.
• SRPS vs SRPS+. Plain SRPS comes from the standard audit of the page HTML. SRPS+ comes from the deep audit, which renders the page and adds extractability, content and lab-performance signals. SRPS+ is absent until a deep audit has run.

The domain score is not the average of your pages. It combines page scores with checks that only make sense for the whole site — robots.txt, the sitemap, whether the apex and www versions agree — so fixing one page moves it a little, and fixing a site-wide setting moves it more.

Why a page scores low

In practice, low scores cluster into a few causes:

• Access. robots.txt blocks a relevant search or retrieval bot, or conflicting Allow/Disallow rules do so by accident. That retrieval path cannot use the page while access is blocked.
• Indexability. A stray noindex, a canonical pointing at a different page, or a non-200 status quietly removes the page from consideration.
• Thin structure. No structured data, or markup that exists but carries no real facts (a Product with no price or availability), or a missing/oversized title and description.
• Graph problems. A page no internal content links to (an orphan), or one buried many clicks deep, is hard for any crawler — and any reader — to reach.

Current manual checks, such as indexation verification and visual review of og:image, are excluded from SRPS. Marking them done records the review but does not change the score.

Fixing it with a technical brief

This is the part that makes the score actionable. Deterministic technical findings with a defined remediation include an implementation brief assembled from the check and its evidence. It can provide:

• the problem and affected scope;
• step-by-step remediation;
• a ready-to-use artifact where appropriate, such as a robots.txt rule, JSON-LD block or corrected meta tag;
• acceptance criteria and relevant implementation cautions.

Because the brief is structured text, you can hand it to a developer, put it in a ticket or give it to an authorized agent. Not every finding has this deterministic artifact. Content-oriented deep findings such as answer-capsule quality or content uniqueness use on-demand brief generation, while some informational checks have no generated fix. After deployment, the next audit re-checks the same signal: a passing check can auto-resolve the technical recommendation; a failing check stays open.

A weekly habit

You do not need a full crawl every day. Free provides the initial audit snapshot; paid plans can refresh audits on their schedule. Use deep-audit quota on high-priority pages, including pages already observed in citations. Read the domain score for site-wide issues, sort pages by SRPS, clear blockers, then work through access, indexability and structure. When checks pass, SRPS rises. Whether AI later retrieves or cites those pages is measured separately.