How Kweri works
Five expert lenses. One clear verdict. A site that keeps improving.
Drop in a URL and five parallel audits run — performance, accessibility, technical SEO, UX, and messaging. You get a prioritised, plain-English report: what’s wrong, why it matters, what to fix first. Then you fix things, come back, and watch the score move.
Three steps
From URL to verdict in seconds.
Paste your URL
Nothing to install. No account needed to start. Just your address and sixty seconds.
Five experts go to work in parallel
Performance, accessibility, technical SEO, UX, and messaging — each assessed by a dedicated agent simultaneously, the way a senior team would split the brief.
You get a verdict, not a wall of numbers
A prioritised, plain-English report: your score, the issues ranked by impact × effort, why each one matters, and exactly what to fix first. Built for founders and marketers, not engineers.
The five lenses
The full picture, not just the easy stuff.
Scroll the five lenses. The report on the right shows the kind of thing each one surfaces on a real page.
Performance
What's making visitors wait — and what it's costing you.
Real-world speed and Core Web Vitals, measured the way Google measures them. Every second of delay costs you conversions. We tell you exactly where time is being lost.
Largest content takes 3.8s to appear
SeriousPerformanceYour hero image is 1.2 MB and loads before anything else. Visitors stare at a blank screen while it downloads — many leave before it finishes.
Accessibility
Who your site is quietly turning away.
Automated WCAG checks plus the human-readable issues automated tools miss. Not a compliance checklist — a genuine read of who can and can't use your site.
Body text fails contrast on your brand grey
SeriousAccessibilityLight-grey copy on white sits at 3.1:1 — below the 4.5:1 minimum. Anyone reading in sunlight or with low vision loses whole paragraphs.
Technical SEO
The foundations search engines — and AI crawlers — need.
On-page signals: title tags, meta, structure, canonical, robots, schema. Plus AI-readiness checks — whether AI crawlers can reach you and whether your structured data is machine-readable. Most tools don't look at this. We do.
No structured data on your key pages
ModerateTechnical SEOThere's no schema markup, so search engines and AI crawlers can't tell a price from a phone number. You're invisible to rich results and harder to cite.
User experience
A designer's eye, not just a checklist.
A genuine heuristic critique — visual hierarchy, CTA strength, cognitive load, journey clarity, trust signals — framed around Nielsen's principles and Laws of UX. The kind of verdict a senior UX designer gives after an hour on your site.
The main action isn't obvious enough
SeriousUX“Book an appointment” sits below the fold and looks like every other link. Visitors who are ready to act can't find the one button that matters, so they leave instead.
Messaging
Does your page say what you do, who it's for, and why — in five seconds?
Whether visitors understand your value proposition before they scroll. Whether your CTA copy does any work. Whether the words on the page are pulling their weight. If they're not, we say so, and we tell you what to change.
Visitors can't tell what you do in five seconds
SeriousMessagingThe headline leads with a slogan, not a service. Someone landing cold can't tell what you sell or who it's for before they bounce.
Honest by design. Every report tells you what an automated check can see and what it can’t. No jargon, no scare tactics, no inflated issue counts.
How the score works
A score that reflects what actually matters.
The overall score is not a mathematical average of the four pillar scores. It’s an AI verdict on how effectively the page converts visitors — and UX and messaging carry the most weight.
A page can load in under a second, pass every accessibility check, and rank perfectly on Google — and still score in the 50s if the messaging is off or the hierarchy fails to guide the eye. Technical excellence doesn’t compensate for UX friction.
That’s the point. Most audit tools give you speed and metadata. Kweri tells you whether a stranger would stay, trust you, and buy.
Full scoring methodology →Score bands
- 85–100Minor polish remaining
- 70–84Meaningful but fixable gaps
- 55–69Notable friction with real conversion cost
- 40–54Significant structural or messaging problems
- Below 40Fundamental issues blocking most visitors
UX & messaging carry the most weight
The improvement loop
The first audit is just the beginning.
Most audit tools give you a report and leave. Kweri is built around what happens next. Fix something, come back, run the audit again — your score updates, your trend line moves, and a cross-page synthesis surfaces the patterns recurring across all your pages.
Re-audit and see the delta
Before: 61. After your fix: 68. Here's exactly what moved and why.
Starter and above
Trends across your pages
Not five disconnected reports — one synthesised verdict across your whole site.
Advanced and above
Close the loop on findings
Mark findings to-do, in progress, done. Flag what you disagree with. Re-audit picks up where you left off.
Starter and above
Re-auditing and fix tracking start on Starter. Trends, scheduling, and cross-page synthesis are Advanced features. See what’s on each plan →
Built for what’s next
Kweri audits for human visitors and AI agents.
AI readiness signals evolve quickly. We review and update our checks as the landscape changes.
Search isn’t just humans on Google anymore. AI crawlers — GPTBot, PerplexityBot, Claude, Gemini — are reading and indexing the web right now. Whether they can reach your site, understand your structure, and cite your content is a new kind of readiness most audit tools don’t check.
Kweri’s Technical SEO pillar includes AI-readiness checks as standard: whether AI crawlers are permitted in your robots.txt, whether your llms.txt is valid, and whether your structured data is machine-readable. It’s the audit for the web as it exists today, not as it existed five years ago.
The Kweri difference
We tell you what we can see, and what we can’t.
Automated accessibility tools catch around 57% of real accessibility issues. We know that — and we say so in every report. Performance data comes from Google’s own infrastructure. UX and messaging are assessed by a language model calibrated against real designer judgement — not a keyword counter — and we tell you so.
No jargon, no scare tactics, no inflated issue counts designed to push upgrades. The diagnosis is always honest, and it’s always free.
What an automated check sees
Measurable- Colour contrast ratios
- Largest Contentful Paint and Core Web Vitals
- Missing alt text and form labels
- Title tags, meta, canonical and schema
- Broken links and redirect chains
What needs a human read
Judgement- Whether your message actually lands
- If the call to action is persuasive, not just present
- Whether the visual hierarchy guides the eye
- If the tone fits who you're selling to
- Whether a stranger would trust you enough to buy
Kweri labels every finding — you always know which is which.
Ready to see what your site is actually doing?
Free audit, no account needed. Start with one page. Come back for the rest.