SkipHireCompare
A UK skip hire comparison directory built on geospatial search, with a paid listings model and direct debit collection for suppliers.
skiphirecompare.com ↗Production deployment
Live
Served by a systemd unit behind nginx with a Let's Encrypt certificate. Verifiable by loading the site.
Supplier payment paths in production
2
GoCardless for direct debit collection, Stripe for card. Both wired, switchable by environment.
Pages and API endpoints in production
42
28 rendered pages and 14 API routes, counted from the Next.js app directory in the repository.
Each figure above is read from a system we operate, and the note says which one. Where a figure would flatter us but cannot be evidenced, it is not here.
Why it exists
Skip hire is a market with real friction and almost no digital surface. Buyers are householders and small builders who need a skip on a specific street in the next few days. Sellers are small operators, often two or three vehicles, who compete on a service radius rather than a brand. Neither side has a good way of finding the other, and price discovery is a series of phone calls.
The commercial thesis is straightforward: build the place where the search happens, then charge suppliers for position in it.
The hard part
The interesting engineering problem is not the listings, it is the question which suppliers actually serve this address. Skip hire radii are irregular, overlapping and change with fuel prices. Answering that from a flat table of postcodes gives wrong answers at the edges, which is exactly where the marginal customer sits.
We put the data in Supabase and used PostGIS for the matching, so a service area is a real geometry and “who covers this point” is a spatial query rather than a string comparison. That removed the need for a separate geospatial service, and it means the radius logic lives next to the data instead of in application code.
The front end pairs that with a Leaflet map that distinguishes paid placements from standard listings visually, so the commercial model is legible to the buyer rather than hidden in ranking.
The commercial model
Suppliers can claim a listing or pay for a featured position. Collection runs through GoCardless, which suits recurring low-value B2B billing in the UK better than card payments do, with Stripe retained as a card path. Both are wired and switchable at the environment level, so the business is not locked to one processor before it knows which one its suppliers prefer.
Where the work sits now
The platform is live and the infrastructure is production-grade, so the effort has moved from engineering to the commercial side: bringing operators onto the directory and growing coverage region by region.
That order is deliberate. A comparison site earns its position by answering the buyer’s question well in the areas it serves, which means supply is built where demand is demonstrable rather than claimed nationally on day one.
What we would do differently
We built the full marketplace before proving that suppliers would pay. The right order was probably the reverse: a much thinner directory in one region, sold to twenty operators by phone, then the platform. The geospatial work was genuinely necessary, but a good deal of the listing and billing surface was built ahead of demand for it.
We also inherited the codebase from a sister directory product rather than starting clean. That bought speed and cost us a period of carrying assumptions from a different market until they were found and removed.
Start with what is in the way.
Most of this work is shaped by the obstacle rather than the ambition. A deal that keeps stalling at the same stage, a process nobody owns, a product that will not sell, a decision that has been open for months. Tell us yours and we will say honestly whether we are the right people for it.