What does this audit produce?

An experience bar: the brand, UX, and design conventions the leading sites in a vertical observably share, plus the gaps even the leaders miss, written as a standard a new build must meet or beat. It is grounded in cross-site patterns rather than aesthetic opinion, and it hands off to the build skills (creative-direction, design-standards, information-architecture, and the relevant build skill) as the bar to meet.

What are the seven dimensions?

Primary-task prominence, layout register and density, merchandising and category surface, primary navigation and search paths, brand register and conviction, trust and conversion signals, and the synthesis of recurring vertical conventions. Each is judged by observable questions across the leading sites and scored as pattern present (the field converges on a convention), mixed (it splits across approaches), gap (no convention emerges or every leader misses it), or not_assessable when the evidence cannot support a judgment.

How does it stay objective about design?

Through the honesty guardrail, which is load-bearing. Report what the leading sites observably do and where they observably fall short, never an aesthetic verdict like 'clean', 'modern', or 'good design' with no observable signal underneath. A pattern is a recurrence across 3 of N or more leaders, not one site doing something distinctive. If a finding cannot point at a specific observable thing, it is taste, not a pattern, and taste belongs in design-standards or art-direction rather than in an experience audit.

Why does it need the rendered sites, not static HTML?

Because most dimensions cannot be judged honestly from un-rendered markup. Static HTML supports only partial assessment (primary navigation, the category surface count, the presence of search), while layout density, brand register, motion, trust-signal prominence, and the rendered hierarchy of the first viewport all need the rendered page. If only static HTML is available for a site, mark every rendered-only dimension not_assessable for that site rather than guessing at design quality you cannot see.

How does it relate to vertical-site-conventions?

They are two sides of one standard. Competitor-experience-audit measures the competitive field on seven dimensions; vertical-site-conventions builds to the same seven, dimension by dimension. The audit's output maps one-to-one onto the build: its patterns name what to build, and its gaps section names the positioning wedge the build can exploit. Run the audit to set the bar, then hand it to the build skill to meet it.

Skill · Competitor experience audit

Competitor experience audit.

The experience bar a new build must meet, from patterns, not taste.

Capture the brand, UX, and site design that the leading sites in a vertical observably share, and where even they fall short. The output is an experience bar: the standard a new build must meet or beat, grounded in cross-site patterns rather than aesthetic opinion. Seven dimensions, each judged by observable questions across the field.

It is the brand and UX counterpart to a technical audit. Design is more subjective than a missing canonical, so the skill makes the subjective assessable: every dimension reports a pattern or a gap, and marks not_assessable whenever the evidence does not support a judgment.

Audience: strategists and design leads setting the experience bar before a build in a new vertical, or pairing a design review with a technical audit.

View the skill on GitHub Browse the full catalog

The framework

Seven dimensions, judged by observable questions.

Each dimension is scored across the leading sites as pattern present, mixed, or gap, and marked not_assessable when the evidence is thin.

01Primary-task prominence: does the site lead with the visitor's primary job or bury it? Count what competes for first attention and name what the field privileges.
02Layout register and density: retail-dense, editorial, or marketing-airy? Count the modules above the fold and describe the register the leaders share.
03Merchandising and category surface: how much of the catalog is surfaced at once, and in what shape (mega-menu, grid, side rail, faceted search)?
04Primary navigation and search paths: do the sites serve both the know-what-I-want path and the browse path, and how prominent is each?
05Brand register and conviction: the posture the leaders take in color, type, imagery, and voice, and whether it holds across the page.
06Trust and conversion signals: the recurring conventions (reviews, guarantees, pricing transparency, stock and fitment) and where they appear.
07Recurring vertical conventions: the synthesis. The conventions that recur on 3 of N or more leaders and define a credible build, plus the gap a positioning opportunity exploits.

The load-bearing rule

Observable patterns, never aesthetic verdicts.

The credibility of an experience audit erodes the moment it slides into taste. Report what the leading sites observably do and where they observably fall short, as cross-site patterns. Never write 'good design', 'clean', or 'premium-feeling' with no observable signal underneath; aesthetic verdicts belong in design-standards or art-direction, not here. If the user asks whether a design is good, redirect to the observable: the field shares X, Y, and Z; this build does X and W but not Y or Z, and that is the gap.

When the evidence cannot support a judgment, say so plainly and mark the dimension not_assessable. Layout and density cannot be judged from un-rendered static HTML, motion cannot be judged from a screenshot, and brand voice cannot be judged from a single page, so the audit works from the rendered sites and marks the rest rather than guessing.

A pattern is a recurrence across the field, 3 of N or more leaders, not one site doing something distinctive, which is a single-site observation. And an audit that names only what the leaders do well is half the value: the recurring weakness across the field is where a positioning opportunity lives, so the synthesis names the gap as deliberately as it names the conventions.

Reference files

Two references that go alongside the SKILL.md.

references/audit-template.md
A fillable experience-audit template, with generalized and named variants both present.
references/experience-dimensions-checklist.md
The observable questions per dimension, in scoring order.

Browse all reference files on GitHub

Bridges to other skills

What consumes the bar, and what it is not.

The audit produces a standard for the build skills and deliberately stays out of the lanes it routes elsewhere.

The build counterpart
vertical-site-conventions
Builds to the same seven dimensions this skill measures. The audit produces the experience bar and names the wedge; that skill composes a build to meet it.
The SEO axis
seo-competitor
Backlink, keyword, and SERP-overlap competitive analysis is a separate axis. Pair the two so a competitive review covers both experience and search.
Single-build taste
design-standards
Production design decisions and aesthetic calls on a known build belong there. This audit describes what the field does, not whether a single design is good.
Direction from scratch
creative-direction
Generating a creative direction for a new brand is its job. The audit supplies the field's register as evidence; it does not invent the new brand's position.
Real users
ux-research
Research with real participants on your own product answers a different question. The audit reads the competitive field, not your users.

Open source under MIT

Read the SKILL.md on GitHub.

The skill source lives in the rampstackco/claude-skills repository alongside dozens of other skills covering the full lifecycle of brand and product work. This page is a structured overview; the SKILL.md is the source. MIT licensed.

View SKILL.md Browse the full catalog

Frequently asked questions.

What does this audit produce?: An experience bar: the brand, UX, and design conventions the leading sites in a vertical observably share, plus the gaps even the leaders miss, written as a standard a new build must meet or beat. It is grounded in cross-site patterns rather than aesthetic opinion, and it hands off to the build skills (creative-direction, design-standards, information-architecture, and the relevant build skill) as the bar to meet.
What are the seven dimensions?: Primary-task prominence, layout register and density, merchandising and category surface, primary navigation and search paths, brand register and conviction, trust and conversion signals, and the synthesis of recurring vertical conventions. Each is judged by observable questions across the leading sites and scored as pattern present (the field converges on a convention), mixed (it splits across approaches), gap (no convention emerges or every leader misses it), or not_assessable when the evidence cannot support a judgment.
How does it stay objective about design?: Through the honesty guardrail, which is load-bearing. Report what the leading sites observably do and where they observably fall short, never an aesthetic verdict like 'clean', 'modern', or 'good design' with no observable signal underneath. A pattern is a recurrence across 3 of N or more leaders, not one site doing something distinctive. If a finding cannot point at a specific observable thing, it is taste, not a pattern, and taste belongs in design-standards or art-direction rather than in an experience audit.
Why does it need the rendered sites, not static HTML?: Because most dimensions cannot be judged honestly from un-rendered markup. Static HTML supports only partial assessment (primary navigation, the category surface count, the presence of search), while layout density, brand register, motion, trust-signal prominence, and the rendered hierarchy of the first viewport all need the rendered page. If only static HTML is available for a site, mark every rendered-only dimension not_assessable for that site rather than guessing at design quality you cannot see.
How does it relate to vertical-site-conventions?: They are two sides of one standard. Competitor-experience-audit measures the competitive field on seven dimensions; vertical-site-conventions builds to the same seven, dimension by dimension. The audit's output maps one-to-one onto the build: its patterns name what to build, and its gaps section names the positioning wedge the build can exploit. Run the audit to set the bar, then hand it to the build skill to meet it.