A searchable database of 1,000+ documented political lies

What the site is

Political Lies is a database of documented lies, falsehoods and misrepresentations made by UK politicians, from Boris Johnson through to Keir Starmer. The site catalogues over a thousand entries, each sourced and categorised by politician, topic and type. Users can filter by category (Brexit, COVID, NHS, Treasury) and search the full archive. It serves as both a journalistic resource and a public accountability tool.

How I came to it

I joined the project in 2021, mainly on the content side, working within the existing system, a Gatsby + Contentful build by Aidan Dunlop, whose architecture served the project well in its early years. As the database grew we wanted more direct control over the site, and with Aidan extremely busy, we migrated the whole system to WordPress. That move bought two things: much better handling of embeds (Twitter and YouTube embeds are everywhere in this content) and proper filtering and tagging, so users can now filter by category and search the posts, which wasn’t possible before. I redesigned the site in the process, keeping the aesthetic close to the original (simple, easy to navigate, easy to read) while improving what it could do.

The challenge

This was a large-scale project, as much content as development. The site needed to present a complex, growing database of political statements in a way that was browsable, filterable and searchable. Each entry has the statement, context, source and categorisation. The project also involved substantial non-development work: right-of-reply correspondence for every entry, and processing large batches of research documents for publication.

What I built

Migration and redesign. Moved the site from its original Gatsby + Contentful stack to WordPress, for better embed control, proper tagging, and a category filter and post search the old system couldn’t offer, with a redesign that kept the original’s plain, readable character.

Custom content architecture. Built using Advanced Custom Fields with custom post types, custom loop templates, and a filterable / searchable front-end. The filtering system lets users narrow entries by multiple categories simultaneously.

Every category carries its own count, and you can stack them to narrow the archive.

Batch content processing. Managed the research, writing, formatting and uploading of content across multiple batches, each containing dozens of documented entries.

Right of reply. Every party a post involved was contacted before publication: formal correspondence prepared and sent, responses tracked and managed, deliverability confirmed, and MPs’ offices phoned to confirm receipt. This isn’t courtesy; it’s the project’s legal protection against libel action. With the research behind each entry on top, the attention to detail had to be absolute: any slip was a legal risk.

The right of reply is published, not just promised: the letters, the replies and the ones who never answered.

Why it matters

Most “filterable database” projects stop at the technical filter. This one had to handle the journalistic and legal weight of the source material: every entry came with a paper trail, every politician got a chance to reply, and the architecture had to scale as the database grew. The dev work was the easy half.