Camp Management Platform Overhaul
I designed the first version of this platform myself. Four months later, I redesigned it completely, because PostHog showed us exactly where we'd got it wrong.
- Project
- B2B2C camp management platform
- Role
- Sole designer across the entire arc
- Team
- One developer
- Client
- CampMap, interactive maps for campsites
- Constraint
- Four months, limited budget
4
Months, MVP to full redesign
1
Developer
2
User types on one design system
4
Focus areas in the redesign
Two users, one platform.
CampMap builds interactive maps for campsites. The platform serves two very different users: camp admins who run their site (tracking inquiries, managing parcels, handling bookings) and campers who need to find their way around and book a spot.
I built the MVP first, by design: we needed something real in users' hands before we could make good decisions about what to build next. PostHog went in from the start, and once enough behavioural data had accumulated, we went back and rebuilt. With one developer and a limited budget, every design decision had to be worth building.
The hardest part wasn't the redesign. It was being the person who built the original and having to argue honestly about why it wasn't working.
Campers weren't failing at the inquiry step. They were failing to reach it.
Campers were landing in the app, getting lost, and leaving without ever completing an inquiry: the one action that directly drove revenue for camp admins. The drop-off happened earlier than the form, because users couldn't find their way to it.
No orientation on landing. Couldn't navigate to a parcel. Inquiry flow too hard to find. Left before taking any meaningful action.
Working around the platform, using external tools to track things that should have lived in the product. Missing features, not bad UX.
MVP, diagnosis, redesign.
MVP
Designed for speed, not polish. Stripped everything not critical to the core loop and prioritised the admin inquiry workflow and the camp map. PostHog from day one.
Diagnosis
Mapped drop-off points against user intent, and read admins' workarounds as a list of missing features.
Redesign
Four focus areas, sequenced with the developer around technical constraints. Some features moved to a later phase when cost outweighed impact.
Design system before screens.
With one developer and a four-month window, the design system wasn't optional. It was the only way to move fast without creating a visual mess. I built it from scratch for both the admin side and the camper-facing product, which have meaningfully different interaction patterns. Building it upfront meant the developer wasn't making design decisions at implementation time.
- Typography
- Colour tokens
- Spacing
- Inquiry cards
- Map overlays
- Navigation
- Notification states
- Onboarding elements
Given the budget, every component had to earn its place: complete enough to be useful, simple enough for one developer to implement consistently.
Two experiences, one system.
An interactive map with GPS navigation to parcels, a streamlined inquiry flow surfaced earlier, onboarding that orients before exploring, and notifications on bookings.
An inquiry management dashboard, real-time notifications, and a cleaner IA that makes the core workflow (track, respond, manage) achievable without training.
Honest about what we have.
Quantitative retention and conversion data from the redesign period isn't available to share. What PostHog showed qualitatively was clear: campers abandoning before ever reaching the inquiry step was the primary problem, and the redesigned navigation and onboarding were built directly against it.
The more meaningful signal is structural: a complete design system, GPS navigation, a rebuilt inquiry flow, onboarding and notifications shipped in four months, with one developer. That took tight prioritisation and sequencing, not just design craft.
Honest about my own work.
As the solo designer on a constrained team, I made consequential calls without a design manager to validate them: pushing GPS navigation onto the roadmap, advocating for structured onboarding when the pressure was to ship features, and doing the design system first against the instinct to go straight to screens.
It's easy to be attached to your own work. The job was to look at the data and respond to what it said, not to what I'd hoped it would say.
New & Improved Performance Tracking →