Design audits for healthcare and fintech SaaS — including AI-readiness for Claude Code, Figma Make, and Cursor.
One to two weeks, async. An OOUX diagnostic, a design-system AI-readiness scorecard, and a prioritized set of strategic moves the org can act on next quarter — handed off as a method, not a heuristic checklist.
— pricing
Three tiers, scaled to the surface.
The audit is the on-ramp. Pick the surface area; the deliverable scales with it. Founding-client pricing — locks for the first three buyers, then resets to $2,500 / $5,000 / $7,500.
Starter
A single product surface or flow. Async Loom walkthrough plus a 15-page PDF report. For teams who want a senior outside read on one screen that isn't moving the metric.
Standard
A full app or one marketing site plus one product surface. Async PDF report, a Notion- or Linear-ready issue list, and a 60-minute Zoom debrief. The default tier for most B2B SaaS teams.
Strategic
Everything in Standard, plus the design-system AI-readiness review: DTCG token compliance, Code Connect coverage, semantic component naming an agent can reason about, and a Component.md spec sample for one anchor component. Where the wedge lives.
— how it works
Three steps. One strategic read.
Async by design — no live workshops, no daily standups. The engagement is structured so the team can keep shipping while the audit runs in parallel.
- 01
Brief and access
A short brief — the surface, the audience, the number you're trying to move. Read access to the product, the analytics, the Figma library, and any prior research. The Strategic tier also asks for read access to the design-system repo and Code Connect mappings, if any exist.
Day 1 — async - 02
OOUX diagnostic + AI-readiness scorecard
I run an OOUX read on the surface — objects, relationships, attributes, CTAs — and map the existing telemetry against a KPI tree built from the brief. The Strategic tier layers a design-system audit against the AI-readiness scorecard: are the tokens machine-readable (DTCG)? Is the component documentation in a format Claude Code, Figma Make, and Cursor can actually parse? What percentage of components have semantic naming an agent can reason about?
Days 2 — 8 - 03
Written brief + 60-minute readout
A stakeholder-ready PDF, a Loom walkthrough, a Notion- or Linear-ready issue list, and a live 60-minute readout. Findings are sized to what the org can act on next quarter — not a wishlist.
Days 9 — 14
— deliverables
What lands on your desk.
The audit document is the artefact. The AI-readiness scorecard is the reason to run it now instead of next year.
- OOUX diagnostic — objects, relationships, attributes, CTAs
- KPI tree mapped to existing telemetry, with measurement gaps named
- Prioritized strategic UX moves, sized by effort and expected lift
- Stakeholder-ready audit document (PDF, ~30 — 50 pages at the Strategic tier)
- Loom walkthrough (async)
- Notion- or Linear-ready issue list
- 60-minute live readout with the team (Standard and Strategic)
- Design-system AI-readiness scorecard — DTCG tokens, Code Connect, semantic naming, agent-parseable docs (Strategic only)
- Component.md spec sample for one anchor component (Strategic only)
- Method handoff — the templates the audit was built against
— fit
An honest read on whether this is the right tier.
— best for
Healthcare and fintech SaaS teams.
The audit pays back fastest in regulated spaces — telemedicine, EHR, patient portals, embedded finance, B2B billing — where loose IA is paid for by the regulator, the operator, and the user.
Design-system leads asking whether their system is AI-ready.
If your team is using Claude Code, Figma Make, Cursor, or v0 and the output keeps drifting off-system, the gap is almost always in how the design system is documented for agents — not in the agents themselves. The Strategic tier exists for exactly this read.
Buyers who want a senior outside read before committing to a rebuild.
Engagement-tier work starts at $18,000. The audit exists so a buyer can put a serious read on the table before that conversation, without committing to an 8-week scope.
— not for
Visual-design refreshes or brand work — the audit is about the object model, the measurement plan, and AI-readiness, not the surface treatment.
Free or low-cost reads — the free pre-audit tool exists for that. The paid tier filters out 'free audit' requests.
Teams that want a heuristic checklist — Nielsen's heuristics are public; the audit is strategic UX work plus a design-system review for agentic context, not a usability sweep.
— before you book
There's a free pre-audit tool if you'd rather start there.
A scorecard that runs on a single URL — heuristic flags, IA observations, and the questions a senior consultant would ask before quoting a real engagement. Email-gated, one report per address, no human in the loop.
— booking · 2026
If the design system won't serve the agents, that's the audit.
A 30-minute fit call costs nothing and is the fastest way to find out whether the audit is the honest tier. If it isn't, I'll say so on the call.