The templates below are the artefacts that come out of every UX audit run with this methodology. Each is shown in full on this page so you can copy the structure immediately. Downloadable file versions in PDF, Notion, Google Docs and Figma are being rolled out — sign up to the newsletter and the link for each lands in your inbox as it ships.
Available templates
Audit report structure
The end-to-end report skeleton. Executive summary, severity rubric, findings format, appendix.
Severity rubric (0–4)
The standard severity scale with definitions, decision criteria, and worked examples for each level.
Recommendation format
One finding per card. Heuristic, WCAG, severity, surface, effort, impact, problem, recommendation.
Audit presentation deck
The 12-slide deck used in the working session. Top three findings, severity overview, roadmap.
Impact-effort grid
The quadrant grid for converting findings into a roadmap. Sprint, quarter, sweep, park.
01 · Audit report structure
The skeleton of every audit report I write. Pages one to three are the document that gets read; the appendix is what gets referenced later.
Report skeleton
Copy this directly into Notion, Google Docs, or whatever your team's documentation tool is. Replace the bracketed sections with the audit's specifics.
PAGE 1 — EXECUTIVE SUMMARY [Audit name] [Scope paragraph from the kickoff] [3-line opening: what we found, in what proportion, what to do first] Top three findings (one paragraph each): 1. [Finding name] — [severity] — [recommended fix] 2. [Finding name] — [severity] — [recommended fix] 3. [Finding name] — [severity] — [recommended fix] By the numbers: Total findings: [n] Catastrophe (sev 4): [n] Major (sev 3): [n] Recommended for this sprint: [n] PAGE 2 — SEVERITY RUBRIC 0 — Not a problem 1 — Cosmetic 2 — Minor 3 — Major 4 — Catastrophe [Brief practitioner note on how scoring was applied] PAGES 3–N — FINDINGS Findings ordered by severity, descending. Each finding follows the Recommendation Format (template 03). FINAL PAGE — ROADMAP Impact-effort quadrant grid (template 05). Sprint 1 commitment list (written in the working session). Two-week follow-up date. APPENDIX - Methodology used - Tools used in evidence collection - Constraints document (from kickoff) - Out-of-scope findings (noted but not actioned)
02 · Severity rubric
The 0 to 4 scale, with decision criteria. Print this and keep it visible while scoring. Avoids the temptation to mark everything as severity 3.
Severity scale
SEVERITY 0 — NOT A PROBLEM The finding turned out to be opinion, not a violation against the framework. Drop from the report. SEVERITY 1 — COSMETIC Visual polish, mild inconsistency. Does not affect task completion. Fix when convenient. Bundle with adjacent work. SEVERITY 2 — MINOR Workable. Users find their way through. Mild friction. Fix this quarter if possible. SEVERITY 3 — MAJOR Slows or stops users frequently. Significant friction. Fix before next release. Should appear in the next sprint backlog. SEVERITY 4 — CATASTROPHE Blocks users, loses data, breaches legal minimum (WCAG AA), or causes financial harm. Fix immediately. Often a same-week patch. DECISION CRITERIA Score on three axes: Frequency — how often will users hit this? Impact — how badly does it affect them when they do? Persistence — does it improve with familiarity, or get worse? If everything is scoring as 3, the rubric is broken. Aim for distribution: ~30% sev 1–2 ~50% sev 3 ~10% sev 4 Remainder unscored / dropped
03 · Recommendation format
One finding per block. The structure forces clarity: you cannot write a vague recommendation against this template.
Single finding format
F-001 · [SEVERITY] · [SHORT TITLE] Heuristic: [Nielsen heuristic number and name, if applicable] WCAG: [Success criterion and name, if applicable] Surface: [URL or screen, device, user role] Effort: [S / M / L / XL — see effort scale] Impact: [Plain language: high / medium / low + qualifier] THE PROBLEM [2–3 sentences describing what happens, who it affects, and why it matters. No opinion language; cite the rule.] THE RECOMMENDATION [2–3 sentences describing the fix in behavioural terms, not visual terms. Reference design system components where they exist. Include a sketch only if the words can't carry the description.] EFFORT SCALE S Under one day M One to three days L One sprint XL More than one sprint
The example findings on the worked report example use this exact format. Copy the structure; don't reinvent it.
04 · Presentation deck structure
The deck for the live working session. Twelve slides. No more. The deck is a conversation starter, not a document.
Twelve-slide deck
SLIDE 01 Title + audit scope (single line) SLIDE 02 Why we did this (one sentence on the trigger) SLIDE 03 By the numbers (findings, severity distribution, next-sprint count) SLIDE 04 Top finding · F-XXX · with screenshot and recommended fix SLIDE 05 Second finding · F-XXX · with screenshot and recommended fix SLIDE 06 Third finding · F-XXX · with screenshot and recommended fix SLIDE 07 Severity distribution overview (one chart) SLIDE 08 Findings by category (heuristic, accessibility, content, performance) SLIDE 09 Impact-effort grid (template 05) SLIDE 10 Recommended Sprint 1 (the nine items) SLIDE 11 Recommended next quarter (the three to five strategic items) SLIDE 12 Two-week follow-up date + tickets-written-live commitment RULES - One screenshot per finding slide, not three - No more than 30 words per slide - The session is 45–60 minutes; the deck supports a conversation, it does not deliver content - Slides 10 and 11 get filled in live, with the team in the room
05 · Impact-effort grid
The grid that converts findings into a roadmap. One slide, four quadrants. The most adopted artefact in the toolkit because teams instantly understand it.
Quadrant grid template
↑ HIGH IMPACT
|
Q1 SWEEP | Q2 SPRINT (ship this sprint)
quiet-week | high-impact, low-effort
backlog items | the quick wins
LOW EFFORT ←——————+——————→ HIGH EFFORT
Q3 PARK | Q4 STRATEGIC (this quarter)
low-impact, | design-system
high-effort | and platform-level fixes
only if context |
changes |
↓ LOW IMPACT
For each finding:
Position based on impact (vertical) and effort (horizontal).
Sprint 1 = everything in Q2.
Quarter = everything in Q4 + selected Q1 items.
Backlog = remaining Q1.
Parked = Q3.
The single grid replaces the priority-ranked list in most teams' eyes
because it is faster to scan and harder to argue with.