NovaBrain severity
Supported findings are ordered by the severity and confidence encoded in the active NovaBrain criterion, not by a promised ranking effect.
Site Health Audit
A source-aware technical checklist that prioritizes supported findings and records fresh rechecks.
Site Health Audit evaluates captured project URLs against supported NovaBrain SEO and AEO checks, orders findings by NovaBrain severity and records later evidence-based rechecks.
Why it is useful
A clear decision, with the supporting evidence close by
A wall of red and green checks tells you something is wrong but not what to fix first or whether a green even means anything. Site Health Audit grades your crawl against a defined criteria library, keeps missing evidence separate from real pass and fail, and turns the result into a ranked work-list that jumps straight to the affected pages.
Guidance where supported
A finding shows its captured evidence and available guidance; an AI draft is offered only when that action and model are configured.
Fresh evidence recheck
A later crawl can re-run the same measured check. Findings update when the required fresh evidence is available.
Data and evidence
See the evidence behind each result
Site Health Audit reads your crawl and grades it against a defined criteria library. Its health percentage is calculated from graded checks only: where the crawl lacks the signal a criterion needs, that criterion is counted as needing evidence rather than as a pass. That separation is why a green here means something was measured.
Primary input
Crawl, sitemap, optional GSC
Product layer
Evidence-gated audit criteria
Primary output
Severity-ranked repair plan
How to read this result
The health score uses graded checks only; checks awaiting another source remain in their own evidence state.
Core capabilities
What you can do with Site Health Audit
The audit turns a crawl into a ranked work-list: errors, warnings and passes with a severity and an affected-page count per criterion, a fix-first queue that jumps straight to the page at fault, the sitemap view alongside it, and the history of earlier runs so a change can be seen rather than assumed.
01
Honest health, not a vanity score
The audit runs your crawl against the active NovaBrain criteria and grades each check as error, warning, or pass, with the health score computed only from checks that could actually be graded. Checks the crawl lacks evidence for are held out as needs-evidence, and offline-only or manual checks get their own states, so a green is always a real pass on captured data. The evidence strip shows exactly which sources, crawl, criteria, Search Console, and domain verification, back the grade.
- Audit health as percent of graded checks passed
- Error, warning, and pass counts per criterion
- Needs-evidence, skipped-offline, and manual states kept separate
- Evidence strip: crawl URLs, active criteria, GSC days, domain verified
02
Fix-first queue with jump-to-page
Every graded issue is ranked by severity and number of affected pages into a single prioritized queue covering both crawl findings and criteria checks. Each row carries its evidence source and a link that opens the precise Crawl Explorer view, the missing-title list, the broken-status lane, the thin-content filter, so you go from problem to page in one click. The same builder powers the standalone Fixes tab, so the priority order never contradicts itself.
- Ranked critical, warning, and opportunity lanes
- Affected page count per issue
- Domain-not-verified and crawl-error (4xx/5xx) blockers surfaced first
- Deep links into the exact Crawl Explorer tab and filter
03
Sitemap and history in one shell
The Sitemaps tab fetches and parses your XML sitemap and cross-references it against the crawl to expose bad entries, indexable pages missing from the sitemap, and declared-but-uncrawled URLs, with a coverage percentage and lastmod freshness. The History tab keeps one row per crawl, date, duration, URLs crawled, errors, and trends health over time from the real run ledger. Both stay honest that Google submission and index coverage are not connected, so acceptance by Google is never implied.
- Sitemap coverage percent and covered-versus-indexable count
- Bad entries: noindex, non-200, or canonicalized-away pages
- Missing-from-sitemap and declared-but-uncrawled URLs
- Per-crawl history: URLs crawled, errors, and health trend
How the work moves
From available evidence to a reviewable next action
Crawl the site, grade it against the criteria, work the queue from the top, then verify and track. Verification re-runs the same criterion against a fresh crawl, so a fix is closed by new evidence rather than by someone marking it done.
- 01
Crawl the site
Novaverb crawls your project and captures the status, content, and directive signals every criterion reads.
- 02
Grade against criteria
The audit engine runs the active criteria library and grades each check, holding out anything it lacks evidence to judge.
- 03
Work the queue
Open the fix-first queue, start with the top blocker by affected pages, and click through to the exact URL list.
- 04
Verify and track
Re-crawl to confirm a fix, then watch the health score and issue counts trend on the History tab.
Built around real work
Useful for teams who need to move from evidence to delivery
In-house SEO teams
Run one audit each crawl and hand engineering a severity-ranked list where every item links to the pages to fix.
Agencies managing many sites
Track each client's health score and error trend across crawls and export the graded criteria list for reporting.
Site owners and business owners
See the single top blocker and its affected page count without needing to interpret a full technical report.
Connected outcomes
Continue from this product into a complete customer workflow
Fix & Improve
How Can I Help Search Engines Discover New Pages Safely?
Make new pages reachable through clean crawl paths, accurate sitemaps, indexable directives and supported submission methods.
Open workflowFix & Improve
Which Technical SEO Issues Should I Fix First?
Fix the issues that combine verified severity, meaningful reach and clear affected URLs before lower-impact cleanup.
Open workflowFix & Improve
Why Is My SEO Work Scattered Across Different Tools?
SEO work fragments when research, crawl evidence, content decisions and reporting lose their shared project context.
Open workflowDiscover & Diagnose
Why Can't Search Engines Find Some of My Pages?
Pages become hard to discover when internal links, crawl depth, sitemaps or indexability signals leave no clear path.
Open workflowMeasure & Prove
How Can I Prove SEO Work Was Actually Completed?
Compare timestamped crawl evidence from before and after the change, then verify that the failing condition is gone.
Open workflowDiscover & Diagnose
Why Did My Organic Traffic Change?
Interpret the change by metric, source and comparison window before deciding whether it is good, bad or only contextual.
Open workflowResources
Read more about Site Health Audit
Written by the team that built it, and free to read without an account.
Guides
Step-by-step walkthroughs, five minutes each.
Definitions
What the terms on this page actually mean.
Try it free
Run the idea on a real URL, no account needed.
In the app
How this works once you are inside a workspace.
Connected products
Keep the work moving without losing the source
Frequently asked questions
Site Health Audit FAQ
Clear answers about data, availability and how the product fits into the wider workflow.
How is the health score calculated?
It is the percentage of graded checks that passed, where graded means error, warning, or pass. Checks the crawl could not provide evidence for are excluded, so the score never inflates by counting ungraded checks as passing.
Why are some checks marked needs-evidence or manual?
A check is needs-evidence when the crawl did not capture the signal it reads, and manual when there is no automated detector for it. Both are shown as their own states rather than being forced into a pass or fail.
Does the sitemap check confirm Google accepted my sitemap?
No. It cross-references your fetched XML sitemap against the crawl to find coverage gaps and bad entries, but it is explicit that Google submission and index-coverage state are not connected.
Can I export the audit?
Yes. The Overview and Issue Explorer views export the graded criteria list, including status, group, severity, affected URLs, the evidence basis, and the exact pass condition, as CSV.
Technical SEO & Crawl
Put Site Health Audit to work on your own evidence.
Start with a real project, keep unavailable data visible and follow the connected workflow when the result is ready for action.