Sources of Data

Quick Answer

Investment bankers pull data from five source types: commercial market databases (Bloomberg, FactSet, Capital IQ, Refinitiv), proprietary internal repositories of the firm's past deals, regulatory filings on EDGAR and EMMA, public-company investor relations sites, and financial media. Each source surfaces a specific category of data (financial, performance, issuance, transaction) used to build comparable-company sets, precedent-transaction sets, and target profiles.

Before a banker builds a single model, the workflow is "find the inputs." The data-collection topic is about where the data lives, not how it gets crunched into a discounted cash flow (DCF). Knowing that Bloomberg is a market database is in scope; building the DCF in Excel comes later in the function.


Commercial and Proprietary Market Databases

The five working source categories cover almost every data pull a junior banker makes during pitch prep, comp building, or due-diligence kickoff.

Source CategoryExamplesWhat Bankers Pull
Commercial market databasesBloomberg, FactSet, Capital IQ, Refinitiv (Thomson Reuters)Pricing, financial statements, performance metrics, transaction history, ownership data
Proprietary databasesThe firm's own internal repositories of past dealsPrecedent transactions, comparable companies, league tables, deal pipeline
Regulatory sourcesEDGAR (SEC), MSRB EMMA, FINRA-issued data, federal and state government filingsPeriodic reports, ownership filings, prospectuses, registration statements
Company internet sitesPublic-company investor relations pages, private-company corporate sites10-K, 10-Q, 8-K, proxy filings, earnings releases, investor presentations
MediaWire services, financial press, industry trade publications, equity researchNews flow, rumored deals, analyst views, industry color

Commercial market databases are the day-job tools. FactSet and Capital IQ are particularly suited to investment banking workflow because of Excel plug-ins, click-through audit trails back to source documents, and screening tools designed for building comp sets. Bloomberg is more trader-centric but is the default for real-time fixed-income data.

Proprietary databases track the firm's own deal history: every offering the firm has underwritten, every M&A engagement, and every comparable transaction the firm has executed or covered. The firm's league-table position (industry ranking by deal volume) matters in pitch decks and is calculated from this history.

Regulatory sources are the public-record backbone. EDGAR (Electronic Data Gathering, Analysis, and Retrieval) is the SEC's filings database and holds every 10-K, 10-Q, 8-K, Schedule 13D / 13G, Form 13F, Schedule 14A, Form 3 / 4 / 5, and registration statement. EMMA (Electronic Municipal Market Access) is the MSRB's parallel database for municipal-securities disclosure.

Company internet sites, specifically the investor relations sections of public companies, mirror EDGAR filings and often add earnings presentations, transcripts, and segment supplements that bankers use to triangulate against the SEC documents. Private-company sites surface business descriptions, product lines, leadership bios, and press releases that anchor the early stages of a sell-side process.

Media includes wire services (Reuters, Bloomberg News, Dow Jones), industry trade publications, and equity research from sell-side firms. Bankers consume research for industry context, but the rules governing how they talk to research analysts at their own firm are heavily restricted (covered in the research-analyst conduct section).

Exam Tip: Gotchas

  • The data-collection topic is about WHERE bankers pull data from, not how they use it in a model. Knowing Bloomberg is a market database is in scope; knowing how to build a DCF in Excel belongs to the analytical-use topic.
  • EDGAR holds public filings; proprietary databases hold the firm's own deal history. They are not interchangeable. EDGAR will not tell you that your firm led 12 healthcare IPOs last year; the proprietary database will.

Categories of Data Collected

Each source serves up one or more of four data categories. The category determines which analysis the data feeds.

Data CategoryWhat It ContainsPrimary Use
Financial dataHistorical and projected balance sheet, income statement, cash flow figuresComparable-company spreads, DCF inputs, credit analysis
Performance dataOperating metrics, key performance indicators (KPIs), same-store comps, backlog, churnIndustry positioning, growth trajectory, business quality
Issuance dataPast securities offerings (debt, equity, equity-linked) including pricing, size, allocationPricing benchmarks for new offerings, league-table tracking
Transaction dataRecent M&A deals including target / acquirer, deal value, multiples paid, consideration mixPrecedent-transaction comps, premium analysis, deal sourcing

Financial data is what the 10-K and 10-Q surface. Three years of historicals and management's forward guidance feed both the comp spread and the DCF projection.

Performance data sits one level below the financials and is industry-specific. A retailer's KPI is same-store sales (also called "comps"); a subscription business tracks churn and annual recurring revenue (ARR); a manufacturer tracks backlog. Performance data tells the story behind the financial trend.

Issuance data captures the firm's deal-tracking universe: every IPO, follow-on, debt offering, and convertible the firm has executed or competed for. Pricing, size, allocation across syndicate members, and discount/premium to a benchmark all feed into pitch materials.

Transaction data is the M&A side: every acquisition the firm tracks, the deal value, the multiples paid, and the consideration mix (cash, stock, or a blend). This is the raw material for precedent-transaction analysis.

Exam Tip: Gotchas

  • Issuance data covers securities offerings; transaction data covers M&A. The exam may test whether you reach for the right database given the question's facts. A pitch for a debt deal pulls issuance comps; a sell-side mandate pulls precedent transactions.
  • Both the firm's own deals AND competitor deals are tracked. League-table position is calculated against competitors, so the proprietary database has to mirror the broader market, not just the firm's pipeline.