> ## Documentation Index
> Fetch the complete documentation index at: https://docs.surnex.io/llms.txt
> Use this file to discover all available pages before exploring further.
# Site audits
> Crawl your site, score it, and see the technical SEO issues holding it back.
**Audits** lists every crawl you've run on the project, newest first. Each audit is a point-in-time snapshot — Surnex doesn't overwrite the last one, so you can compare crawls over time.
## The audit list
Each row shows:
* **Score** — the audit score, colour-coded: green at 80+, amber at 50–79, red below 50
* **Pages · issues** — `120 pages · 38 issues`
* **When it ran**, as a relative time
* A **status badge** — pending, processing, completed, or failed
Click a row to open its [detail view](/audits/issues).
## Running an audit
Click **Run Audit**. The panel pre-fills **Target URL** with `https://` and your project domain, so most of the time you just click **Start Audit**.
| Field | Notes |
| ---------------------- | ----------------------------------------------------------------------------------------------------------- |
| **Target URL** | Where the crawl starts. Pre-filled with your project domain; change it to crawl a subdirectory or subdomain |
| **Max pages to crawl** | Default **50**, minimum 1, maximum **10,000**. "Higher values take longer but provide more coverage" |
The crawl runs in the background — you can leave the page. The row appears immediately with a pending status and updates as the crawl progresses.
Audits do **not** run on a schedule. One runs automatically when the project is created, and after that only when you click **Run Audit**. An audit score months out of date usually means nobody has crawled since. See [How data collection works](/projects/data-collection).
### Choosing a page limit
The default 50 is a sample, not a site audit. It's enough to catch sitewide template problems — a missing meta description pattern, no schema, mixed content — because those repeat on every page.
Raise it when you need per-page findings: orphan pages, duplicate titles, and thin content only surface if the crawl actually reaches the pages involved. For a full picture, set the limit above your real page count.
Larger crawls take proportionally longer, and audits count against your plan's monthly audit allowance regardless of size. See [Plans](/billing/plans).
## The audit score
The score starts at 100 and deducts for each issue found, weighted by severity:
| Severity | Deduction each | Maximum total deduction |
| ----------- | -------------- | ----------------------- |
| **Error** | −2 | 50 points |
| **Warning** | −0.5 | 30 points |
| **Notice** | −0.1 | 10 points |
Three bonuses add back up to 15 points:
* **+5** if the site uses HTTPS
* **+5** if `robots.txt` is present
* **+5** if a sitemap is present
The result is clamped to 0–100.
Two consequences are worth understanding. Because deductions are capped per severity, a site with 25 errors scores the same as one with 250 — past the cap, the score stops distinguishing. And because the bonuses are flat, a small site missing a sitemap is penalized as heavily as a large one.
Treat the score as a coarse indicator for reporting, and work from the [issue list](/audits/issues) for anything actionable.
## Issue categories
Every issue falls into one of eight categories:
| Category | Covers |
| --------------- | -------------------------------------------------------------------- |
| **meta\_tags** | Titles, descriptions, H1s — missing, duplicated, too long, too short |
| **links** | Broken links, redirect chains and loops, orphan pages |
| **performance** | Slow pages, large pages, large images, uncompressed resources |
| **images** | Missing alt text, broken images |
| **indexing** | `noindex`, robots blocks, missing canonicals, non-indexable pages |
| **content** | Thin content, duplicate content |
| **security** | Mixed content, missing HTTPS |
| **schema** | Missing structured data |
See [Audit issues](/audits/issues) for every issue type, its severity, and how to fix it.
## Empty state
**No audits yet** — "Run a site audit to check your website for SEO issues". Projects created normally have one audit from their initial run, so an empty list usually means that first crawl failed.
## Related
Every issue type and what to do about it.
One crawled page's issues and metadata.
What you fixed and what regressed.
Core Web Vitals, measured separately from the crawl.