> ## Documentation Index
> Fetch the complete documentation index at: https://sourcetable.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Deep research

> Use AI to research complex topics across multiple web sources and compile structured findings into your spreadsheet.

Deep research goes beyond a simple web search. When you need comprehensive information synthesized from multiple sources, the AI browses the web, reads multiple pages, cross-references findings, and delivers a structured summary — all directly in your spreadsheet.

## How it works

The AI uses the `deep_research` tool to:

1. Break your research question into sub-queries
2. Search multiple sources across the web
3. Read and extract relevant information from each source
4. Cross-reference and validate findings
5. Compile results into a structured format in your spreadsheet

## Example prompts

**Market research:**

* "Research the electric vehicle market — key players, market size, growth projections, and recent trends"
* "Compile a competitive analysis of the top 5 project management tools with features, pricing, pros, and cons"
* "Research the SaaS pricing landscape for AI writing tools"

**Company research:**

* "Research Stripe's product offerings, pricing model, and recent acquisitions"
* "Find information about Series A funding rounds in fintech for Q1 2024"
* "Compile a list of the top 20 YC companies from the last 3 batches with descriptions and funding"

**Industry analysis:**

* "What are the key regulations affecting cryptocurrency exchanges in the EU?"
* "Research supply chain trends in semiconductor manufacturing"
* "Summarize the current state of AI regulation globally"

**Technical research:**

* "Research best practices for building real-time data pipelines with Kafka"
* "Compare PostgreSQL vs. MySQL vs. Snowflake for analytics workloads"
* "What are the most popular Python libraries for time series forecasting?"

## Output formats

The AI delivers research results in whatever format makes sense:

* **Structured tables** — comparisons, feature matrices, company lists
* **Written summaries** — executive briefings, market overviews
* **Data-ready formats** — columns and rows you can immediately chart, filter, or analyze further
* **Cited sources** — links to the original sources it referenced

## Deep research vs. web search

| Feature         | Web search                  | Deep research                                  |
| --------------- | --------------------------- | ---------------------------------------------- |
| Sources checked | 1-3 results                 | 10+ pages across multiple queries              |
| Depth           | Surface-level snippets      | Full page content, cross-referenced            |
| Output          | Quick answers               | Comprehensive structured analysis              |
| Time            | Seconds                     | 1-3 minutes                                    |
| Best for        | Quick facts, current prices | Market research, competitive analysis, reports |

## Use cases

* **Analysts** — competitive intelligence, market sizing, trend analysis
* **Sales** — prospect research, industry background before calls
* **Marketing** — content research, competitor messaging analysis
* **Finance** — company due diligence, industry risk factors
* **Operations** — vendor evaluation, technology comparison
* **Founders** — market validation, investor landscape research