mh studio Book a free call
← All work
Compliance · B2B SaaS · Fictional client

Clearpass: SOC 2 without months of spreadsheets.

A web app, mobile app and website that help small security teams get audit-ready, and stay that way. It reads your systems, and only changes them when you approve.

Role
Lead product designer, end to end
Deliverables
Web app, mobile app, website, design system
Benchmarks
Vanta, Drata
Team
Solo concept project
Try the product

Don't just read it. Pass the audit.

Every screen is a working prototype. Approve fixes, upload evidence, grant access that expires. All data is fictional.

app.clearpass.io
clearpass Northwind Health
Northwind Health / Overview

Overview

SOC 2 audit in 23 days
72%audit ready

Readiness by framework

  • SOC 271%
  • ISO 2700169%
  • HIPAA83%

Path to audit

  1. Scope and connect systemsDone Aug 30
  2. Controls monitored18 controls, 24/7
  3. Close the gaps28% left in all frameworks
  4. Auditor reviewStarts Dec 8 with Harbor & Lane
  5. Report deliveredTarget Jan 9

Score, last 12 weeks

+34 pts since August
Week 12: 72% ready (this week)

Needs your attention 5

Live monitoring

  • TLS scan passed on 14 endpoints2 hours ago
  • GitHub still lists 2 former employees3 hours ago
  • Security training reached 100%Yesterday
  • Ledgerly review due in 6 daysYesterday
  • AWS root account unused for 90 days2 days ago
  1. 1Switch frameworks at the top and watch the score ring and bars change.
  2. 2Open Needs your attention, pick a failing control, tap Review fix, check what will change, then approve it.
  3. 3Check the control's Activity list, upload evidence for AC-02, then press Ctrl K to jump anywhere.
  4. 4Filter Vendors by risk and send a reminder.

Web app

A top navigation and a list-with-detail layout instead of the usual sidebar dashboard. Four connected pages and a command palette.

Open full screen ↗

The brief

Clearpass is a fictional company that sells SOC 2 readiness to startups with 10 to 50 people. SOC 2 is a security audit that US business customers often ask for before they sign a contract. The buyer is usually one security lead, or an engineer who was given the job.

Business problem (assumed for this brief): accounts that stall below 50% readiness never book their audit, and churn before renewal. Design goal: get every new account past 50% in its first 30 days, without asking for risky access to production.

The client asked for three things: show exactly what is left, help people fix it fast, and make the evidence collect itself.

Constraint 1

Read-only by default

Security teams will not connect a tool that can change GitHub or AWS without asking. Anything that writes has to be asked for, scoped and logged.

Constraint 2

Auditor-proof

Every action needs a person, a time and the exact change, in a form an auditor accepts.

Constraint 3

One person, often on a phone

Most buyers have no compliance team. Alerts and approvals have to work from a phone, between other jobs.

01 · What I owned

My role, and what this project is.

Owned

  • Problem framing, constraints and the research plan
  • Benchmark teardown of Vanta and Drata
  • Information architecture for web, mobile and website
  • All UI, the permission model and the audit trail
  • The design system and the working prototypes, built in code

Team and honesty

  • Solo concept project, framed as a client brief
  • No real client, users or engineers were involved
  • In a real engagement I would share success metrics with a product lead, and the permission model with security engineering
  • Not mine in any version: backend, integrations, legal wording of audit reports

Planned timeline for this concept

About 7 weeks, split into five phases. This is how I would plan the work as a real engagement, not a record of time spent.

  1. Week 1
    Discover
    Vanta and Drata teardown, personas and the research plan
  2. Weeks 2 to 3
    Design
    Information architecture, the permission model and audit trail flows
  3. Weeks 4 to 5
    Prototype
    Working prototypes of the web app, mobile app and website
  4. Week 6
    Build
    Design system, accessibility pass and developer handoff
  5. Week 7
    Test and launch
    Task tests with an admin and an auditor, fixes and launch plan

02 · Who we designed for

One buyer, one engineer who gets the work.

Personas are working assumptions from desk research, not interview findings. They are what I would validate first.

Primary persona

Jordan, Head of Security

45-person health-tech company, first SOC 2 audit booked for next month

"I don't need another checklist. I need to know the three things that will fail the audit, and fix them before Friday."

Persona voice, not a real quote

Secondary persona

Maya, platform engineer

Owns the AWS and GitHub controls, receives the work

"Send me the exact change, not a compliance lecture. And don't touch production without asking."

Persona voice, not a real quote

Jobs to be done
  1. See how close we are, for each framework
  2. Fix gaps without hunting through five tools
  3. Hand the auditor clean, current evidence
  4. Approve access fast without breaking the rules
  5. Answer security questionnaires quickly (v2, see below)

03 · Research

What I did, and what I would test next.

Kept

  • One readiness score, broken down by framework
  • Automated tests mapped to each control
  • Shared controls across SOC 2, ISO 27001 and HIPAA
  • An auditor-facing workspace

Refused

  • A 14-item sidebar on every screen
  • Status colours with no words next to them
  • Fixes that only link out to docs
  • Hiding the hard part behind "talk to an advisor"

Done: a teardown of Vanta and Drata public onboarding and control views, November 2026. This is the only research behind this concept.

Not run yet

Research plan

  • 6 interviews with security leads at 10 to 50 person startups
  • 2 interviews with auditors about what evidence they reject
  • Unmoderated test of the Fix flow with 5 security leads. Success: 4 of 5 approve the right change without opening docs
Riskiest assumption

Will security leads approve a write at all?

If they won't, the fix flow falls back to "send the exact change to the owner as a ticket". The design already supports that, so the bet is cheap to lose.

04 · The core flow

Five steps from scattered to audit-ready.

01

Connect

Read-only links to cloud, code, identity and HR tools. Nothing can write by default.

02

Monitor

Hourly tests turn every control into one readiness score. The prototype shows a trimmed set of 18.

03

Fix

Each failing control gets a prepared fix. You see exactly what will change, approve one scoped write, then it re-tests.

04

Prove

Evidence collects itself, shows how fresh it is, and reminds the owner before it expires.

05

Audit

The auditor gets a read-only, time-boxed seat with controls, evidence and the activity trail.

05 · Key decisions

Five decisions, and what I said no to.

The riskiest decision

Read by default, write by consent

Integrations connect read-only. A fix is prepared, not run: Clearpass lists exactly what will change, an Admin approves one write scoped to that change, then it re-tests and logs who approved it.

Rejected: Auto-remediation with standing write access. Faster, but a wrong auto-fix in GitHub or AWS costs far more than a slow one, and no security lead would connect it.

Top navigation, not a sidebar

Security leads live in two places: the score and the list of controls. A top bar gives the table the full width, and the detail panel opens beside it, so nobody loses their place.

Rejected: The 14-item sidebar most compliance tools use. It puts every module on equal footing, when the daily job is mostly the score and the controls.

One score, explained

The ring is the headline, but every point links to a control you can open. Switching framework re-scores everything instantly.

Rejected: A separate headline score per framework. Controls are shared, so three numbers moving at once read as noise.

Approvals that expire, admin needs two

Access requests default to the shortest sensible window. Admin access needs a second approver and ends after 4 hours. Every decision is logged as evidence.

Rejected: A warning on admin requests. Compliance leads expect separation of duties for production admin, so a warning alone would fail review.

Critical alerts can't be snoozed

Non-critical alerts can be snoozed for 24 hours, and the snooze is logged with a name. Critical ones must be fixed or reassigned.

Rejected: Snooze anything, silently. Easy to build, and exactly what an auditor flags.

06 · Roles and permissions

Four roles. Writes and admin access always need an Admin.

In the prototype you are Jordan, an Admin, so you can approve fixes. Owners prepare them. The auditor seat is designed but not prototyped.

Admin

Jordan

Can: Connects systems, approves fixes that write, approves access (one of two for admin), manages members.

Can't: Cannot approve their own admin access request.

Control owner

Maya, Devon, Avery

Can: Sees and prepares fixes for their controls, uploads evidence, snoozes non-critical alerts.

Can't: Cannot approve a write. A prepared fix goes to an Admin.

Contributor

Anyone asked for evidence

Can: Uploads the file they were asked for and comments on it.

Can't: No settings, no fixes, no access approvals.

Auditor

External firm

Can: Read-only, time-boxed seat that ends with the audit window. Sees controls, evidence and the activity trail; can comment and request evidence.

Can't: Cannot see vendor contracts or change anything.

Scale beyond 200 seats (SSO, SCIM provisioning, bulk owner assignment, multiple workspaces) is in the v2 list below.

07 · Audit trail

For an auditor, the log is the product.

Every row has a person, a time and the exact change. Rows are append-only, so nobody can edit history, and the auditor can export them as CSV.

  • Fixes record the scoped write: which system, what changed, when access ended
  • Access approvals record both approvers and the expiry
  • Snoozes record who, for how long, and why

Try it: in the web app above, open AC-03, approve the fix, and watch its Activity list update. The frame on the right is a static design of the auditor's view.

Northwind HealthAuditor view, read-onlyAccess ends Jan 9
ControlsEvidenceActivityRequestsExport CSV
  • Nov 14, 10:02Jordan Reyes, AdminApproved fix on AC-03: remove 2 GitHub accounts. One-time write to GitHub, ended 10:02.
  • Nov 14, 10:02ClearpassRe-ran 3 tests on AC-03. All passed.
  • Nov 14, 10:03ClearpassEvidence attached to AC-03: GitHub members export.
  • Nov 13, 16:40Taylor BrooksRequested GitHub organization admin. Approved by Jordan Reyes, waiting on Maya Singh.
  • Nov 12, 09:15Avery ParkSnoozed "Ledgerly review due" for 24 hours. Reason: vendor contact on leave.

08 · Edge and failure states

What happens when it goes wrong.

Specified, not prototyped

Integration disconnected

Tests show "Not checked, GitHub disconnected 3h ago". Affected controls count as unknown, never as passing.

Specified, not prototyped

Fix fails halfway

Steps stop at the failed one. Nothing is marked passing, the partial change is logged, and you get Retry and "Open in GitHub".

Specified, not prototyped

Write permission declined

The prepared fix becomes a ticket for the control owner, with the exact change list attached.

Partly in the prototype

Evidence goes stale

Freshness shows days left. Owners get a reminder 30 days before expiry; expired evidence flips the control to "Needs evidence", in words.

Specified, not prototyped

No second approver in time

Admin requests expire after 24 hours unanswered. The requester is told; nothing is granted.

In the prototype

Empty and all-clear states

No search results, everything passing, all caught up: each has one plain sentence instead of a blank panel.

09 · Accessibility

Built to a WCAG 2.2 AA target.

A target, not an audit: no external accessibility review has been done on this concept.

Status in words, not just colour

Every badge pairs a colour with a word: Passing, Failing, Needs evidence. Readable in greyscale and by screen readers.

Contrast checked

Body text #475467 on white is 7.7:1, secondary text #667085 is 5:1. Filled buttons moved from teal #0E9384 (3.8:1 with white) to #107569 (5.6:1).

Keyboard first

Every row is a real button. Ctrl K opens the command palette, Esc closes it, and inputs show a visible focus ring.

Touch and motion

Mobile targets are at least 44px. Animations stop when the system asks for reduced motion.

10 · Design system

Calm navy, one trustworthy teal, and words on every status.

Navy #0C1A2B

Teal #0E9384

Mint #2ED3B7

Canvas #F6F7F9

Passing #079455

Failing #D92D20

Needs evidence #DC6803

Aa AC-03

Inter for everything people read. JetBrains Mono for control IDs, so they scan like the codes they are. Status badges always pair a colour with a word.

11 · Success metrics

Targets I would propose. Not results.

This is a concept, so nothing has shipped and nothing has been measured. Baselines are assumptions to replace with real data in week one.

Accounts past 50% readiness in 30 days

The north star, from the business problem

Baseline
Unknown. Assumed most stall near 40%
Target
80% of new accounts
Measured
Readiness score on day 30 after the first integration, per monthly cohort

Time to audit-ready

Matches the website promise

Baseline
Assumed 4 to 6 months with spreadsheets
Target
9 weeks, median
Measured
First integration connected to 100% readiness for the chosen framework

Evidence automated

Baseline
Manual uploads only
Target
80% of evidence items
Measured
Share of evidence with source "Automated"

Fixes approved or ticketed in 24 hours

Baseline
No baseline
Target
70% of failing controls
Measured
Alert time to fix approved or ticket sent

Fixes reverted within 7 days

Counter-metric: speed must not cost trust

Baseline
No baseline
Target
Under 2%
Measured
Approved fixes later undone in the connected system

Weekly active admins after the audit

The renewal signal: people still use it once the audit is done

Baseline
No baseline
Target
60% of accounts
Measured
Admins who open web or mobile at least once a week, 8 weeks after the report

12 · What I left out of v1

Scope cuts, and why.

  • Security questionnaire auto-answerA top request, but it does not move audit readiness, the north star. Planned for v2, once evidence coverage passes 80%.
  • Risk acceptanceDocumenting an accepted risk with an owner, reason and review date. Needed before a first real audit, so it is first in line for v1.1.
  • SSO, SCIM and multiple workspacesNeeded above about 200 seats. The target buyer has 10 to 50 people.
  • Custom frameworksSOC 2, ISO 27001 and HIPAA cover the first buyers. Custom mapping adds a lot of UI for a few accounts.
  • Standing write accessLeft out on purpose, not for later. Every write stays scoped, approved and logged.

13 · Reflection

What I learned designing it.

What I would change

I started with one-click auto-fix as the hero. Writing the permission model showed it was the riskiest idea in the product, so the fix became "prepared change, a person approves". Next time I would design permissions first, not the happy path.

What I am unsure about

Whether engineers trust a fix that comes from a compliance tool. If they don't, the better default is "send to owner as a ticket", and approval moves to the engineer.

What I would test first

5 security leads, unmoderated test of the Fix flow. Success: 4 of 5 approve the right change without opening docs.

Book a free call →See all work
Next project
Crewday

Payroll and HR for small teams.

View case study →