Every page read, every problem shown with its evidence
Our own crawler reads your whole website, over 150 deterministic checks run across twelve categories, and every finding opens the pages it affects and the element itself. The health score shows its working, right down to the one critical problem that capped it.
Over 150 checks across twelve categories, each finding opening the pages it affects.
Latest analysis
68Health
Capped at 68 by one site-wide finding
Missing structured data on 212 of 240 pages. Clear it and the score is free to move.
4 critical17 high38 medium154 checks run
By category
Weighted, with every deduction shown
On-page SEO81
Indexability94
Technical SEO72
Content58
Performance64
Internal links77
AI crawler access100
Structured data41
Worst findings
Pages affected
CriticalMissing structured data212
HighThin content46
HighImages without alt text184
MediumOrphan pages6
HighSlow largest paint on mobile62
Hasten AIReads yourwebsite.com
Is our website being cited by AI engines?
Hasten AIWorked through 3 steps
Yes. Your pages are cited in 18% of answers across 1,284 tracked answers, most often on ChatGPT and Perplexity. Rival One is cited in 31%, mostly for one page that answers the question in its first 90 words.
The quickest gain: add an answer block to your 14 product pages.
Every crawler can tell you that 312 pages have a meta description problem. Almost none of them will tell you which one to fix first, what it is costing you, or whether the fix you shipped last week actually worked.
Most scores are unexplained, too. A number between 0 and 100 with no working shown is a number you cannot argue with and cannot act on, and it will happily sit at 78 while one line in robots.txt removes your whole website from search.
GrowthHasten shows its working on every score, caps the score when something site-wide is broken and names the finding that capped it, and puts the element itself next to each finding so you can see the problem rather than read about it.
Issues and health score
The audit runs deterministically before anything is allowed to interpret it. Findings are grouped by section and by type, ranked by severity and by how much of your site they touch, and the health score shows every category weight, every deduction and anything it could not measure.
Over 150 named checks across twelve categories
Category weights, deductions and unscored categories, all on screen
Score ceilings: a site-wide critical problem caps the score, and the finding that capped it is named
Severity bands, and the count of pages each finding affects
Every count opens its rows. 12 pages is a link to those 12 pages
Over 150 checks across twelve categories, each finding opening the pages it affects.
Latest analysis
68Health
Capped at 68 by one site-wide finding
Missing structured data on 212 of 240 pages. Clear it and the score is free to move.
4 critical17 high38 medium154 checks run
By category
Weighted, with every deduction shown
On-page SEO81
Indexability94
Technical SEO72
Content58
Performance64
Internal links77
AI crawler access100
Structured data41
Worst findings
Pages affected
CriticalMissing structured data212
HighThin content46
HighImages without alt text184
MediumOrphan pages6
HighSlow largest paint on mobile62
Hasten AIReads yourwebsite.com
Is our website being cited by AI engines?
Hasten AIWorked through 3 steps
Yes. Your pages are cited in 18% of answers across 1,284 tracked answers, most often on ChatGPT and Perplexity. Rival One is cited in 31%, mostly for one page that answers the question in its first 90 words.
The quickest gain: add an answer block to your 14 product pages.
Ask Hasten AI↑
Page explorer
The full inventory: status, indexability, click depth, links in and out, word count and every issue found on that page. Filter it, sort it, export it, and open any page to see the whole story for that URL.
Status codes, indexability and click depth for every page
Pages with issues, error pages and pages not linked from the home page
Words per page, and the distribution across the site
CSV export of any filtered view
Audit / PagesIllustration
Pages crawled
240
Indexable
214
89%
With issues
137
↓ 18
Orphans
6
↓ 2
URLCodeDepthIssuesWords
/200001,240
/pricing20012860
/product/ai-visibility200112,110
/blog/old-post404340
/resources/legacy301410
Core Web Vitals
The Core Web Vitals that decide how Google treats your pages, measured on mobile and desktop across a sample of your site, with the improvements that come up most often pulled out so you fix the pattern rather than the page.
LCP, INP and CLS, mobile and desktop
Speed score with its band, and the spread across sampled pages
The improvements suggested most often across your site
Joined to the audit, so a slow template shows up as a site-wide finding
Audit / PerformanceIllustration
Speed score
78
↑ 11
LCP
2.1s
↑ 0.4slargest paint
INP
142ms
↑ 38msinteraction
CLS
0.04
↑ 0.03layout shift
Mobile vs desktop
Desktop92
Mobile64
Suggested most often
Pages
Serve images in next-gen formats184
Reduce unused JavaScript121
Defer offscreen images96
And in the same place
The parts that make the rest worth having
Visual evidence
Each finding shows you the thing it is about: the link in its context, the image, the title as it would appear as a search result, the social card, the structured data, a clipped screenshot of the page. Reading about a problem and seeing it are different jobs.
The search result preview, as Google would draw it
The social card, as it would unfurl
The link in the surrounding paragraph
The image, at the size it is actually being served
Structured data, parsed
Content analysis
Content scored page by page, with the spread shown so you can tell a few bad pages from a systemic problem. Suggested fixes draft a title, a description, an H1 or alt text from the page as it reads now and the queries it already earns.
Thin content, duplicate content and missing metadata, counted and opened
A content score per page, and the spread across the site
Suggested titles, descriptions, H1s and alt text, drafted from the page itself
Opportunities ranked by what the page already earns in search
Internal links
Your internal link graph as a structure rather than a list: how deep pages sit, which have no way in, which have no way out, and the specific internal links worth adding, with the anchor text to use and the reason.
Orphan pages: reachable by nobody
Click depth, and everything four or more clicks from the home page
Links per page, and pages with no outbound links
Linking opportunities with suggested anchor text and the reason
Images
Every image the crawl found, what is wrong with it, and what fixing it is worth in real kilobytes. Accessibility and performance in the same list, because they are the same list.
Missing alt text, counted and opened
Images far larger than they are displayed
Formats that should be modern and are not
Potential savings, in real bytes
Crawls and crawl control
Start a crawl and watch it work: pages read, URLs found, average response time, duration, and what it noticed about your site files. Choose how pages are rendered, which paths to skip, and how often it runs by itself.
Live progress, pages crawled, URLs found and average response
Site files: robots.txt, sitemap and llms.txt, found or missing
Render mode: served HTML, fully rendered, or automatic
Paths to skip, so a staging folder or a calendar never eats the crawl
List check: crawl exactly the URLs you paste, apart from the site's analyses
Crawl history and comparison
Every analysis is kept with its own stored score and findings. Open an old one exactly as it was, or compare any two: the health score, the issues that got better or worse, and the pages added, removed or changed.
Every crawl, with when it ran, what it read and what it found
Any past analysis opened on its own stored score
Compare two analyses: score, issues and pages
What changed since the last crawl, as a standing screen
Website audit report
Technical SEO, content, performance and search for this website, as one printable document with a read-only share link. The thing you send a client instead of a screenshot.
One document covering the audit, the search data and the work
A read-only share link anyone can open
PDF, with your own name and logo on paid plans
Where every figure came from, on the figure
Why this one
What you get here that you do not get elsewhere
A score you can argue with
Category weights, deductions, anything unmeasured and any ceiling are all on screen. You can trace a 62 back to the exact findings that made it a 62.
Evidence, not descriptions
Every finding can show you the element: the link in context, the title as a search result, the image at the size it is served. You see the problem instead of reading about it.
Hosted, scheduled, watched
Nothing to install and nothing left running on your laptop. Crawl on demand or on a schedule, and a lighter daily check catches a page going down the day it happens.
It proves the fix
Mark something fixed and GrowthHasten re-reads the affected pages, re-runs the same rule, and records the result. Verified means verified.
The crawler handles up to 100,000 pages. What your account may crawl is your plan's page limit, which runs from 500 pages on Free to 50,000 on Advanced, and higher by arrangement on Enterprise.
What does the health score actually mean?
It is a weighted average of twelve category scores, and the weights are on screen. A finding costs its category in proportion to how much of your site it affects, with a floor per severity so one critical page is never rounded away. A critical problem that reaches the whole site caps the score, and the screen names the finding that capped it.
Is it a hosted crawler or do I install something?
Hosted. There is nothing to install. Crawls can run daily, weekly or monthly on a schedule, and a lighter daily check watches for a page going down, a noindex appearing or robots.txt changing.
Can I control how it reads my site?
Yes. Choose whether pages are read as served HTML or fully rendered, set paths to skip, set the schedule, and run a list check against exactly the URLs you paste when you only want to verify a handful.
And it is only a third of the platform
Website Audit shares a workspace, a queue and an agent with the other two. That is the whole point.