About UK Data Services

UK Data Services is the trading name of DataNorfolk Ltd, a company registered in England and Wales (number 08576932) and incorporated on 20 June 2013. We are based in Hickling, Norfolk, and work with clients across the UK remotely.

The business is run by Peter Foster, its founder and Managing Director. Peter scopes every project, writes most of the extraction and reconciliation code himself, and is the person you deal with from the first call to the last delivery. There is no account manager between you and the engineer.

We started because most web scraping tools either broke constantly or needed more engineering time to maintain than to write from scratch. The alternative we settled on is managed extraction: we build the collector, run it, watch it, and fix it when the source changes, and you receive clean data on a schedule.

Since then we have delivered over 500 projects across e-commerce, finance, property, legal, manufacturing and healthcare. Recent work includes daily regulatory document archives for an international law firm, stock and price aggregation across distributor portals for an electronics business, and pricing research for a management consultancy. Several of those are written up, with the client anonymised, in our case studies.

Fixed scope and price. Every project starts with a written scope and a fixed quote. If the job changes, the quote changes in writing before any extra work starts.

A sample before you commit. We extract a few hundred rows first so you can check the fields, formats and coverage against what you expected. Most problems surface here, when they are cheap to fix.

Reconciled, not assumed. Each delivery is checked against a count or figure you already trust: the number of listings a portal shows, a regulator's own published total, last month's file. That is how we hold the 99.8% accuracy standard quoted across this site, and it is why a delivery sometimes arrives with a note explaining a difference rather than a silent gap.

Traceable to source. Each record carries the URL it came from and the time it was captured, so any figure can be audited later. Our methodology page goes through this in detail, including how we handle bot-protected sites and PDFs.

Watched after launch. Collectors are monitored, and when a source changes its layout you get a change log entry and a fixed collector, not a quiet run of empty files.

Delivered where you work. Data lands in your S3 bucket, SFTP server, database or spreadsheet, in the format you asked for. We do not hold your data longer than the job needs.

We assess each source against UK GDPR, the Computer Misuse Act 1990 and the site's own terms before we start, and we say no when the answer is no. We do not collect personal data without a lawful basis, and we do not scrape behind logins we are not authorised to use.

We do not pass work offshore. Clients work directly with the people scoping and building their data solution. There are no invented biographies or stock portraits here, and no fabricated reviews.

The same code that runs client projects also runs a few things anyone can use: a scrapeability checker that tells you how hard a site is to extract from, a cost calculator, a data converter, a robots.txt analyser, and the CMA cases dataset, a reconciled record of Competition and Markets Authority cases built and maintained by Peter Foster.