The website audit that pays for itself in the first finding.

We use real browser automation to visit every page of your site, capture optimised screenshots, and read the rendered HTML and HTTP headers — exactly like a real user. A vision-capable AI then analyses each page across 9 categories and generates findings with severity ratings and fix suggestions.

Most scans complete in 5–15 minutes. Larger sites take longer — but you can leave the tab and we'll email you when it's ready.

No catch. Our cost-per-scan is genuinely low because the AI does the work, not a $200/hour consultant. We make it back on volume — and on customers who come back monthly to re-scan.

Page counts on every tier are caps, not quotas. We start at the URL you give us and crawl outward through your internal links until we either run out of pages or hit your tier's limit. If your site has fewer pages than the cap, you still get the full audit at the tier price — there's no "unused page" credit, but there's also no penalty for picking a tier that's bigger than you ended up needing. Pick the tier that comfortably covers the size of your site.

We'll crawl the first 500 pages following your internal link graph (homepage outward, breadth-first). For sites larger than that, run multiple scans rooted at different sections, or contact us — we can quote a custom enterprise scan.

Yes. We only crawl publicly-accessible pages. Screenshots and HTML are stored on your scan only and accessed via a private proxy — never exposed publicly. See our Privacy Policy for full detail.

Screenshots stay around for 90 days after a paid scan completes (30 days for free trial scans, 14 days for failed scans), then they're automatically deleted from blob storage. This keeps storage costs down — without a sweep, every scan would accrete cost forever for content nobody opens. Your audit findings, executive summary, persona reviews, scores, and page metadata are kept indefinitely; only the image files themselves are removed. Reports continue to render and PDF-export normally — they just show a placeholder where the screenshots used to be. If you need a long-term archive, download the PDF while the report is fresh.

An executive summary with health score, findings across 9 categories (Errors, UX, Accessibility, SEO, Performance, Design, Content, Security, Security Headers), severity ratings, an Impact × Effort matrix with your Quick Wins highlighted, page-by-page screenshots (at desktop and/or mobile, your choice), a security headers grade, and a downloadable PDF.

Yes. When you start a scan you pick Desktop, Mobile, or Both. Single-viewport scans are billed at the tier price you see on this page; choosing "Both" bills at 2× the tier price (a Standard scan goes from $5 to $10) because the worker captures every URL twice — 1440 × 900 with a desktop Chrome user-agent for desktop, 390 × 844 with an iPhone profile (touch input + iOS Safari user-agent string) for mobile, with a fresh browser context each time, both rendered in headless Chromium via Playwright. Two real renders, twice the AI cost, hence the doubled price. In return, device-specific findings (a hamburger nav that won't open, a hero image that's offscreen on iPhone, a CTA that needs scrolling on a 375px screen) actually appear in the report instead of hiding behind whichever single viewport you chose. Page-by-page view shows desktop and mobile screenshots side-by-side; findings carry a small viewport chip so you can spot mobile-only issues at a glance. The executive summary and the eight persona reviews are merged into one cohesive narrative that names the viewport when it matters. Note: this is Chromium-based emulation, not the actual WebKit engine — most layout, touch and viewport bugs reproduce, but a small class of true Safari-only rendering quirks (e.g. specific WebKit CSS bugs) won't.

Yes — pick a lens when you start the scan: Pre-sales, Pre-launch, Conversion, Investor or Brand polish. The executive summary, finding ranking, and persona ordering all shift to that voice (e.g. Conversion pushes CRO and UX to the top; Investor leads with security, performance, and trust signals). Lens is optional — leave it off for a neutral audit.

Same audit, three lenses on it. Story shows your top 5 findings as a pitch-ready narrative (great for stakeholders or clients). Quick Wins is an Impact × Effort 2×2 matrix — high-impact, low-effort fixes sit top-left so you can plan a sprint in 30 seconds. All Findings is the full filterable inventory by persona, category and severity, with rule IDs and copy-pasteable evidence for engineers. Switch between them with one click; the URL updates so you can bookmark a view or send it to a teammate.

Yes. Every finding carries an effort tag — Quick (<1 hour), Moderate (<1 day) or Involved (>1 day) — alongside its severity. That powers the Quick Wins matrix and lets you sort the backlog honestly: the cheapest, highest-impact work first.

Yes — every report comes with a unique private shareable link. No account required to view.

Yes — every paid scan unlocks a verified-domain badge. Prove ownership of your domain once (DNS TXT record, .well-known file or HTML meta tag — your choice), and we generate a signed, embeddable badge plus a public /verified/<your-domain> trust page that anyone can visit to confirm the score. The badge stays accurate because it always shows your latest scan, and the SVG is HMAC-signed so it can't be forged. Drop it in your site footer, README or pitch deck just like a code-coverage badge.

Yes. We use headless Chrome under the hood, so single-page apps render fully before we capture the page.

You get to retry it for free as many times as needed. If we genuinely can't access your site, we'll refund you.