A static website that acts as the clean decision layer for an Alaskan ATV/UTV rental fleet — routing every rental choice to the right Xola checkout, and the complicated cases to a human.
The business
An ATV/UTV rental operation where customers choose between self-guided pickup rentals (one to five days, or longer custom trips) and delivered trailhead packages. Xola handles scheduling, checkout, and payments. The problem is everything before checkout: helping a visitor pick the right vehicle, duration, and option without confusion.
The real problem
Rental choices multiply fast — vehicle × duration × pickup-versus-delivered — and a generic website turns that into a maze. Long custom rentals involve trailers, deposits, and custom pricing that genuinely need a conversation. And fleet reality changes: when a vehicle is out of service, it must disappear from the customer's view entirely without anyone editing page layouts.
My responsibility
Design and build of the entire decision layer: the site architecture, the booking-data model that drives it, the routing logic into Xola checkout, and the rules for what customers never see.
What was built
01
The website is a decision layer, not a booking engine. Customers make two or three clear choices and land on exactly the right Xola checkout — the site never handles money, availability calendars, or payments.
02
Three pickup vehicles across one-to-five-day durations map to fifteen direct Xola checkout links — each choice goes straight to the correct pre-configured checkout instead of a generic catalog.
03
Rentals of six days or more involve custom pricing, trailer configurations, and deposits — so that path deliberately opens a text-message conversation with the operator instead of pretending automation can scope it.
04
Every bookable option — price, duration, vehicle, checkout URL — lives in a single data file. Operational notes for the owner live alongside, in a field the site is designed never to render publicly.
05
An unavailable vehicle is flagged once in the data file and disappears from the fleet grid, selectors, and pricing everywhere — while remaining documented internally for when it returns.
06
Delivered trailhead packages filter by location, date, and vehicle so only genuinely available combinations are offered.
System view
Self-guided pickup rental, or a delivered trailhead package.
Only available vehicles are shown; one to five days for direct booking.
1–5 days → direct Xola checkout link (15 pre-mapped routes)
Six or more days needs trailers, deposits, and custom pricing.
6+ days → opens a text-message conversation with the operator
Scheduling, payment, and confirmation stay in the proven booking backend.
Inspectable proof
Sanitized artifacts show the underlying work without exposing client identities, credentials, private dashboards, or customer data.
The routing model is shown without checkout URLs, prices, phone numbers, or equipment identifiers.
The build exists, but the status remains validation until production data and hosting are independently confirmed.
Evidence
Astro codebase with the bookingData source-of-truth file: per-option pricing, durations, checkout URLs, and the hidden-equipment flag — inspectable and version-controlled.
Documented decision model: 15 direct 1–5-day checkout routes, the 6+ day SMS gate, and customer-facing hiding of unavailable equipment.
Every checkout URL, price, and availability value is being verified against the live Xola configuration before launch.
Cloudflare Pages deployment, canonical domain, and redirects are pending final confirmation.
Results & current state
The full decision layer is built and modeled around one editable data file.
Repository evidence: routing logic, fleet data, and checkout mapping exist and build; no claims about live performance are made.
Complex bookings route to a human by design.
The 6+ day path intentionally opens SMS — scope decisions stay with the operator instead of a form pretending to price custom trips.
Describe your setup in a few sentences — you'll get a plain answer about whether and how I can help.