For Research Analysts ·
What you'll accomplish
By the end of this guide, you'll be able to upload PDFs and long-form documents — industry reports, academic papers, competitor white papers — to Claude and get synthesized summaries, key statistics extractions, and cross-document comparisons in minutes. A literature review that used to take 3–4 hours of reading becomes a 30-minute AI-assisted process.
What you'll need
Before uploading anything, decide what you're trying to find out.
Troubleshooting: Claude can't read images embedded in PDFs. If a report is mostly charts, you'll need to paste key statistics manually.
What you should see: Each uploaded file appears as a thumbnail or file name above the message box, confirming it's been processed.
Before asking specific questions, get Claude to tell you what's in the documents.
What you should see: A structured overview of the uploaded documents with key themes and statistics flagged.
Now ask your specific research questions:
Based on these documents, answer the following questions with specific citations (document name + approximate location):
1. [Your research question 1]
2. [Your research question 2]
3. [Your research question 3]
For each answer, cite which document and what section supports it.
What you should see: Answers to each question with document-level citations. This is your raw material for the "Secondary Research" section of your report.
Once you have your specific answers:
What you should see: A clean secondary research synthesis ready to paste into your report appendix or background section.
Document overview:
I've uploaded [N] documents on [topic]. List main topics per document, flag key statistics, and note any contradictions.
Targeted extraction:
Based on these documents, answer with citations: 1. [Question 1]. 2. [Question 2]. 3. [Question 3]. Cite document name and section for each.
Market context synthesis:
Write a 350-word market context section on [topic] for a [client type] audience. Include: size/growth, key trends, main challenges, 3–5 statistics. Source from the uploaded documents.
Contradiction analysis:
Where do the uploaded documents disagree or give conflicting data? Explain what might account for the differences.