Automotive Marketplace & Dealer Stock Feed Development
I build used-car marketplaces and dealer websites that pull live vehicle stock in automatically — including from supplier systems that publish no API or export. My proof is CarVendors, a live UK used-car marketplace where stock arrives with no re-keying and safety checks protect live listings when a source breaks.
This service is for UK dealers and dealer groups whose stock lives in a system with no usable export, for marketplace operators who need ingestion from mixed sources, and for anyone who has been told by a platform vendor that the integration they need does not exist.
- Vehicle stock ingestion from sites with no API
- XML and CSV dealer feed integration
- Vehicle data normalisation (make, model, colour, fuel, transmission)
- UK registration (VRM) validation and recovery
- Faceted vehicle search and filtering
- Vehicle detail pages and enquiry capture
- Multi-dealer isolation and onboarding
- Feed monitoring, alerting and safeguards
- A scope review of where your stock lives today and what shape the data is in
- Per-source terms assessment before any feed is built
- Safety checks that refuse to remove live listings when a source fails
- Staging environment and automated deployment
- Full repository handover — README, runbook, environment matrix
- PHP
- MySQL
- Next.js
- React
- TypeScript
- Node.js
- Laravel
- cron and CLI pipelines
- cURL
- GitHub Actions
- cPanel and Vercel hosting
CarVendors is a live UK used-car marketplace. Stock is ingested from dealer sites that publish no API, normalised into the marketplace schema and published with no manual step. On the one occasion a source site failed mid-import, the run aborted in 48 seconds and no live advert was deleted.
Read the CarVendors case study- Stock feed integration onto an existing site: quoted per source
- Dealer website with stock ingestion: quoted after a scope review
- Full automotive marketplace: fixed quote after a scope review
- Feasibility test on one source: fixed price, agreed up front
Scope drives the number — how many sources, what shape the data is in, stock volume and refresh frequency. Every quote is fixed after a scope review.
Questions, answered honestly.
Usually yes. Where a supplier system publishes no API, no export and no feed, stock can be read from its own pages, normalised into your schema and published automatically. Where a feed does exist, that route is simpler and cheaper. Either way the marketplace side stays the same, so mixing sources is not a problem. Any ingestion is set up only on stock you are entitled to list, with the source operator’s knowledge, and each source’s terms are assessed before a feed is built.
Safety checks. The import runs before anything touches live inventory, active stock is counted before and after, and independent checks on success rate and post-import inventory ratio must both pass before a single record can be switched off. If they fail, the run preserves live inventory and reports why. Removals are staged and reversible for a grace period rather than deleted outright. This is the failure mode most integrations get wrong, and it is the first thing I build.
A subscription platform is cheaper to start and fine if your stock already fits its integrations. It stops working when you need a source it does not support, search logic it does not offer, or your data out of it. A bespoke build costs more up front and wins on three-year total cost, control and exit: you own the code, the domain, the data and the feed configuration, and you can move to another developer.
Often yes. The CarVendors ingestion runs on a standard shared cPanel account with no VPS and no new monthly bill. Whether that holds for you depends on stock volume and refresh frequency — a linear import has a practical ceiling per run, and past that it needs chunked scheduling or a bigger box. That is one of the things a scope review settles before you commit.
Yes. The work is built around UK conventions — sterling pricing, UK registration plate formats and their historic variants, postcode-based search, and UK GDPR cookie consent. I am based in Chattogram, Bangladesh and work hours that overlap the UK working day.
Yes. You receive the repository, hosting handover, deployment notes and documentation. No lock-in, and no dependency on me to keep running.

