UX Audit
✓ a prioritized list, not a 40-slide deck
On this page
- TL;DR
- What a UX audit is, and what it is not
- When to run a UX audit
- Step 1: Define the scope and success criteria
- Step 2: Gather existing evidence
- Step 3: Walk through the experience as the user
- Step 4: Evaluate against explicit criteria
- Step 5: Rate severity
- Step 6: Report findings so people can use them
- What a good UX audit finding looks like
- Validate the findings that matter
- Where AI can help with a UX audit
- Common UX audit mistakes
- UX audit checklist
- Turn the audit into a research plan
TL;DR
A UX audit is a structured review of a product or experience designed to identify what may be hurting the user experience and prioritize the problems worth addressing first.
A useful audit goes beyond an expert clicking through a product and listing everything that feels wrong. It defines a clear scope, evaluates the experience against explicit criteria, brings in existing customer and product evidence, rates findings by severity, and identifies which assumptions still need to be validated with users.
The result should not be a 40-slide deck that disappears after the readout. It should give your team a defensible, prioritized list of what to investigate or fix next.
What a UX audit is, and what it is not
A UX audit is a systematic evaluation of a product, website, or specific user flow against usability principles, existing evidence about user behavior, and the goals the experience is supposed to support.
Think of it as an evidence-informed expert review.
The auditor works through the experience looking for potential usability problems, but the strongest audits do not rely on expert judgment alone. They also draw from evidence the organization already has, such as analytics, support tickets, previous user research, customer interviews, and usability studies.
A UX audit is related to usability testing and heuristic evaluation, but the three are not interchangeable.
- A UX audit combines expert evaluation with available product and customer evidence to identify and prioritize potential UX problems.
- Usability testing puts the product in front of real users and observes what happens as they attempt specific tasks. It provides direct evidence about where people actually struggle. Our usability testing guide covers the method in more detail.
- Heuristic evaluation reviews an interface against an established set of usability principles. It can be one part of a broader UX audit, but an audit can also incorporate analytics, customer feedback, accessibility considerations, existing research, and product goals.
That distinction matters because an audit generates informed hypotheses about the experience. It does not automatically prove that users will experience every issue the auditor identifies.
When to run a UX audit
A UX audit is particularly useful when you need to understand an experience before deciding what research, design, or product work should happen next.
- Before a redesign. Audit the current experience before replacing it. You want to understand not only what appears to be broken, but also what is already working and should be protected.
- When a product metric changes unexpectedly. If conversion, activation, retention, or task completion has moved, an audit can help identify plausible UX explanations that you can investigate further.
- When you inherit a product. Reviewing a product systematically can help a new product, design, or research team understand its flows, inconsistencies, assumptions, and known customer pain points.
- Before usability testing. An audit can help you identify the areas most worth putting in front of users instead of trying to test an entire product at once.
- After significant product changes. New features and incremental design decisions can create inconsistencies over time. Periodic audits can surface that accumulated friction.
Poor reasons to run one
If the redesign has already been approved and the audit exists only to justify it, you are not really auditing anything. And if nobody owns what happens after the findings are delivered, the audit is likely to become documentation rather than a catalyst for improvement.
Step 1: Define the scope and success criteria
One of the easiest ways to make a UX audit unmanageable is to audit “the product.”
A large product can contain dozens of journeys, user types, devices, and states. Reviewing all of them at once produces an enormous list of findings with very little indication of where the team should start.
Scope the audit around a particular experience and audience instead. For example:
- New administrator completing first-time setup on desktop
- Existing customer renewing a subscription on mobile
- Prospective customer comparing plans on the website
- Team member creating and sharing their first report
Then define what the person is actually trying to accomplish. “Complete onboarding” is the product's description of the task. “Get my team set up so we can start using the product this week” is closer to the customer's goal.
Before starting the audit, answer three questions:
- Whose experience are we evaluating? Define the relevant audience or segment.
- What are they trying to accomplish? Describe the goal in the user's terms.
- What does success look like? Define successful outcomes for both the user and the business.
These criteria give you something more meaningful to evaluate than whether an interface simply looks good.
Step 2: Gather existing evidence
Before you start clicking through the product, find out what your organization already knows.
This helps reduce the risk of forming an opinion about the experience first and then unconsciously looking for evidence that confirms it.
Depending on the product and flow, useful evidence may include:
Analytics
Look for abandonment, repeated actions, unusual paths, or other behavioral patterns that suggest a place worth investigating. Analytics can show you where something unusual is happening, although you will usually need additional research to understand why.
Support tickets
Search for terminology associated with the flow and look for recurring questions or complaints. Pay attention to both how often an issue appears and how customers describe it in their own words.
Sales, renewal, and churn conversations
These conversations can reveal expectations, objections, and problems that may never make it into a support ticket.
Existing research
Search previous interviews, usability studies, surveys, and other research related to the experience. A searchable research repository makes this much easier than hunting through old decks and individual documents. If your research is difficult to find or reuse, our UX research repository guide covers how to improve that infrastructure.
Session recordings
If your team collects them, review multiple relevant sessions rather than relying on one memorable example. Individual recordings can generate useful hypotheses, but patterns across sessions are more informative.
The goal is to build an evidence base before making judgments. Instead of writing that the export process is confusing, you may eventually be able to write that customers repeatedly contact support about the export process, and product analytics show increased abandonment at the same step.
Now you have a hypothesis supported by more than an expert impression.
Step 3: Walk through the experience as the user
Once you understand the existing evidence, go through the experience yourself.
Use the devices and conditions relevant to the audience you defined in step one. If you are evaluating onboarding or another first-use experience, start with a fresh account whenever possible.
Familiarity is one of the challenges of auditing your own product. Once you know where everything lives and what internal terminology means, it becomes difficult to experience the interface as someone encountering it for the first time. Document:
- Each step in the flow
- Important decision points
- Places where the next action is unclear
- Unexpected system behavior
- Errors and recovery paths
- Terminology that requires interpretation
- Inconsistencies between screens
- Accessibility concerns
- Moments where you have to stop and think
Screenshots with concise annotations are especially useful because they let stakeholders see the issue in context.
Document what works, too. A UX audit should identify elements worth preserving, not simply produce a list of defects. If customers already understand one part of the experience particularly well, the team should know that before redesigning it.
Step 4: Evaluate against explicit criteria
A useful UX finding needs a standard behind it. Without one, the audit can quickly turn into a debate about personal preference.
Choose the criteria that fit the experience you are evaluating. These might include:
- Established usability heuristics. Nielsen Norman Group's usability heuristics are a common framework for evaluating issues such as system feedback, error prevention, consistency, user control, and recognition rather than recall.
- Accessibility standards. Evaluate relevant accessibility requirements, including the Web Content Accessibility Guidelines, while recognizing that a UX audit is not necessarily a substitute for a formal accessibility audit.
- Your design system. Inconsistent components, interactions, terminology, and patterns can introduce unnecessary friction.
- Task-success criteria. Return to the goals you defined in step one. Does the experience make it reasonably clear how the user can accomplish the task?
Depending on the product, you may also want to evaluate:
- Findability
- Clarity of language
- Error prevention and recovery
- Feedback and system status
- Consistency
- Accessibility
- Mobile experience
- Performance
The criteria should reflect the product and research question rather than becoming a generic checklist applied identically to every experience.
Step 5: Rate severity
Finding problems is only half the job.
If your final report contains 40 findings with no indication of which ones matter most, the team still has to do the prioritization work after the audit.
Rate findings using consistent criteria. Two useful dimensions are impact and frequency.
- Impact: what happens when someone encounters the problem? Does it prevent task completion, create substantial friction, introduce uncertainty, or primarily affect polish?
- Frequency: based on the evidence available, how often does the issue appear to occur and how much of the relevant audience may encounter it?
You can then use a simple severity scale:
- Critical: prevents completion of an important task or creates a significant barrier that requires immediate attention.
- Serious: creates substantial friction or failure, although users may have a workaround.
- Moderate: slows people down, creates uncertainty, or makes the experience unnecessarily difficult.
- Cosmetic: primarily affects consistency or polish without substantially interfering with task completion.
Do not let the labels create false precision. Your severity rating is still an assessment based on the evidence you have.
Where possible, pair severity with an effort estimate from design and engineering. That lets the team identify high-impact problems that may also be relatively straightforward to address.
Most importantly, keep frequency and importance separate. A problem that appears frequently is not automatically the highest priority. A less common issue may deserve immediate attention if it prevents an important audience from completing a critical task.
Step 6: Report findings so people can use them
The UX audit deliverable should make prioritization easier, not give stakeholders more information to sort through. A practical report can include:
- A one-page summary. Surface the handful of findings that matter most, their severity, and why they matter. Someone who reads only this section should understand the major issues.
- A prioritized findings list. Include the severity, supporting evidence, estimated effort where available, and recommendation for each issue.
- Detail for each major finding. Show the relevant screen or flow, explain what happens, identify the evidence or principle supporting the finding, and describe the consequence for the user.
- What is working. Explicitly identify parts of the experience the team should preserve.
- Open questions. Document what the audit could not determine and what additional research would answer it.
That last section is particularly important. A UX audit should not pretend to answer questions that require user research. Instead, it can help determine what research should happen next.
What a good UX audit finding looks like
Write findings as an observation and consequence rather than a verdict.
A weak finding might say that the save-state feedback is confusing. That leaves everyone debating what confusing means.
A stronger finding would say that the interface provides no visible confirmation after a user saves, making it difficult to tell whether the action succeeded.
If you have supporting evidence, make the finding stronger: the interface provides no visible confirmation after a user saves, and support tickets show that customers have reported duplicate submissions after repeating the action because they were unsure whether the first attempt succeeded.
Now the finding explains:
- What happens
- Why it matters
- What evidence supports it
Avoid adjectives such as confusing, cluttered, and unintuitive unless you immediately explain the observable behavior behind them.
The goal is not to make the finding sound severe. It is to make it specific enough that another person can examine the evidence and decide whether they agree.
Validate the findings that matter
A UX audit produces hypotheses about where an experience may be breaking down. That is useful, but expert judgment is not the same thing as observing real users.
Take high-severity or uncertain findings into additional research when the decision warrants it.
Usability testing can show whether people actually encounter the problems you identified and how those problems affect task completion.
If the issue involves navigation or information architecture, tree testing can test whether people can find information through the proposed structure, while first-click testing can help you understand whether people initially head in the expected direction.
For an unfamiliar workflow, ethnographic research or user interviews may reveal that the team's assumptions about the task itself need revisiting.
The point is not to validate every line of an audit. Use the audit to narrow the field, then spend research time on the findings where being wrong would materially affect the decision.
Where AI can help with a UX audit
AI can make parts of a UX audit considerably faster, especially when the evidence is spread across a large amount of qualitative material. Useful applications include:
- Finding patterns in existing research. Instead of manually opening every previous study, AI-assisted search and synthesis can help surface research related to the flow you are auditing.
- Analyzing open-ended feedback. AI can help group support tickets, survey responses, and other customer feedback into recurring themes.
- Checking consistency. It can help identify variations in terminology or patterns across a large experience that are easy to miss screen by screen.
- Creating a first pass. AI can help organize observations, summarize existing evidence, and suggest areas worth investigating.
The limits matter just as much. AI should not be treated as the final authority on severity or priority. Those decisions require product context, business priorities, knowledge of the affected audience, and an understanding of what the organization can realistically change.
It can also make a frequently mentioned problem look more important simply because there is more text about it.
Use AI to make the evidence easier to inspect, not to replace the judgment required to interpret it. Great Question's AI research capabilities are designed to keep AI-assisted analysis connected to the underlying research, so teams can move from a generated pattern back to the customer evidence behind it.
Common UX audit mistakes
- Auditing too much at once. A tightly scoped audit usually produces more actionable findings than trying to evaluate an entire product in one pass.
- Relying on opinion without evidence. Connect findings to a usability principle, customer evidence, product data, or ideally more than one source.
- Giving every finding equal weight. Without severity and prioritization, the team is likely to address whatever is easiest rather than whatever matters most.
- Skipping follow-up. Decide who owns the next step for important findings. Otherwise the audit becomes documentation rather than action.
- Ignoring what works. Tell the team what should be preserved alongside what needs improvement.
- Auditing only under ideal conditions. Consider the devices, screen sizes, connection speeds, accessibility needs, and environments relevant to the people actually using the product.
- Treating the audit as proof. An expert finding is evidence-informed judgment. When a high-stakes decision depends on whether the problem really affects users, validate it with research.
UX audit checklist
Before wrapping up your audit, check that you have:
- 01Defined the specific user or customer segment
- 02Defined the task or flow being audited
- 03Documented what successful completion looks like
- 04Reviewed relevant analytics
- 05Reviewed support and customer feedback
- 06Searched existing research
- 07Walked through the flow under realistic conditions
- 08Evaluated the experience against explicit criteria
- 09Documented what works as well as what does not
- 10Connected each major finding to evidence
- 11Rated findings by severity
- 12Added effort estimates where appropriate
- 13Identified the highest-priority findings
- 14Documented questions that require additional research
- 15Assigned owners or next steps for findings that will move forward
Turn the audit into a research plan
A UX audit is most valuable when it helps a team decide what to do next.
Some findings will be straightforward enough to fix. Others will expose assumptions that need to be tested. And some will point back to research your organization has already done but nobody knew existed.
That is why the infrastructure around an audit matters as much as the audit itself.
When customer interviews, usability studies, surveys, and other research live in a shared research repository, teams can start an audit with the evidence they already have and preserve the new findings for the next person who needs them.
From there, Great Question can support the research that follows, from recruiting participants through your research panel to running different UX research methods and bringing the results back into the same body of evidence.
Frequently asked questions
What is a UX audit?
A UX audit is a structured evaluation of a product, website, or user flow against usability principles, available customer and product evidence, and defined user goals. The result is a prioritized set of potential UX problems and recommendations for what to investigate or improve.
What is the difference between a UX audit and usability testing?
A UX audit is primarily an expert assessment informed by existing evidence. Usability testing observes real people attempting tasks with a product. An audit can identify likely problems and help determine what to investigate, while usability testing provides direct evidence about what happens when representative users interact with the experience.
How long does a UX audit take?
There is no universal timeline. The time required depends on the size of the flow, number of user types and devices involved, amount of existing evidence to review, and depth of the final deliverable. A tightly scoped flow can be audited much faster than an entire product.
What should a UX audit include?
A useful UX audit should define its scope and evaluation criteria, document supporting evidence, identify what is working as well as what may be causing problems, rate findings by severity, prioritize recommendations, and identify open questions that require additional research.
Do you need a UX audit if you already conduct usability testing?
They serve different purposes and can complement each other. Usability testing provides direct evidence about how participants interact with the flows you test. A UX audit can review a broader experience, incorporate evidence from several sources, and identify areas worth investigating in future research.
What is a UX audit checklist?
A UX audit checklist is a structured list of the evidence, evaluation criteria, and reporting steps to cover during an audit. It can include defining the user and task, reviewing existing research and analytics, evaluating the flow against usability and accessibility criteria, rating findings by severity, and documenting next steps.
