Worked example
From a 120-page annual report to a cited two-page brief
A public annual report PDF becomes page-anchored Markdown, a set of extracted tables, answers with page citations, and an executive brief a client can verify.
- Segment
- Independent consultant
- Industry
- Professional services
This is a reproducible walkthrough on public material, not a customer testimonial. Metrics are derived from published unit prices and limits.
Published · Reviewed
Situation
A consultant needs to brief a client on a listed company's annual report by tomorrow morning and must be able to point to the page for every figure.
Problem
Text copied from the PDF broke tables and lost page numbers, so figures could not be traced and the numbers had to be retyped.
Approach
Extract once with page anchors, pull the tables as XLSX, and let Ask This Source answer the client's likely questions with page citations before writing the brief.
Implementation
- Upload the report to the Document Extractor (1 unit). Plain text is immediate; the Docling Markdown with `<!-- page N -->` anchors and the headings outline follows.
- Export the detected tables as XLSX for the financial summary.
- Ask This Source: revenue by segment, headcount change, stated risks; each answer cites the page (10 units per question).
- Create From Source, executive brief format, formal tone (10 units); open it in the Markdown Editor to add the client's context and export DOCX.
- Annotate the client's own copy of the PDF in the PDF Editor with the page references, in the browser.
Result
- Every figure in the brief carries the page it came from, so the client can open the report and check it.
- Tables arrive as spreadsheets rather than retyped columns.
- The original PDF was processed in memory and never stored.
Metrics
| Metric | Value | Basis |
|---|---|---|
| Quota used | about 41 units | 1 extraction + 3 questions at 10 + 1 brief at 10, at published unit prices; a scanned report would add 3 units per OCR page. |
| File size allowed | up to 200 MB on Pro | The documented paid-plan upload limit; 25 MB on Free. |
Related capabilities
Document Extractor
Extract clean text and LLM-ready Markdown from PDF, DOCX, PPTX, XLSX, TXT, EPUB and HTML, with OCR for scans, then export in nine formats.
About Document ExtractorAI Summary, Ask and Create
Summarize any source into an overview, key points and timestamped chapters; ask questions with citations; create posts, briefs and scripts from it.
About AI Summary, Ask and CreateMarkdown Editor
Write, edit and preview Markdown live, open a document extraction to polish it, and download .md, DOCX or HTML; nothing leaves the browser.
About Markdown EditorPDF Editor
Annotate, sign, fill, watermark, merge, split, reorder, rotate, protect and unlock PDFs in the browser, with nothing uploaded for editing.
About PDF EditorReproduce this on your own source
The steps above use the same tools and prices you get; start with the first one.