Uploading Documents
Documents are the raw material of your analysis — transcripts, field notes, articles, anything text-based. This page uploads two real, fictional demo interview transcripts into the Remote Team Trust Study project from the previous page, so you can see exactly what happens on screen — and so those same two documents are ready to go when we build a codebook and start coding next.
Watch it happen: uploading two transcripts
Every screen below is real — these two files actually get uploaded to a real project. Click through it yourself, or let it play.
Supported formats
| Format | Extension | Notes |
|---|---|---|
| Microsoft Word | .docx | Most common for interview transcripts |
.pdf | Text-based PDFs only (not scanned images) | |
| Plain Text | .txt | UTF-8 encoded, unformatted text — what the demo above uses |
Monthly upload limits
Every plan includes a monthly document quota — how many new documents you can upload for AI analysis in a billing period. Documents you've already uploaded stay fully accessible even after you hit the limit; the quota only gates new uploads.
| Plan | Documents / month |
|---|---|
| Free | 2 |
| Researcher | 25 |
| Pro | 100 |
Your remaining quota is always visible in the sidebar under Free Plan Storage (or your plan's name) — exactly like the "2 / 2" shown in the walkthrough above. If you try to upload more than your remaining quota allows, Paideias shows exactly how many slots you have left and how many the current selection would use.
Why documents wait in queue
When you upload a document, it enters a processing queue — exactly like P01_Interview.txt and P02_Interview.txt in the walkthrough above. A document stays queued until your project has at least one code defined in its codebook. This is by design: the AI needs a codebook to know what to look for in your documents.
Without codes, the AI has no framework for analysing the text. Once you create your first code — manually or by generating codes with AI — every queued document begins processing automatically.
How documents are processed
Once a document enters the processing pipeline, Paideias:
- Extracts text — reads the file and pulls out all text content, stripping formatting
- Chunks the text — breaks the document into meaningful segments (typically paragraphs or semantic units)
- Indexes chunks — stores each chunk for individual coding and retrieval
You don't need to manage chunks manually — the system handles this automatically. When you open a document in Focus Mode, you'll see the full text with chunk boundaries invisible to you.
Document naming
Documents are displayed using the filename you uploaded — like P01_Interview.txt above. You can rename documents after upload without affecting the file content. Use clear, consistent naming, for example Participant_01_Interview.docx or Focus_Group_B.docx.
Removing documents
You can remove a document from your project at any time using the trash icon in its row. Removing a document also removes all codes applied to passages within it. This action cannot be undone, so make sure you've exported any data you want to keep first — see Exporting a project.