UX Audits
Senior UX/UI audits that give you a defensible answer to “what should we fix first?” — across strategy, design, accessibility, and content. Delivered in 1 to 10 days.
Anyone can run Lighthouse and dump the output into a deck. The audit is the toolkit. The offer is judgment applied to the toolkit — which of these 39 findings actually matter for your business, and what should you do first.
You're shipping fast and the site reflects it. An audit tells you which of the dozens of “we'll fix that later” items matter now — and which are noise.
I won't step on toes. I audit the unowned territory — SEO, accessibility, content hygiene, conversion path, AI visibility — and hand the design team a brief, not a critique.
Pick by the question you need answered. Most engagements start with Discovery or Strategic Health Check and expand from there.
Fast strategic read. What does the site say, who is it for, and where is it leaking attention?
Includes
For: First-time clients, intro engagements, decision-makers who want a strategic read before committing.
Non-design audit for teams with an in-house designer. The marketing-site layer that designers don't own.
Includes
For: Pre-redesign briefing, post-launch checkup, teams with a designer they don't want me to step on.
Strategic + light UX. Hands your designer a defensible brief instead of starting from "what do you want?"
Includes
For: Teams about to brief a designer or agency. The audit pays for itself in scope clarity.
Everything. Strategic, IA, design craft, heuristics, accessibility — with a prioritized action plan.
Includes
For: Founders without a designer, pre-launch product teams, agencies needing a senior contractor.
WCAG 2.1 AA compliance audit with keyboard, screen reader, cognitive, and inclusive-language layers.
Includes
For: Compliance-driven engagements, post-ADA-letter clients, enterprise procurement, public-sector contracts.
What ChatGPT, Claude, Perplexity, and Google AI Overviews say about your brand — and what to do about it.
Includes
For: Thought-leadership brands, consultancies, B2B services. Almost nobody else is selling this yet.
How pricing works
Every package is fixed-fee — the price on the card is the price. A 20-minute scoping call confirms which package fits and whether the engagement needs an add-on. Get in touch.
Three published reports that show the framework end-to-end. Same severity coding, same tier structure, same prioritized action list every engagement delivers.
Pick from the menu. Run 4–6 per engagement, not all 39. The right mix depends on whether you're pre-redesign, post-launch, in content phase, or in fundraising mode.
Tier 1
Strategic
What the site says, who it's for, where it sits, what it asks
01
What does a first-time visitor learn in 30 seconds?
Read the homepage as a stranger. Write down what the company does, who it's for, and what they want you to do. Then list the questions you still have.
02
Who is the site speaking to vs. who should it be?
Reverse-engineer the implied audience from the language, examples, and case studies. Compare against the target audience the team says they want to reach.
03
What's the claim, what's the proof, what's the ask?
Strip the homepage to its three jobs. Score each on clarity 1–5. Surface where each is doing less work than it could.
04
Who else is in this space, what do they own, what's unclaimed?
3–7 direct or adjacent competitors mapped on a positioning grid. Where they own ground, where you can.
05
What is the site semantically about, according to its own copy?
Three columns — terms you own (proprietary, distinctive), terms you compete on (generic), terms missing that should be present.
06
What proof of credibility is on the site, and what's hiding?
Catalog clients, partners, press, awards, speaking, citations, credentials. Then catalog what should be there but isn't.
07
Can a visitor actually complete the ask?
Walk the primary CTA end-to-end. Note every friction step. Score each. Recommend the friction-to-cut order.
08
What do ChatGPT, Claude, Perplexity say about you?
Test the brand across 4 AI surfaces with 10 questions. Citation quality scorecard. robots.txt strategic review.
09
When was the site last meaningfully updated?
Blog dates, case study currency, team roster, footer copyright. Flag where the site looks dormant.
10
Are the boring credibility basics present and current?
Privacy policy, ToS, cookie compliance, broken outbound links, current "we are hiring" claims, partner logo currency.
Tier 2
UX & Interaction
Task flows, navigation, forms, states, microcopy, responsive
11
Does the structure match how users think?
Map the nav tree. Identify orphans, duplications, depth issues, category overlap. Optional tree test with 5–10 users.
12
Can a visitor find what they came for, fast?
5–8 representative tasks timed. Note dead-ends, mis-labels, back-button moments. Specific label and structure recommendations.
13
Does each user goal flow end-to-end without breaks?
Top 3–5 jobs the site is hired for. Walk each. Map friction. Score completion confidence per job.
14
Is the form doing its job or filtering out leads?
Every form scored. Field-by-field recommendations. Benchmark data on conversion impact per fix.
15
Does first-run experience get users to value fast?
Sign up cold. Document every step. Time-to-first-value. Define a recommended "first value" milestone.
16
Are empty, loading, error, edge states designed?
Coverage matrix per state per surface. Green/yellow/red. Punch list of missing states ranked by severity.
17
Does the language sound like the same person, doing its job?
Pull every CTA, button label, error, empty state, confirmation. Score for clarity and consistency. Before/after rewrites.
18
Does it hold up across the actual device mix?
iPhone, Android, iPad, desktop. Screenshot grid, parity issue list, breakpoint refinement recommendations.
40
Is the copy free of typos, and does the voice stay consistent?
Proofreading + copy editing pass across every published page. Typos, doubled words, missing prepositions, voice drift, brand-name spelling consistency. Most recent addition to the menu — numbered 40 to preserve the original 39 framework.
Tier 3
Visual & UI
Hierarchy, type, color, brand, components, motion
19
Does the eye land where it should?
Squint test, blur test, attention-flow trace per surface. Annotated screenshots showing intended vs. actual hierarchy.
20
Is the type system intentional, scalable, readable?
Catalog every font/size/weight/leading. Map against a type scale. Accessibility check (min 16px body, scalable units).
21
Is the palette intentional and accessible?
Extract every color. Map to a defined palette. WCAG AA contrast checks on every text/background pairing.
22
Does the brand look and sound the same across surfaces?
Site, email, social, app stores, sales deck. Cross-surface grid. Drift list. Canonical brand reference recommended.
23
Real design system, or each surface bespoke?
Component inventory. Variant rationalization. Hardcoded values that should be tokens. Starter token set recommendation.
24
Are icons and images brand-coherent, or filler?
Style consistency (one icon family, not three). Purpose check. Quality. List of stock placeholders to replace.
25
Does motion serve the experience or distract?
Animations, transitions, parallax, hover, scroll. Timing, easing, accessibility (prefers-reduced-motion). Keep/cut list.
Tier 4
Research & Insight
Heuristics, usability, personas, journeys, analytics, VoC
26
Does the experience follow established usability principles?
Walk each surface against Nielsen's 10. Severity-rated violations (cosmetic, minor, major, catastrophic). Top 5 fixes.
27
Where do real users stumble?
5–8 representative users, moderated or unmoderated. Per-task success rates. Friction pattern themes. Prioritized fixes.
28
Who are we actually building for?
Evidence-grounded personas. JTBD, behaviors, pain points, decision triggers, quotes. No stock photos, no imaginary demographics.
29
What does end-to-end look like across owned + unowned touchpoints?
Awareness to loyalty. Search, social, ads, site, product, support, email. Emotion, friction, time per stage.
30
Does data match team assumptions?
GA4, Mixpanel, Amplitude. Entry, exit, scroll, click maps, funnel drop-off. Assumption-vs-reality gaps. Experiments to run.
31
What are customers saying — and what patterns are missed?
Support tickets, NPS, App Store, public reviews, sales calls. Themed and ranked. Trends. Top 5 recommendations.
Tier 5
Accessibility & Inclusion
WCAG, keyboard, screen reader, cognitive, language
32
Is the experience legally and ethically accessible?
Automated scan (axe-core, WAVE, Pa11y) + manual review. Violation report, severity-ranked. Legal risk framing.
33
Can keyboard-only users complete every primary task?
Tab order, focus visibility, modal trap, skip links, shortcut conflicts. Per-surface fix list.
34
Does it work for VoiceOver / NVDA / JAWS?
Alt text, ARIA, heading order, form labels, dynamic content announcements, live regions. Optional audio recordings.
35
Usable for ADHD, dyslexia, low literacy, ESL?
Reading level, sentence length, jargon density, time-pressured flows, recall vs. recognition patterns.
36
Does language welcome the audience, not exclude?
Ableist language, gender assumptions, idioms that don't translate, "easy"/"obvious" condescensions. Before/after rewrites.
Tier 6
Technical Supporting Layer
Performance, technical SEO, analytics implementation
37
Does the site load fast enough?
Lighthouse, WebPageTest, PageSpeed Insights. LCP, CLS, INP, TTFB. Prioritized fix list with engineering handoff.
38
Can search engines crawl, index, rank correctly?
Meta tags, structured data, sitemap, robots.txt, canonicals, redirects, internal linking, image alts. Paste-ready fixes.
39
Are events firing correctly?
Trace key user actions through analytics layer. Event-by-event verification, gaps and duplicates, recommended fixes.
How every audit gets framed, delivered, and followed up.
01
Pick by buyer need, not by completeness. Running every audit signals junior. Picking the right 4–6 signals senior.
02
Deliverable beats process. Clients pay for the document. Deliver fast. Offer follow-on work.
03
Frame as gift, not critique. Every finding is "here's what I'd brief into the next phase," not "here's what your team got wrong."
04
Always name what wasn't audited. Scope boundaries signal senior. Protects against "you missed X" pushback.
05
End with "want to go deeper?" Every audit starts the conversation, doesn't end it.
Native app audits. B2B SaaS dashboard audits. E-commerce / checkout audits. Email and transactional UX. Each of those becomes its own companion menu when there's a real client to test it on. If your engagement spans one of these, we'll scope it together.
20-minute scoping call. No commitment. We pick the audits that fit your situation and put a fixed fee on the engagement.