Crawl Explorer
See every page the way Google does — and turn every technical issue into a task you can act on.
SEO problems hide in the gap between what a page claims and what a crawler actually retrieves, and most audits flatten that into a single score you can't act on. Crawl Explorer reads your site's own crawl page by page, so every title, canonical, directive, and rendering signal is evidence you can open, filter, and export.
Crawl Explorer indexes every URL on your site and lays out its status, depth, metadata and health signals in one filterable table. No sampling, no black box — the same crawl the rest of Novaverb runs on.
What you get
Nothing hides
Every crawled URL with HTTP status, canonical, indexability, depth and internal links — filter and export any signal in seconds.
Find the real blockers
Duplicate clusters, JS-render gaps and data-quality flags are surfaced automatically, so you fix what actually stops pages from ranking.
Evidence you own
Each crawl is a timestamped record you keep — compare scans over time and prove a fix actually landed.
Inside Crawl Explorer
Eighteen focused intelligence cockpits
One crawl feeds eighteen grouped tabs, each a purpose-built workbench for a single decision: titles, meta descriptions, H1s, content depth, images, canonicals, indexation directives, hreflang, structured data, internal links, JavaScript, security, URL hygiene, performance, redirects, HTTP status, sitemaps, and Search Console overlay. Every tab carries its own filters, quick-stat drill tiles, priority lanes, and a click-to-open drawer that ties each finding to a what/why/how fix card. You never leave for a second tool to answer a title, canonical, or directive question.
- Title state: missing, duplicated, too short, too long, weak
- Meta description length band (healthy 120-160) and duplication
- H1 count, duplicate H1, and title-mismatch flags
- Canonical relationship: self, cross, or missing plus target health
Data quality that never fakes clean
The bundled Data Quality view runs seven trust checks live over your crawl and leads with evidence coverage, how many checks it can actually run, instead of a made-up cleanliness score. Crawl-provable checks like soft-404 inferred, thin content, and needs-recheck return real counts; checks that require Search Console, a render pass, or Google's selected canonical are marked insufficient, never green by absence. A page is only called clean when a check actually ran and passed.
- Evidence coverage shown as N of 7 checks runnable
- Soft-404 inferred: 200 status with almost no extractable content
- Thin content and needs-recheck from word count and extraction
- GSC render gap, index mismatch, canonical shadow gated as insufficient
Duplicate and render risk, bundled
Duplicate Clusters groups pages that share an exact title, meta description, or H1, then risk-ranks each cluster by canonical posture, competing versus already-consolidated, and suggests the safest primary URL to keep. JS Render Health proves render-blocking load per page and the server-rendered text already in your initial HTML, and stays honest that raw-versus-rendered delta needs a render pass. Both are exact-match and crawl-derived, so body-level near-duplicate and rendered-only claims are surfaced as insufficient rather than asserted.
- Exact title, meta, and H1 clusters with unique URLs affected
- Canonical posture: consolidated, competing, missing, mixed
- Suggested primary URL from indexability, inbound links, depth
- Render-blocking resource count and server-rendered word count
What you can see
Concrete signals this system surfaces — measured, never fabricated.
From signal to action
Run a crawl
Point a project at your domain and Novaverb crawls it, capturing status, content, directives, links, and rendering signals into the latest scan.
Open a cockpit
Pick the tab for the decision at hand and read its KPI strip, priority lanes, and drill tiles scoped to that scan.
Filter and drill
Use the smart-filter dropdowns, status facets, and issue chips to narrow to the exact URLs that prove a problem.
Fix or export
Open any row's drawer for a diagnosis and fix steps, or export the current tab and filters as CSV for your team.
What's included
Built for how you work
Triage a large site by opening the exact tab and filter for a title, canonical, or indexation issue instead of scrolling one giant report.
Hand clients an evidence-backed URL list per problem, exported to CSV, with the canonical or directive signal that proves it.
Find thin, duplicate-title, or weak-H1 pages and see the fix guidance in the same drawer before rewriting.
Built to be trusted
Common questions
Do I need to connect anything to start?
No. Crawl Explorer works entirely from your own crawl, though checks that need Google-selected canonical, index coverage, or a render pass stay marked insufficient until you connect Search Console or run rendering.
What does an insufficient or needs-evidence label mean?
It means the check could not be proven from the signals captured, so Novaverb refuses to call it pass or fail. The URLs are neither confirmed clean nor flagged until the missing evidence lands.
Can I export what I am looking at?
Yes. Every tab streams the current tab plus its active filters to CSV, so the export matches exactly what is on screen.
How does it handle JavaScript-rendered content?
The static crawl proves render-blocking load and the text already present in your initial HTML. Raw-versus-rendered delta, render timing, and rendered-only content are shown as insufficient until a render pass runs, rather than guessed.
Crawl Explorer, and the whole platform behind it.
Request your invite to start, or see exactly what's included at each tier.