Skip to main content
Search and Document Platforms
Last updated: October 21, 2026
When a document set grows past what one person can read, such as a FOIA production, a leak or years of agency records, you need a platform that indexes every file and lets a team search it. The choice turns on two questions: where the documents will live, and who needs access. Sensitive or source-protected material belongs on a platform you host yourself.
| Platform | Model | Strengths |
|---|---|---|
| ICIJ Datashare | Local / server | Batch search, NER pipelines, privacy-preserving |
| OCCRP Aleph | Server | Cross-dataset entities, investigations workspace |
| DocumentCloud | Hosted | Public publishing, annotation, embeds |
| Open Semantic Search | Self-hosted stack | Facets, thesaurus, Solr-class search (see Open Semantic Search product) |
Platform notes
- ICIJ Datashare is open-source and self-hosted, built by the International Consortium of Investigative Journalists. It ingests PDFs, emails, office documents, images and archives, runs OCR on scans, and detects people, organizations, locations and email addresses. Queries can combine facets with boolean, wildcard and fuzzy operators. A server mode supports several users with shared tags. Because it runs on your own machines, no outside cloud service sees the documents.
- OCCRP Aleph indexes both documents and structured data (CSV, spreadsheets, SQL), and it cross-references names of people and companies against watchlists. Status as of 2026-10-21: OCCRP says it has rewritten Aleph from scratch as Aleph Pro and is focusing its resources there. It committed to maintaining the open-source repository only until December 31, 2025, after which the core team provides no further updates, bug fixes or support. Check the current terms before starting a project on self-hosted Aleph.
- Open Semantic Search is free software you run on your own server. It provides full-text, thesaurus-based and faceted search, collaborative tagging and annotation, alerts and watchlists, and automatic text recognition.
- Google Pinpoint is a hosted research tool for journalists. It searches documents, transcribes audio and video, and turns scanned tables into spreadsheets. Google says collections are private by default.
Search tactics (quick reference)
- Phrase search for statutory citations and unique contract clauses.
- Boolean blocks for (A OR B) AND NOT C.
- Proximity where the platform supports it — co-location of two payment codes, for example.
- Facets to isolate date ranges and file types before deep reading.
- Fuzzy search to catch OCR errors and spelling variants of a name. Datashare and Open Semantic Search both support it.
- Save every query in your research log, so a colleague can rerun it and a reader can see how a document was found.
For the processing steps that come before search, see Document Analysis — Overview.
Related skill
document-research-specialist
Sources
- ICIJ, Datashare README (read 2026-10-21)
- OCCRP, Aleph repository README, including the Aleph Pro transition notice (read 2026-10-21)
- Open Semantic Search (read 2026-10-21)
- Google Journalist Studio, Pinpoint (read 2026-10-21)
Expanded with AI research and drafting assistance from the primary sources cited; editorial review pending.
