UX Research
How Many Users Do You Need for a Usability Test? The Numbers Behind the Five-User Rule
The âfive users find 85% of usability problemsâ line is a rule of thumb, not an iron law. It is a specific consequence of a specific model in a specific scenario, not a universal law of usability. The Nielsen-Landauer modelâĶ
New on the desk
WCAG 2.2 vs 2.1: Every New Success Criterion, in Plain English
Web Content Accessibility Guidelines (WCAG) 2.2 arrived with only nine new requirements compared to its predecessor, but many of them bringâĶ
How to write error messages that help: A UX checklist, with 'Before' and 'After'
Error messages are one of the most important parts of user experience design. When an error message appears, the user is already confused orâĶ
Mega Menu or Nested Navigation: What Your Site Size Actually Calls For
The size of a site moves mega menus and dropdown menus into consideration as more top-level entries are added. One exposes several layers of linksâĶ
The System Usability Scale: How to Score It and What the Number Means
The system usability scale assigns a one number score from 0 to 100 to the overall usability of a system as subjectively rated by its users. HoweverâĶ
The 2003–2006 archive is still here. 421 index entries on usability, accessibility and information architecture, 138 pages on the people who wrote about them, and the original listings — all at their original addresses. Open the archive.
UX Research
All in UX Research →- How Many Users Do You Need for a Usability Test? The Numbers Behind the Five-User RuleUX Research
The âfive users find 85% of usability problemsâ line is a rule of thumb, not an iron law. It is a specific consequence of a specific model in aâĶ
- Moderated vs Unmoderated Usability Testing: A Decision Table That Ends the ArgumentUX Research
Usability testing comes in two primary flavors: moderated and unmoderated. And yet most usability teams don't seem to know which one they're reallyâĶ
- How Much Does Usability Testing Cost: A Detailed Budget BreakdownUX Research
When planning a usability test budget, itâs tempting to start with a quoted price from a vendor or an agency. But those âfromâ prices are composites.âĶ
Accessibility
All in Accessibility →- WCAG 2.2 vs 2.1: Every New Success Criterion, in Plain EnglishAccessibility
Web Content Accessibility Guidelines (WCAG) 2.2 arrived with only nine new requirements compared to its predecessor, but many of them bringâĶ
- Does the European Accessibility Act Apply to My Business?Accessibility
The European Accessibility Act, which entered into force in June 2019, sets out accessibility requirements for certain products and services in theâĶ
- How Much Does an Accessibility Audit Cost?Accessibility
If you're comparing accessibility audit quotes and wondering why they differ by an order of magnitude, that's because an "audit" is not a singleâĶ
Interface
All in Interface →- How to write error messages that help: A UX checklist, with 'Before' and 'After'Interface
Error messages are one of the most important parts of user experience design. When an error message appears, the user is already confused orâĶ
- How Many Fields Should a Form Have?Interface
The idea that fewer form fields mean better conversion sounds like it should be a rule. And there is strong foundation for the rule. Yet, it's farâĶ
- When to Use a Modal DialogInterface
When it comes to user interfaces, modal dialogs are a common tool for designers and developers. They are inert windows that temporarily block userâĶ
Architecture
All in Architecture →- Mega Menu or Nested Navigation: What Your Site Size Actually Calls ForArchitecture
The size of a site moves mega menus and dropdown menus into consideration as more top-level entries are added. One exposes several layers of linksâĶ
- When to Use Breadcrumbs NavigationArchitecture
Breadcrumbs are a secondary navigation tool, helping users understand their location within a complex site hierarchy. They are not decorative; toâĶ
- Why Is My Website Search So Bad: How to Diagnose and Fix ItArchitecture
If your site search is failing users, the ones pulling you into the diagnostic pit will probably be zero-result queries and searches that do showâĶ
Metrics
All in Metrics →- The System Usability Scale: How to Score It and What the Number MeansMetrics
The system usability scale assigns a one number score from 0 to 100 to the overall usability of a system as subjectively rated by its users. HoweverâĶ
- Task Success, Time on Task, and Error Rate: Which Answer Which QuestionMetrics
Task success, time on task, and error rate are the classic trio of usability metrics. But each measures something different. Task success rate showsâĶ
- How Long Should an A/B Test Run? Sample Size Without the Statistics DegreeMetrics
The sample size for an A/B test comes from four inputs: the baseline conversion rate, the minimum detectable effect, the desired significance levelâĶ
AI and Users
All in AI and Users →- Chatbot vs. Site Search for Support: The "Right" Tool Depends on the TaskAI and Users
When it comes to providing customer support, the age-old debate between using a chatbot and a site search engine continues to rage. But the truth isâĶ
- How to Explain an Automated Decision to Users: Patterns That Survive a ComplaintAI and Users
Explaining an automated decision to the person it affects is a crucial user experience task, and one with real legal and regulatory requirements.
- Synthetic Users vs. Real Participants: What AI Research Tools Can and Cannot ReplaceAI and Users
Synthetic users promise to generate text feedback indistinguishable from real research participants. However, this technology cannot substitute forâĶ
Industry references
Sites and standards bodies this desk links to, and partners who support the publication.
