By Mehdi HassanMost explanations of a UX audit stop at a definition and a list of heuristics. That's not what you're buying. You're buying a document your team can turn into tickets on Monday. This hub explains what a UX audit is, how one runs, and exactly what the deliverable should contain, with an example structure of severity-ranked findings, each paired with a copy fix and a UI fix. Each section links to a deeper guide.
What is a UX audit?
A UX audit is a structured review of a digital product, such as a web app, SaaS product or marketing site, against usability principles, the product's own goals and whatever behavioral data exists. An experienced reviewer walks through key journeys, records where users are likely to get stuck or confused, rates how serious each problem is, and recommends fixes.
It's also called a UX review, a usability audit or a product UX audit. The names overlap; what changes is the depth. A light review might be one person checking a signup flow against a checklist. A deep audit combines a heuristic evaluation using Nielsen's 10 heuristics, analytics, session replays and stakeholder interviews across several journeys.
A UX audit is not the same as user testing. An audit predicts problems based on expertise and evidence. Testing observes real people hitting them. Audits are faster and cheaper; tests are more convincing. We compare the two in detail in UX audit vs usability testing.
What does a UX audit include?
A UX audit includes five parts: a scope, a review method, an evidence base, a severity-ranked findings list and recommended fixes. If any of the five is missing, you get either an opinion piece or a pile of screenshots that nobody acts on.
Part | What it covers | What good looks like |
|---|---|---|
Scope | Which journeys, devices and user types | Two or three named journeys with a clear goal each |
Method | How the review is done | Heuristics plus a journey checklist, stated up front |
Evidence | What supports each finding | Analytics, replays, support tickets, not just opinion |
Findings | The problems found | Severity-ranked, one problem per finding, with location |
Fixes | What to change | Specific copy and UI changes, sized by effort |
Summary | Where to start | A short list of the highest-impact fixes |
The scope matters most. 'Audit the app' produces a long, shallow list. 'Audit signup through first project created, on desktop, for new trial users' produces findings deep enough to fix. If you want the full list of what to check inside each journey, our UX audit checklist with 60 checks is organized by journey for exactly this reason.
What does a good UX audit deliverable look like?
A good UX audit deliverable is a ranked list of findings where each row names the problem, where it happens, why it matters, how severe it is, and the specific copy and UI change that fixes it. It reads like a backlog, not a critique.
Here's the structure we recommend. Every finding gets the same fields, so the list can be sorted, filtered and pasted into a ticket tracker.
Field | Purpose | Example |
|---|---|---|
ID and location | Find it fast | F-07, Billing settings, plan change modal |
Problem | One sentence, user's view | Users can't tell if downgrading deletes data |
Evidence | Why we believe it | Heuristic: visibility of system status; replays show hesitation |
Severity | How bad it is | High: blocks a task and risks churn |
Copy fix | Exact words to change | 'Your projects stay. You'll lose access to exports.' |
UI fix | Interface change | Add a consequences list above the confirm button |
Effort | Rough size | Small: copy and one component |
The copy fix column is where most audits fall short. A large share of usability problems are word problems: unclear labels, vague error messages, buttons that don't say what happens next. Writing the replacement text into the audit removes a round of back-and-forth. Our UX audit report template gives you this format ready to copy.
For example, suppose: three findings from a hypothetical SaaS signup audit
This is a hypothetical example to show the format, not a real client.
F-01, High. Signup form, password field. Problem: password rules appear only after a failed submit. Copy fix: show 'At least 10 characters, one number' under the field before typing. UI fix: live checklist that ticks as rules are met. Effort: small.
F-02, Medium. Onboarding step 2. Problem: 'Workspace URL' field with no explanation, so people stall. Copy fix: 'This is the link your team will use, like acme.yourapp.com. You can change it later.' UI fix: prefill from company name. Effort: small.
F-03, Low. Empty dashboard. Problem: blank chart area with no next step. Copy fix: 'Connect a data source to see your first report.' UI fix: one primary button to connect. Effort: medium.
How does a UX audit work step by step?
A UX audit works in six steps: agree the goals and scope, gather existing data, walk the journeys, log findings, rank them, and present fixes. The walk-through is where reviewers spend the most time, but the ranking and fixes are what make the audit useful.
Agree goals and scope
Name the business goal (more activated trials, fewer support tickets), the journeys that serve it, the user types and the devices. Write it down; it's your filter for every finding.
Gather existing evidence
Pull funnel data, session replays, support tickets, sales objections and any past research. Our guide to product analytics for UX covers the events and replays worth setting up first.
Walk each journey as a user
Complete every task in scope with a fresh account, on each device. Note friction, confusion, errors and dead ends as you go, with screenshots and the exact screen.
Check against heuristics and the checklist
Review each screen against usability heuristics, accessibility basics and a journey checklist. This catches issues a walk-through misses, such as missing error recovery.
Rate severity and effort
Score each finding on impact, frequency and persistence, then estimate effort. Merge duplicates so one root cause is one finding.
Write fixes and present
Write the copy and UI fix for each finding, group the top fixes into a short action plan, and walk the team through it live so questions get answered once.
How are UX audit findings ranked by severity?
UX audit findings are ranked by severity based on three factors: how much the problem hurts the user's task, how many users hit it, and whether they hit it once or every time. A common scale has four levels, from cosmetic to critical, and it's adapted from the severity ratings Nielsen Norman Group has published for years.
Severity | Meaning | Typical example | When to fix |
|---|---|---|---|
Critical | Blocks a core task or causes data loss | Can't complete checkout on mobile | Now |
High | Major delay or frequent abandonment | Unclear pricing step in signup | This sprint |
Medium | Slows users or causes errors they recover from | Vague form error messages | Next few sprints |
Low | Cosmetic or rare | Inconsistent icon styles | Backlog or design system work |
Pair severity with effort and you get a simple priority matrix: high severity and small effort first, then high severity and large effort, then the quick cosmetic wins. Low-severity issues that repeat across many screens often point to UX debt in the design system, which is better fixed once at the component level than screen by screen.
When does a product need a UX audit?
A product needs a UX audit when a key metric stalls and nobody can say why, before a redesign, after fast growth has added features on top of features, or before you raise money or launch to a new segment. It's also useful when you're about to add something big, like AI features.
- Activation or conversion is flat and you've run out of obvious fixes.
- Support tickets repeat the same 'how do I…' questions.
- You're planning a redesign and want to know what's actually broken before changing everything.
- The product grew feature by feature and screens no longer feel like one product.
- You're adding AI features. Our UX readiness checklist for adding AI to a SaaS product is a focused audit for that case.
A UX audit is usually not the first move for a product with no users yet; there's no behavior to analyze and too little to review. In that case, research and testing come first. A UX audit is also different from an SEO or technical site audit. If you're not sure which one you need, see website audit vs SEO audit vs UX audit.
How much does a UX audit cost?
UX audit prices quoted in 2026 typically run from about $1,500 to $25,000, based on the ranges agencies publish, and they vary widely. The spread comes from scope (one journey or the whole product), depth (checklist review or a full study with analytics and interviews), and whether you get findings only or redesigned screens too.
- Typical UX audit price range quoted by agencies
- $1.5k–$25k
- Designify brand and design sprint, fixed price
- $3,000
- Website or product explorations in the sprint
- 10
The cheapest audits tend to be findings-only reports. The most expensive include research with users. In between is the option most startups need: an expert review that also produces redesigned screens for the top problems. We break down each tier in our UX audit cost guide, and if you're comparing it to conversion work on marketing pages, see what a CRO audit costs.
How do you choose a UX audit service?
Choose a UX audit service by looking at a sample deliverable, not a sales deck. Ask to see how a finding is written: if it has a severity, evidence and a specific fix, the team knows what it's doing. If it's an annotated screenshot saying 'this feels cluttered', keep looking.
Other signals worth checking: whether they ask about your goals and data before quoting, whether they scope journeys rather than counting screens, and whether anyone on the team can design or build the fixes. Auditors who've never shipped product tend to recommend changes that are expensive to build. Our guide to UX audit services and what to expect lists the red flags, and if you're also picking a longer-term partner, see how to pick an AI product design agency for AI-heavy products.
What should you read next about UX audits?
Every guide in this cluster goes deeper on one part of the audit. Start with the one that matches where you are.
If you want to… | Read |
|---|---|
Run the review yourself | |
Apply a proven method | |
Present findings that get fixed | |
Find friction in data first | |
Decide between audit and testing | |
Budget for one | |
Hire someone | |
Fix the systemic issues |
If your product includes AI actions, two more are worth reading before an audit: AI product UX and designing features users trust and human-in-the-loop design for approvals and undo. AI flows fail in ways a standard heuristic pass can miss, such as unclear confidence or no way to reverse an action.
