One-day UX Audit: A Checklist That Finds Critical Problems to Users

Most UX problems don’t need to be looked for with expensive research and weeks of testing. You can see them with the naked eye if you know where to look.

A one-day UX audit is not a compromise. This is a specific technique: a structured product inspection based on proven criteria that identifies 80% of critical issues without recruiting users.

The remaining 20 percent -- nuances, contextual issues, unexpected scenarios -- yes, we need real tests. But a day's audit is enough to prioritize, justify a redesign, and not spend your budget researching what's already obvious.

## Preparation: 30 minutes before the audit

An audit without preparation is just a walk through the interface with the feeling of “something is wrong.” Preparation turns it into a diagnosis.

### Define the scope

What exactly are you checking? The whole product is a bad idea. In one day, you can really check the quality:

• One key flow (onboarding, checkout, activation)
• One section of the product
• Specific user scenario from start to finish

If the product is large, choose the float with the greatest losses according to analytics. No data - start with onboarding. There's always something there.

### Gather context

Before opening the interface, record:

• Who are the main users (experienced / beginners, age, devices)
• What is the main task of the user in this flow
• What data about problems is already available (complaints in support, drop-offs in analytics)

This protects against “designer audits” – when you find problems that annoy you, not those that bother the user.

### Prepare the tools

Minimum set:

• Devices: computer + smartphone (if mobile is important)
• How to fix: Notion, FigJam, or just Google Doc
• Screenshots: Any screen capture tool with annotations

The optimal set: add a test account of the beginner (not logged in advance) to pass the float as the first user.

## Block 1: First screen and orientation – 45 minutes

The first thing a user sees determines everything else. This is the highest cost of error.

### Rule of 5 seconds

Open the main page/first screen and record 5 seconds. Then close it. Answer three questions:

• What is this product?
• For who?
• What should we do next?

If one answer isn’t obvious, it’s the first screen problem. It’s not subjective, it’s a failed test that a real user will pass.

### First screen checklist

• The title explains what the product does (not the next generation platform, but what it does)
• The target audience is visible (“for teams”, “for freelancers”, “for e-commerce”)
• One major CTA, not three competing
• No autoplay video with sound
• No full-screen poppies in the first 5 seconds
• Download no longer than 3 seconds (check DevTools → Network → Slow 3G)
• On the mobile CTA is visible without scrolling

### Navigation and orientation

• The user understands where he is (breadcrumbs, active menu item)
• It is clear how to go back or to the main
• Logo clickable and leads to the main
• Mobile menu opens and closes without problems
• Search is where you need it (if the product has a lot of content)

## Block 2: Key Flow - 2 hours

That's the basis of the audit. Go through the main user script from start to finish – as if you’re doing it for the first time.

### How to pass the float

The important thing is, don’t help yourself. If something is unclear, fix it, but do not look for a solution yourself. It is confusion and uncertainty that is the problem.

At every step, ask:

• Where should I look now?
• What am I being asked to do?
• Do you know what will happen if I press?
• Is there a fear of being wrong?

### Onboarding checklist

• Registration takes less than 2 minutes
• The first useful action is possible immediately after registration
• Empty State Explains What to Do (Not Just “You Have No Data”)
• The progress of the onboarding is visible (the user knows how much more is left)
• Each step has a clear goal (not step 3 of 7, but add the first project)
• There is an opportunity to skip optional steps
• After the onboarding is completed, it is clear what to do next

### Form checklist

The form is the concentration of UX pain. Check each shape in the product:

• Only mandatory fields (each extra field reduces conversion)
• Labels over fields (not a placeholder instead of a label - it disappears when entered)
• Keyboard type corresponds to the field (email → email keyboard, phone → digital)
• Errors appear near the field, not at the top of the form
• The error message explains what to do (not “incorrect format”, but “enter email in name@domain.com format”)
• The submit button describes the result ("Create an account", not "Next")
• The form is not reset in case of error – the data is saved

### List of key actions

• The main action is available for 1-2 clicks from any screen
• The button clearly stands out as the main action (not lost among others)
• After the action, there is a clear feedback (something has changed, confirmation has appeared)
• Destructive actions (remove, cancel) require confirmation
• Irreversible actions are clearly marked as irreversible
• It is possible to cancel an accidental action (undo or basket)

## Block 3: Readability and texts - 45 minutes

UX problems often live not in structure but in words. Bad texts break even a good interface.

### Readability checklist

• The main text size is at least 16px (on mobile)
• Text contrast corresponds to WCAG AA (minimum 4.5:1 for plain text)
• Line no longer than 65-75 characters (otherwise hard to read)
• Interline interval not less than 1.4
• Text is not centered in long blocks (only in headings and short phrases)

### Checklist of UX texts

• Section headings describe content rather than decorate
• CTA buttons describe the action and the result (verb + object: “Download report”, “Add participant”)
• Error messages do not blame the user (“Incorrect password” instead of “You entered the wrong password”)
• Empty states explain what to do, not just report lack of data
• There is no technical jargon without explanation (unless the audience is technical)
• Confirming messages are specific (“Project Save” not “Successful”)

## Block 4: Mobile and cross-platform – 45 minutes

If a significant part of the audience on mobile is a separate full-fledged audit. Mobile isn’t a “smaller desktop”: it has different interaction patterns and other problems.

### Mobile UX checklist

• All buttons and links at least 44×44px (minimum tappable size)
• No items that require hover (no mouse on the phone)
• Forms do not cause zoom when focusing (field font size ≥ 16px)
• Content does not go beyond the screen horizontally
• Poppas shut down and don't block everything on the small screen
• Swipe gestures work where expected (galleries, cards)
• Bottom navigation is achievable with your thumb

### Cross-browser checklist (fast)

• Check in Chrome and Safari (they give different results)
• Check in incognito mode (without cache and extensions)
• Check for slow connection (DevTools → Network → Slow 3G)

## Block 5: Errors and extreme cases - 30 minutes

Most designers check the happy path when everything goes according to plan. Real users live in edge cases.

### How to check

Do the “wrong” things intentionally:

• Enter incorrect data into forms
• Click "Back" in the browser in the middle of the flow
• Interrupt halfway
• Open deep links without authorization
• Try empty states (0 projects, 0 notifications, 0 search results)

### List of states

• 404 page clear and offers a way back
• Download state is everywhere there is waiting (not an empty screen)
• A network error is handled (not a white screen, but a clear message)
• “0 Elements” Search Results Explain Why and What to Do
• Very long content does not break the layout
• Empty input fields show placeholder or hint
• The interrupted download is restored or clearly reports an error

## How to document: a system of fixing problems

Finding problems should be fixed so that you can work with them later.

### Structure of one problem

Problem: [One sentence is what doesn't work]
Where: [screen, step flow, specific element]
Severity: Critical / Major / Minor
Why this is a problem: [the impact on the user]
Screenshot: [attached]
Possible solution: [optional, if obvious]

### Classification of certification

*Critical - Blocks the execution of a key task. The user can not register, pay, create the main object. Correct immediately.

Major – significantly complicates the task or creates a high risk of error. The user copes, but with difficulty or accidentally does the wrong thing. Correct in the nearest sprint.

**Minor - annoys, reduces the quality of the experience, but does not block. Correct if possible.

## Final report: from list of problems to plan

At the end of the day, you have a list of problems. The next step is to turn it into a plan.

### Prioritization matrix

For each problem, evaluate:

• Frequency: How often do users experience it (units/part/most)
• Impact: How profoundly impacts experience (Minor/Major/Critical)
• Effort: How much it takes to fix (hours/days/weeks)

Start with High frequency + High impact + Low effort – quick wins with maximum effect.

### What to include in the report

1. Executive summary - 3-5 sentences: what was checked, main findings, recommendations
2. Top 5 Critical Problems - With Screenshots and Explained Influence
3. Complete list - grouped by severity
4. Recommended next steps Prioritized plan

## AI and UX Audit: How Claude Speeds Up Diagnostics

AI does not replace the passage of the interface with your own hands – but helps in the stages of preparation, analysis and documentation.

### Prompt: prepare a checklist for a specific product

I conduct a UX audit [product type: SaaS/e-commerce/mobile app/marketplace].

Main audience: [Description]
The main flow that I check: [description]
Known problems or complaints: [if any]

Create a specialized checklist for this audit:
- Add criteria specific to this type of product and audience
- Prioritize What to Check First
- Indicate where critical problems occur most often in this type of product

### Prompt: Get a second look at the screenshot

Download the screenshot in Claude:

Here is a screenshot of our product.

Targeted user action on this screen: [description]
Target audience: [Description]

Perform an express audit:
1. What is the first thing that catches your eye – and is this the right priority?
2. Are there elements that can confuse or create false expectations?
3. What prevents the user from performing a targeted action?
4. Evaluate the texts: title, CTA, captions - are they clear?

### Prompt: Prioritize the list of problems

Here is a list of UX issues found in the audit:

[problem list]

Prioritize them by impact × effort.

For each problem:
- Evaluate the impact on the user (High / Medium / Low)
- Estimate the approximate correction effort (Low: hours / Medium: days / High: weeks)
- Ultimate priority

Highlight the top 5 "quick wins" - high impact, low effort.
Highlight the top 3 "strategic" - high influence, high effort, it is worth planning.

### Prompt: Write an executive summary of the audit

I did a UX audit of the product [name/description].

The main problems identified are:
[List]

Write an executive summary on a half page for guidance:
- What was checked and how
- Main conclusion (1 sentence)
Top 3 Critical Business Impact Problems
Recommendation: What to do first

Tone: Concrete, without UX jargon, with an emphasis on impact on users and businesses.

## Block 6: Availability - 30 minutes

Accessibility is not just about people with disabilities. This is about the quality of the interface in principle. A product with good availability is usually better for everyone.

### Quick check without special tools

• Try to pass the key flow with only the keyboard (Tab for transition, Enter for confirmation). Does it work?
• Browser zoom to 200% - nothing broke or left?
• Turn off CSS (in DevTools → ... → More tools → Rendering → Emulate CSS media feature prefers-color-scheme) and look at the structure. Is it logical?

### Availability checklist (basic)

• All images have alt text (except decorative)
• Text contrast corresponds to WCAG AA (4.5:1) – check with WebAIM Contrast Checker
• Focus states visible (not removed via outline: none)
• Interactive elements reachable from the keyboard
• Forms have correct labels (not just placeholder)
• Modal windows capture the focus and return it when closed

### Quick verification tools

• axe DevTools (browser extension) – automatically detects WCAG violations
• Lighthouse (Chrome DevTools → Lighthouse → Accessibility) – Scores 0 to 100
• WAVE (wave.webaim.org) - Visual Accessibility Inspector

The purpose of a quick audit is not a perfect score, but critical blockers. Check out Lighthouse: If a score is below 70, there are serious problems.

## Block 7: Performance as UX – 20 minutes

Speed is design. A slow product is a bad UX no matter how beautiful the screens are.

### What to check

Run Google PageSpeed Insights (pagespeed.web.dev) for product homepages. Look at this:

• **LCP (Largest Contentful Paint) ** When the user sees the main content. Normal: < 2.5 seconds
• FID/INP (Interaction to Next Paint) - How fast does the interface react? Normal: < 200ms
• CLS (Cumulative Layout Shift) - whether items jump when booted. Normal: < 0. 1st

If any of them in the red zone is a UX problem, not just a technical one.

### Performance checklist

• LCP < 2.5 sec on mobile (most critical)
• No CLS – Content doesn’t jump when images load
• Heavy images optimized (use WebP, correct dimensions)
• Slow 3G download in DevTools is acceptable
• Skeletons or placeholders are shown while the content is loaded

## Typical audit findings by product type

Different types of products have different “typical” problems. Knowing them, you check them first.

### SaaS/B2B

** Most common problems:**

• Empty screens without direction after registration
• Overloaded dashboards with 20+ metrics without hierarchy
• Requirement to fill out profile before accessing functionality
• The Unobvious Difference Between Paid and Free Features

### E-commerce

** Most common problems:**

• Slow loading of catalog pages and product cards
• Complex or multi-step checkout
• Lack of “guest” options when placing an order
• Mobile shopping cart works differently than desktop

### Mobile applications

** Most common problems:**

• Too small tap targets (especially in lists)
• No deep linking – the notification leads to the main screen, not to specific content
• Request permissions without explaining why (geolocation, camera, notifications)
• Bottom Sheet is not closed by swipe down

### Marketplaces

** Most common problems:**

• Search returns irrelevant results without filters
• Search results page doesn’t remember filters when returning
• No comparison of several products/offers
• The placement process (for the seller) is too complicated

## After the audit: how not to lose the findings

Audit done, list of problems. The most common further story: the document is opened once, then it is lost, most problems are not fixed.

### Discovery system

Task tracker, not document. Each problem found is a separate problem in Jira/Linear/Notion. With description, screenshot, severity and effort evaluation.

Results meeting. 30-45 minutes with product and team. Pass the top 10 problems, negotiate priorities, fix solutions.

** Quick wins separately.** Select 5-10 tasks that can be closed in a day without a developer (text, color, order of elements). Close them right away – it gives momentum and shows the value of the audit.

**Review through the block. ** What's been corrected? No what? Why? Are there new problems? This turns an audit from a one-off event into a regular practice.