Financial Data Integration into Generative Search

Generative Search integrates structured financial data directly into AI-generated responses

• Unified Financial and Qualitative Insights in Generative Search
• What are the Core Financial Data Components?
• How Dataset Selection Works in Generative Search
  • When Generative Search Uses Equity Screener
  • When Generative Search Uses Canalyst
  • FAQs
  • Summary
• Example Generative Search Questions You Can Ask

Unified Financial and Qualitative Insights in Generative Search

AlphaSense now delivers more complete and trusted answers in Generative Search by combining structured financials, company KPIs, and transaction data with premium qualitative content like filings, broker research, transcripts, and expert calls.

This integration bridges traditional financial analysis with AI-powered research, enabling you to ask natural-language questions about financial performance and trends, combine qualitative insights with quantitative data in a single response, and receive answers grounded in both financial data and underlying source documents.

What are the Core Financial Data Components?

Generative Search is built on a rich foundation of structured and qualitative data, allowing it to deliver complete, trusted, and auditable answers. Below are the core data components that power Generative Search results.

Historical Financials & Consensus

• Access financial performance and consensus estimates across more than 22K public companies with 20+ years of historical data.
• When financial data is used in a Generative Search response, citations link directly to our Equity Screener so you can continue to refine your screen.

Company-Specific KPIs (Canalyst)

• Generative Search can answer questions using sector-specific KPIs across 4K+ companies, including metrics like Rule of 40 for SaaS companies, Load Factor for Airlines, and RevPAR for Hotels.
• All KPIs are fully auditable within Canalyst Models.

M&A Transactions

• Create natural language screens and explore insights across approximately 1 million M&A deals.
• Generative Search can generate rationale-rich summaries with one click, complete with citations back to the underlying transaction data in the M&A Screener.

Funding Rounds

• Generative Search incorporates data from over 700K funding rounds to enhance company overviews, peer comparisons, and market landscape analysis or help benchmark valuation multiples with precedent rounds.

Market Data

• Global equities and pricing provide quick market context within Generative Search responses, helping you ground insights in real-time market conditions.

Institutional Investor Holdings

• Generative Search can provide answers about institutional ownership of specific companies like number of shares, notional value, % change in shares, and % of portfolio sourced from 13-F filings.

By combining structured data (financials, KPIs, deals, and market data) with premium qualitative content (filings, broker research, transcripts, and expert calls), Generative Search delivers domain-specific AI that produces more accurate, trusted, and actionable answers.

The result: faster decision-making, full auditability back to source documents, and powerful insights without leaving your existing research workflow.

How Dataset Selection Works in Generative Search

Generative Search automatically selects the most appropriate financial dataset based on:

• The metric referenced
• The company’s coverage
• Whether the query requires standardized fundamentals or company-specific KPIs

This selection is expected behavior and is designed to return the most reliable dataset for the type of question asked.

When Generative Search Uses Equity Screener

Equity Screener is used for standardized financial fundamentals across broad public company coverage.

This includes:

• Revenue
• EBITDA
• Net income
• Cash flow
• Margins
• Valuation multiples (EV/EBITDA, P/E, EV/Revenue, etc.)
• Consensus estimates
• Market data

Example prompts that will typically pull from Equity Screener:

• “Compare NVIDIA and AMD revenue growth over the last 3 years.”
• “What is Tesla’s current EV/EBITDA multiple?”
• “Show Apple’s gross margin trend since 2020.”
• “What are the top 10 semiconductor companies by revenue growth?”

If your query focuses on standard financial statement metrics or valuation ratios, Generative Search will generally pull from Equity Screener.

When Generative Search Uses Canalyst

Canalyst is used when queries require company-specific segment data or KPIs.

Canalyst provides both:

• Company-specific KPIs
• Industry-relevant KPIs (as modeled within specific companies)

This includes:

• Segment revenue (e.g., by geography or product line)
• ARR, NRR, same-store sales
• Load Factor, RevPAR
• Other modeled operating metrics
• Detailed financial breakdowns tied to analyst-built models

Important:

Canalyst data applies only when a specific company is referenced.  It does not power screening queries.

For example:

• “Show all SaaS companies with ARR above $500M” → will not use Canalyst.
• Canalyst requires a specific company to be specified.

Example prompts that will typically pull from Canalyst:

• “Break out Microsoft’s revenue by segment for the past 3 years.”
• “What is Snowflake’s net revenue retention?”
• “Compare Lululemon’s US vs international same-store sales growth.”
• “Using Canalyst data, show Amazon’s AWS revenue growth.”

If your query references segments, KPIs, or operating drivers for a specific company, Generative Search will use Canalyst data where coverage exists.

FAQs

What If Both Datasets Contain Similar Metrics?

Some historical financial datapoints may exist in both datasets. However, strict rules guarantee that for any given metric, the data will always come from a single data provider and is never mixed.

In those cases:

• Standardized metrics → typically sourced from Equity Screener
• Segment or KPI-driven analysis → sourced from Canalyst

Both datasets are auditable and source-linked within AlphaSense.

Can I Force Generative Search to Use Only Canalyst?

Currently, dataset selection is automated.

To increase the likelihood of pulling from Canalyst:

• Explicitly reference “segment,” “KPI,” or the specific metric (e.g., ARR, NRR, same-store sales)
• Specify “according to Canalyst” in your prompt

Example:

• “Using Canalyst segment data, break out Amazon’s AWS revenue growth.”

Important clarifications:

• Canalyst does not support screening queries.
• Canalyst requires a specific company.
• Standardized financial statement metrics and screening logic are intentionally sourced from Equity Screener.
• Generative Search is programmed not to retrieve overlapping fundamentals across multiple datasets to prevent conflicts.

If a company is not covered in Canalyst, Generative Search will default to available standardized financial data.

Summary

• Equity Screener → Broad, standardized financial fundamentals.
• Canalyst → Deep, model-driven segments and industry KPIs.
• Canalyst requires a specified company.
• Screening queries use Equity Screener.
• Overlapping fundamentals are intentionally controlled to prevent data conflicts.
• Generative Search selects the dataset that best fits your query.

Example Generative Search Questions You Can Ask to Leverage Financial and Qualitative Data

Use the examples below as a guide for how to frame effective, natural-language questions in Generative Search and what to expect in return. 

Each row pairs a sample question with the type of structured, source-backed output you’ll receive when querying across M&A, funding, equity screeners, KPIs, and financial segments, helping you quickly see how numbers and narrative come together in practice.

M&A, Funding & Equity Screeners:

KPIs & Segments:

Institutional Investor Ownership: