Surfaced

Find the story your documents never gave up.

For journalists and researchers.

The problem

Most investigative work starts with documents — thousands of them. PDFs that have never been indexed, never been searchable, never been read by anyone outside the organisation that produced them. Reading them manually takes months. Ctrl+F gets you nowhere. The patterns, the anomalies, the connections that make a story — they stay buried. Not because they are not there, but because there has never been a practical way to surface them.

How it works

01

Point Surfaced at your document collection

A folder of PDFs, scanned or text-based

02

Surfaced extracts and cleans the text

Applying OCR where needed

03

NLP and clustering analyse the collection

Language patterns, topics, and statistical relationships across the full collection

04

Receive a structured research report

Cluster analysis, topic models, classifications, and statistical findings ready for further investigation

The output

A timestamped research folder containing a cluster analysis report, topic model output, document classifications, and statistical reports including chi-square and ANOVA results. Everything is exportable and ready to feed into your own analysis or reporting workflow.

Who it is for

Surfaced is built for investigative journalists, academic researchers, and policy analysts working with large collections of unstructured documents. If you have ever stared at a folder of PDFs and wondered what is actually in there, Surfaced is for you.

Book an AI Discovery Call

Tell us about your workflow. We will tell you exactly what we can build and how long it will take.

Book a call