Live controlled intake

RMC Device Repairs

Making a technically capable repair service feel clear before a customer posts anything.

Product strategy · Service and interaction design · Front-end implementation · Serverless workflow engineering · Release verification

Live RMC Device Repairs homepage showing the UK mail-in service, board-level repair positioning and transparent prices.

The challenge

Mail-in repair asks for trust before the technician has seen the device. The website needed to explain specialist board-level capability, real prices and service limits while making it impossible to mistake an enquiry for a booking or permission to post a device.

Built around the real constraint.

I designed the public journey as a precise repair ledger: clear scope and prices first, then a deliberately bounded request form. The Astro front end keeps the language direct and the mail-in rules visible. Behind it, a signed Netlify webhook validates the submission again, derives an opaque reference and sends separate owner and customer messages through an idempotent delivery path.

Role

One connected product responsibility.

I held the thread from product direction and interaction design through delivery and evidence.

  • Product strategy
  • Service and interaction design
  • Front-end implementation
  • Serverless workflow engineering
  • Release verification

What was built

  • Seven focused public routes covering repairs, prices, process, parts, warranty and the service boundary
  • A repair-request journey that asks for useful fault detail while warning against passwords, serial numbers and private files
  • Client and server validation, unsafe-battery stops, honeypot handling and a fail-closed intake switch
  • Signed webhook verification, opaque request references and separate owner and customer email delivery
  • Responsive image delivery, strict security headers and a deliberately separate boundary from RepairOS

Verification

  • Custom-domain, TLS, route, header and production-function checks on the live service
  • A controlled production request proving capture, owner alert, customer acknowledgement, reference consistency and reply routing
  • 136 Vitest checks, a production build and CSP verification at the delivery checkpoint
  • 118 Playwright journeys across desktop, tablet and mobile at the delivery checkpoint
  • Search indexing remains deliberately held while the remaining operational launch gates are closed

Lessons carried forward

What the work taught me.

  1. Trust grows when a service says what has not happened yet as clearly as what happens next.
  2. A simple form can still carry a serious security and delivery contract behind the interface.
  3. Release gates are part of product quality when the website opens a real operational path.

Have an idea worth building?

You do not need a finished brief. Tell me what you are trying to make—or what is getting in the way—and we can work out the right next step.