Common Heuristic Sets

Nielsen's ten usability heuristics are the most widely known: visibility of system status, match between system and real world, user control and freedom, consistency and standards, error prevention and recovery, recognition rather than recall, flexibility and efficiency, aesthetic and minimalist design, error messages that are helpful, and help and documentation. A common violation of visibility of system status appears in forms that submit silently, leaving users unsure whether a click registered or the page has simply frozen. Other frameworks focus on specific areas–accessibility heuristics emphasize perceivability and operability, mobile heuristics address touch-sized targets and screen size constraints. Choose heuristics suited to your product type rather than applying a generic list to everything. A booking widget, for instance, benefits more from checking date pickers against touch-target guidance than from desktop-oriented conventions alone.

Running an Evaluation

ux audit scope

Gather a small group of evaluators–typically three to five people–and have each independently inspect the interface against your heuristic set. Evaluators do not need to be usability experts, though some training helps consistency. Each person notes problems they observe, rating severity (minor inconvenience versus blocks core workflow). After individual passes, consolidate findings and discuss. Severity ratings help prioritise fixes. A single evaluator catches perhaps forty to fifty percent of problems; five evaluators catch roughly eighty to ninety percent. Adding more evaluators yields diminishing returns. A simple four-point scale–cosmetic, minor, major, catastrophic–keeps ratings consistent across evaluators and makes comparing notes during consolidation far easier than free-text descriptions alone.

Strengths and Limits

design audit

Heuristic evaluation works well for quick feedback on mature designs and for early-stage concepts where building a prototype for user testing is premature. It is faster and cheaper than recruiting participants. The risk is that evaluators sometimes spot false problems–issues they think should be problems but users do not actually experience–or miss real problems because they lack specific user context. An evaluator might flag a colour contrast choice that rarely troubles users with typical vision while overlooking a labelling problem that only surfaces once someone attempts an unfamiliar task. A feedback loop between heuristic evaluation and user testing balances strengths. Use evaluation to narrow possibilities, then test assumptions with real users. Documentation of findings matters; unsorted lists of problems create confusion. Link each issue to the specific heuristic it violates and describe the consequence for users.

ux audit scope

Structured Analysis

Select three to five key competitors and define analysis dimensions: core features, navigation structure, information hierarchy, visual design approaches, pricing strategies, error handling, onboarding flows. Document findings in a matrix or spreadsheet. Capture screenshots and note specific implementation choices–button labels, placement of key actions, how form validation works. Walk through key user journeys on each product. For a calendar app, that means creating an event, sharing it, and checking a group's schedule. Document where flows diverge. Some products require confirmation before deleting; others undo after deletion. Neither is objectively correct; each reflects a design philosophy. Competitor analysis highlights such choices and their consequences.

Identifying Patterns and Gaps

Look for patterns–approaches that multiple competitors share are likely established conventions users expect. If five project management tools place timelines on the left and task details on the right, deviating from that pattern requires strong justification. Conversely, gaps reveal opportunities. If no competitor offers real-time collaboration notifications, that could be an underserved need or a feature with problems you have not discovered. Analyse why competitors might have skipped it. Sometimes gaps exist because the feature is hard to do well; building it poorly wastes effort. Other gaps are genuine opportunities. Competitor analysis feeds design decisions but never determines them. Your product has different users, different business constraints, and different values than competitors. Use analysis to inform, not dictate.

Beyond Surface Features

Deep analysis goes beyond feature lists. Study competitor support documentation and community forums to see where users struggle. Review user reviews and ratings to identify consistent complaints. Check competitor job postings for product manager and designer roles–they signal what companies are investing in. Follow competitor blogs and public communications to understand their strategic priorities. This context helps you understand not just what competitors built, but why they built it and how successful those choices have been. Shallow feature comparison leads to feature bloat; strategic analysis leads to purposeful design decisions grounded in market understanding.

Key Discovery Activities

User research reveals actual problems and desires. Interviews with target users uncover frustrations with current solutions. Analytics reveal where people abandon products and what paths they take. Support tickets highlight problems users encounter using existing products. Competitive analysis shows how similar problems are currently solved. Stakeholder interviews surface business constraints and opportunities. Synthesis across these sources identifies patterns and validates problems worth solving. A team might discover that customers struggle with invoicing in their accounting software not because invoicing is hard to use but because they do not understand invoice numbers for tax purposes. The problem is not software design; it is education. Discovery surfaces that distinction.

ux audit scope

Narrowing and Prioritising

Most organisations have more problems to solve than resources to address them. Discovery helps prioritise. Which problems affect the most users? Which do users care about most? Which problems create business opportunity? A problem affecting a tiny segment may not be worth solving despite genuine severity. A problem affecting everyone but minor in impact may not warrant development investment. Prioritisation matrices balancing impact and effort help teams decide. Validate problem framing before advancing: if you solve this problem, will it actually change user behaviour or satisfaction? Solving non-problems wastes effort. Run small experiments or prototypes to test problem framing. A mockup showing a proposed solution often reveals whether users actually want what you are proposing.

a fuller account of accessibility audit scope

Rapid Testing and Learning

ux audit scope

Discovery is not indefinite exploration. Set time boundaries–two weeks, one month–for discovery before moving to build decisions. Use rapid prototyping to test ideas cheaply. Paper sketches, clickable wireframes, or crude prototypes reveal problems with approaches before engineering investment. Testing with users is fast and inexpensive at this stage. Five users reviewing a prototype reveal most major issues. Quick feedback loops beat extended speculation. Document discovery findings clearly so building teams understand problem context and constraints. Discovery that goes unshared wastes knowledge. Successful products often trace back to strong discovery work that validated both problem and solution direction. Discovery is not a phase you complete then leave behind; revisit discovery when product changes or new problems emerge. Markets shift, competitors innovate, and user needs evolve. Continuous discovery keeps products aligned with reality.

a template for briefing a design audit properly

Audit and Classification

Start by inventorying existing content and structure. Identify every major content type and category. Map current hierarchy. Spot redundancy–when the same content appears in multiple places–and orphans–pages nobody links to. Note whether page titles, labels, and category names are consistent and clear. Are they jargon-heavy or accessible? Document navigational patterns–how do users move between related content? Can they easily backtrack? Look for mental model mismatches: labelling a feature “Settings” when users call it “Preferences” creates confusion. Category labels should match language users employ. Audit reveals current state; it does not judge. A site with extensive archives naturally has deeper hierarchy than a small blog.

User Testing of Structure

Card sort exercises reveal how users naturally group related content. Participants sort cards representing content into categories and name those categories. Open sorts (participants create their own categories) reveal natural mental models. Closed sorts (participants assign content to predefined categories) test whether designed structure matches expected groupings. Tree testing presents a proposed structure and asks participants to locate items. A participant should find “return policy” in fewer than three clicks. If participants struggle, the structure does not reflect their expectations. Five to eight participants typically reveal whether structure works. Testing is quick and inexpensive but powerful for validating approach before investment.

a realistic look at what a screen inventory actually contains

Redesign and Implementation

Structure redesigns often feel disruptive but are usually worth the effort if current structure is genuinely confusing. Gradual transitions help–running old and new navigation alongside each other, using search to bridge gaps during transition, clear communication about changes. Navigation design is separate from IA but intertwined–clear IA becomes unusable if navigation is hidden or confusing. Documentation of structure helps future teams understand why categories exist, preventing degradation over time. IA reviews work best as ongoing practice. As products grow and add features, structure that once worked becomes strained. Annual reviews catch creep before it requires major remediation. Content changes outpace design; structure review keeps the two aligned.