Industries

The method holds. The constraints change completely.

We research before we draw in every sector we work in. What differs is who we can talk to, what we're allowed to remove, and what counts as proof that the new version is better. A nurse can't pause a shift for a usability session. A payments team can't drop an identity check because it tests badly. Those two facts change a project more than any style decision ever will.

Fintech product design →Ask about your sector

Where we've done the work, and what makes each one awkward.

Six sectors, listed with the constraint we actually plan around rather than a list of buzzwords. Fintech has a full page because it's where most of our current work sits.

Fintech and payments

Onboarding, KYC, merchant tooling and money movement — where a design decision has a compliance consequence.

The constraintVerification steps cannot simply be deleted, so the work is about sequencing them around what a person can answer from memory.

Fintech product design →

Healthcare and clinical software

Charting, triage and bedside tools used standing up, one-handed, mid-shift.

The constraintThe bottleneck is almost never the feature list. It's navigation cost measured in minutes per shift.

Read the Terrace Health case study →

Logistics and operations

Dispatch boards, exception handling and freight visibility for people who work in a queue, not a dashboard.

The constraintOperators keep a parallel spreadsheet when software hides state. Designing for that habit is the job.

Read the Atlas Freight case study →

B2B SaaS and developer tools

Trials, activation, admin surfaces, permissions and the settings nobody wants to own.

The constraintActivation is usually lost between signup and the first useful output, not in the marketing funnel.

SaaS design service →

Education platforms

Course delivery, assessment and administrative tooling used by three audiences with conflicting needs.

The constraintStudents, instructors and administrators want opposite defaults. Choosing whose day gets easier first is a product decision.

Read the Orbit School case study →

Commerce and marketplaces

Discovery, checkout, seller onboarding and the returns flow that decides whether someone buys twice.

The constraintConversion work fails when it optimises a page in isolation from the fulfilment promise behind it.

Read the Kindred Market case study →

How sector knowledge shows up

Not in the pitch. In week one, when we ask better questions.

Who we recruit, and how: In fintech we test with merchants and account holders mid-task, on their own devices. In clinical software we shadow instead, because interrupting a ward round is not an option. Recruitment method is a scoping decision, and we write it into the proposal so nobody discovers it late.
What we're allowed to remove: Some steps exist because a regulator, an auditor or a payer requires them. We ask for the source of each mandatory step early, then design around the ones that are real — usually by resequencing rather than deleting.
What counts as proof: Completion rate for onboarding. Minutes per shift for charting. Exceptions cleared per operator for dispatch. We agree the measure with you before design starts, then report against it — including when a change moves nothing.
Where the system fits: Regulated products carry disclosure, error and empty states that generic component libraries ignore. Our design systems work treats those as first-class components, not edge cases bolted on at the end.

Sector work

Three projects, three different rules.

All work →

Shop owner completing a Nova Pay account setup on a phone at their counter

Fintech · 2025

Before you ask.

If your sector isn't listed, say so in the form below — we'll tell you honestly whether we've worked in it.

Do you only work in these six sectors?: No, but these are the ones where we can show shipped work. We'd rather tell you we're new to a domain and price the extra research than pretend otherwise.
Does industry experience shorten the project?: It shortens discovery, not design. Knowing the vocabulary and the usual failure points means fewer wasted sessions — the drawing and testing still take the time they take.
Will you work with our compliance or clinical team?: Yes, and earlier than most teams expect. We invite them to the first flow review so mandatory steps land in the design instead of arriving as objections at sign-off.

Working in a domain with real rules?

Tell us the constraint you keep designing around. We'll tell you how we'd approach it.

Start a project