Help

File Uploads - Troubleshooting Guide

Most upload issues come from one of a handful of causes. Work through the sections below in order and you'll resolve the vast majority on your own. If you're still stuck after step 5, contact us with a screenshot and we'll dig in.

Supported file types

What you can drop into any analyzer (BHR, TPI, Impecuniosity).

PDF

Both text-based and scanned. Up to 50 MB.

Word (.docx)

Modern Word format. Read in your browser.

Images

JPG, PNG, HEIC, TIFF, WebP. OCR'd server-side.

Your files don't leave your device

PDFs and Word documents are read directly in your browser - the file itself is never uploaded. Only the extracted text is sent for analysis, and personal identifiers are masked first. Scanned PDFs and images are the one exception: they need server-side OCR, which runs on our hardened infrastructure with the same masking rules applied.

Common issues - work through these in order

This usually means one of the following:

  • The file is a legacy .doc (binary Word, pre-2007). Open it in Word and choose File → Save As, then pick .docx. The newer format unzips correctly in your browser.
  • The extension has been changed manually (e.g. a .docx renamed to .pdf). Open the file in its original program and re-export.
  • It's a format we don't read yet (e.g. .xlsx, .pptx, .rtf). Copy the text out and paste it into the text box directly.

This is almost always a corporate firewall blocking part of the PDF reader. We've moved the reader inside our app to minimise this, but some networks still interfere. Try these in order:

  1. Hard-refresh the page with Ctrl + Shift + R (Windows) or Cmd + Shift + R (Mac) to clear cached scripts.
  2. Try a different browser (Chrome, Edge or Firefox).
  3. Tether to your phone's hotspot for one upload to confirm whether your office network is the cause. If it works on the hotspot, your IT team needs to allow our domain.
  4. As a last resort, copy and paste the text directly into the text box.

Some PDFs are made up entirely of scanned images with no underlying text layer. The system automatically detects this and runs OCR - but OCR can fail on poor scans. Try:

  • Re-scanning at a higher resolution (300 DPI minimum).
  • Splitting very long documents (50+ pages) into smaller chunks.
  • If the original is on paper, photographing each page in good lighting often beats a low-quality scanner.
  • Saving the file as PDF/A from Adobe Acrobat - this often regenerates a clean text layer.

The platform caps individual files at 50 MB. To get under the limit:

  • Use a free online tool such as Adobe's "Compress PDF" to shrink scanned PDFs significantly.
  • Split multi-statement bundles into one file per statement before uploading.
  • If the file is full of high-resolution photos you don't need, re-export "Print quality" rather than "High quality".

We can't read PDFs that have a password. Open the file in Adobe Reader, type the password, then use File → Print and choose "Save as PDF" (or "Microsoft Print to PDF" on Windows). This produces an unprotected copy you can upload.

Some Word documents store content inside complex elements (text boxes, comments, tracked changes) that our extractor doesn't yet pull out. If the result looks incomplete:

  • In Word, accept all tracked changes before saving.
  • Convert text boxes to inline text where possible.
  • Or save a clean copy as PDF and upload that instead - our PDF extractor handles complex layouts well.

Still stuck?

Send us the file name, the case reference, a screenshot of the error, and the browser you're using. We'll usually have an answer within the working day.

Contact support →