Product UI

We design the toolsteams run on.

Alongside client websites, we design and build product interfaces: dashboards, review queues and admin consoles. Here is the CRM we run M Studio on, and a working concept for an AI security console.

Book a callSee our client websites

Internal tool · In daily use

The CRM we run M Studio on.

Every lead, campaign and draft lives here. A long research pipeline becomes a clear list: what we know about each business and the one step that moves it forward. Nothing goes out until one of us approves it.

Leads: Every business sits in one clear stage, with its next step and contact details beside it.
Campaigns: Each campaign records its audience, its offer and what is still unconfirmed.
Human approval: Drafts wait for a person to review them before anything can be sent.

Dark CRM leads screen with stage counts, a search field, a business list of three example roofing companies at different stages, and a detail panel for Example North Roofing showing its next step, a Move to audit button and contact details. — Leads: every business, its stage and the one next step that moves it forward.Open full size: Leads screen

Dark CRM campaigns screen with a saved campaign list beside the detail for an example roofing campaign: target customers, the offer, a next step card with an Update campaign files button, and counts of linked leads and attributed results. — Campaigns: who we are trying to reach, what we offer and what is still unconfirmed.Open full size: Campaigns screen

Screens show sample data, not real prospects.

Concept · Synthetic data

A policy console for AI traffic.

Security teams need to see what their AI tools are sending and change the rules without surprises. We built this working concept to show how we approach that kind of product: every decision is explained, every change shows its impact before it goes live, and every version can be rolled back.

A studio concept with invented data. It is not a client product. Try it: open a request, change a policy, then roll it back.

Policy console

Concept · synthetic data

Recent AI requests and the decision for each
Time	Team	Decision
10:42:07	Finance	Redacted
10:41:55	Engineering	Blocked
10:41:30	Support	Allowed
10:40:48	Clinical ops	Held for review
10:40:12	Finance	Allowed
10:39:51	Support	Redacted

How we approach product interfaces.

Start from the real task: We watch the job an interface is for before we change how it looks.
Show the impact first: Scope, consequences and a way back are visible before anyone commits a change.
Design every state: Empty, loading, error and permission states get the same care as the main path.
Accessible from day one: Keyboard paths, visible focus and WCAG 2.2 AA checks are part of the build, not a later pass.