docAnalyzer

Get your first answer in 5 minutes

Upload a file, wait for it to process, ask your first question, follow the citation back to the source.

The fastest way to learn docAnalyzer is to use it. This article walks you through the end-to-end loop on a single document: about five minutes, no setup beyond having something to upload.

1. Upload a document

  1. 1

    Click Add Documents in the top right from any page in the app.

  2. 2

    Drag your file onto the drop zone, or click to browse. You can drop several at once, and they upload in parallel.

  3. 3

    Wait for processing. A progress bar shows up next to each file; a typical 10-page PDF finishes in 15–30 seconds.

  4. 4

    When the green check appears, the document is in your library and ready.

While the file processes, the document's info panel shows each stage: extraction, structure analysis, and indexing. Scanned pages are OCR'd automatically; if the result is poor, you can run Enhanced OCR from the document's action menu later.

2. Open the document

Click the document in your library. The right-hand info panel opens with metadata and a Chat button. Click it to open a Focus: a chat narrowed to this one document.

You don't have to open Focus directly. If you're not sure what to ask, you can talk to Ask docAnalyzer first and let it suggest where to dig in. For this walkthrough we're going straight to Focus.

3. Ask a question

Type a question in plain English. Some shapes that work well on a first try:

  • A specific fact: "What's the renewal clause in section 12?"
  • A summary: "Summarize the executive findings in three sentences."
  • A list: "List every email address mentioned in this document."
  • A comparison: "How does this proposal differ from a standard MSA?"

Hit return. The answer streams in. Cited claims appear as clickable links that point to the exact place in the source: a page, a section, or a cell.

4. Follow the citation

Click a citation. The document viewer opens scrolled to the cited page (or section, or cell for spreadsheets). Read the surrounding context to verify the answer is grounded in the source.

This is the loop docAnalyzer is built around: ask a question, get a cited answer, click through to the source, decide whether to use the answer or sharpen the question.

What's next

You've done the core flow. From here:

Was this helpful?