Healthcare · Telehealth
SkinOS
A web-based skin analysis platform for spas and salons: ten skin factors read from a phone photo, turned into treatment and product recommendations a practitioner can act on in the room.
- Flagship build
- Seed
- 16 weeks
- Focus
- SAAS, A.I, CUSTOM-WEB-APP
- Delivered
- 01 Feb 2026
- Location
- United Kingdom
Outcome
Ten skin factors scored from one smartphone photograph, running in the browser with no install and built HIPAA-aware.
The working product supported the client's seed fundraising materials.
What it is
SkinOS is a skin analysis and client management platform for spas, salons, skincare clinics and independent aestheticians. A practitioner photographs a client with a smartphone; the platform reads ten skin factors from the image and returns personalised treatment and product recommendations.
It is positioned deliberately as a cosmetic assessment and client education tool, not a diagnostic device. That line is drawn in the product itself rather than in the small print, and it shapes what the interface is allowed to assert.
What we built
The platform covers the whole consultation, not just the analysis:
- Skin analysis scoring ten factors from a single smartphone image.
- AI recommendations mapping those scores onto the treatments and products a given location actually offers.
- Progress tracking that compares before-and-after images so a practitioner can show a client measurable change rather than describe it.
- Client management holding consultation history across visits.
- Multi-location and team management, so a group can run several sites and several practitioners under one account.
The platform is built HIPAA-aware — access controls, audit trails and data flows designed for health information from the start — and it runs entirely in a browser: no install, no device management, nothing for a salon owner to roll out.
SkinOS 2.0
The second version reordered the consultation around how it actually happens. Recommended treatments now appear before products, because a practitioner is having a treatment conversation first and a retail one second, and the original order made the platform feel like it was selling.
Interface elements that surfaced information a practitioner could not act on were removed rather than improved. A shorter assessment flow that ends in a decision beats a longer one that ends in a dashboard.
Keeping three platforms up
SkinOS runs alongside two sibling products for the same client. We moved to a weekly end-to-end pass across all three, exercising the real user journeys rather than waiting for reports to arrive.
It is unglamorous work and it is the reason the platforms stay boring, which for a system a salon opens in front of a paying client is the whole objective.
How we built it
Built the analysis as a browser workload so a salon could run it on the phone already in the room. The obvious build for a computer-vision product is a native app. For this audience it would have been the wrong one: practitioners work across multiple locations on whatever device is to hand, and an install step between a client sitting down and an assessment starting is a step most consultations would never take. Everything runs in the browser instead. A practitioner opens a URL, photographs a client's skin, and has ten scored factors and a recommendation set before the consultation has properly begun.
Have a project like this in mind?
Tell us what you're building. We'll show you how we'd approach it and what it would take to ship.