← Back to blog

The Hidden Goldmine: Public Data Sources You're Not Using

By Annie

Everyone chases the same Bloomberg terminals and expensive data subscriptions. Meanwhile, some of the richest financial datasets in the world sit in public repositories, completely free, waiting for someone to actually read them.

I'm talking about SEC filings, investor relations presentations, earnings call transcripts, Freedom of Information Act (FOIA) responses, municipal bond disclosures, patent filings, and hundreds of government databases that most investors and analysts completely ignore.

Not because they're hard to access. Because they're boring to access.

The data is there. The insights are real. But the work to extract, structure, and analyze them? That's where most people tap out.

Let's fix that. Here are the hidden public data sources that could change how you invest, analyze risk, or build products—and how to actually use them.

SEC EDGAR: The overlooked treasure trove

Every public company in the US files detailed financial disclosures with the SEC. These aren't marketing materials—they're legal documents where companies have to tell the truth, or executives go to jail.

What's in there:

  • 10-K annual reports – Full financial statements, risk factors, management discussion
  • 10-Q quarterly reports – Quarterly financials and updates
  • 8-K current reports – Major events (acquisitions, executive changes, bankruptcy)
  • 13F filings – What hedge funds and institutions actually own
  • Form 4 insider trading – When executives buy or sell their own stock
  • Proxy statements (DEF 14A) – Executive compensation, board composition, shareholder proposals

Why it matters:

  • See what institutional investors are actually buying before it hits the news
  • Track insider buying patterns—executives don't buy their own stock unless they're confident
  • Dig into risk factors sections to find what companies are really worried about (often buried in legalese)
  • Compare executive comp to company performance—misalignment is a red flag

How to access it:

  • Official: SEC EDGAR
  • Better UX: sec.gov API (free, rate-limited)
  • Even better: Tools like kibble.shop that parse and structure this for you

Real example:

Before Silicon Valley Bank collapsed, their 10-K showed a massive unrealized loss position in their bond portfolio. It was public. In a PDF. Most people didn't read it. The ones who did got out early.

Investor Relations Decks: Strategy in slide form

Every quarter, public companies release earnings presentations alongside their financial results. These aren't required by law—they're marketing—but they reveal a lot about strategy, priorities, and what management thinks is going well (or not).

What you find:

  • Forward-looking guidance and projections
  • Segment breakdowns (revenue by geography, product line, customer type)
  • Key performance indicators (KPIs) the company chooses to highlight
  • Competitive positioning and market share claims
  • Capital allocation priorities (buybacks, dividends, M&A, R&D)

Why it matters:

  • Earnings calls often sanitize bad news. The deck shows what they wanted to emphasize before the call.
  • Compare decks quarter-over-quarter—when a KPI disappears, it's usually because it got worse.
  • Watch for language shifts: "investing in growth" → "disciplined capital allocation" = they're cutting costs.

How to access it:

  • Company investor relations websites (usually under "Events & Presentations")
  • SEC 8-K filings (companies often attach the deck as an exhibit)
  • Aggregators like Seeking Alpha, but you lose the ability to scrape/analyze at scale

Real example:

When tech companies started removing "monthly active users" from their investor decks in 2023-2024, it was a signal that growth was slowing. The metrics didn't lie—they just stopped showing them.

Earnings Call Transcripts: What they actually said

Earnings calls are where executives talk to analysts about quarterly results. The prepared remarks are scripted, but the Q&A is where the real insights hide.

What you find:

  • Executive tone and confidence (or lack thereof)
  • Questions analysts are asking—these reveal what the Street is worried about
  • Hedging language: "We're cautiously optimistic" = trouble ahead
  • Defensiveness when pressed on specific issues

Why it matters:

  • Sentiment analysis works here: nervous executives use more hedging words
  • Repeated questions on the same topic = market concern is real
  • Compare what they say in the call vs. what's in the 10-Q—discrepancies are red flags

How to access it:

  • Free transcripts: Seeking Alpha
  • Company IR websites (some post audio/transcripts)
  • Paid: FactSet, Bloomberg (but free sources work fine for most use cases)

Real example:

Before Enron collapsed, analysts on earnings calls were asking increasingly pointed questions about off-balance-sheet entities. The company's evasive answers were a signal—if you knew what to listen for.

FOIA Requests: Government data on demand

The Freedom of Information Act lets you request documents from federal agencies. State-level equivalents exist too. Most people don't use this because it sounds bureaucratic. But it works.

What you can request:

  • FDA inspection reports for pharmaceutical plants
  • EPA violation records for chemical companies
  • DOT safety complaints for auto manufacturers
  • USDA inspection records for food companies
  • Regulatory correspondence (warnings, enforcement actions)

Why it matters:

  • Public companies often bury bad news. Regulators document it.
  • Safety violations, quality issues, and compliance failures show up here first.
  • This is primary source data—not filtered through PR teams.

How to access it:

  • Federal: FOIA.gov
  • State/local: Varies by jurisdiction
  • Some agencies post common requests online (FDA, EPA)

Real example:

FDA inspection reports for manufacturing facilities often reveal quality control issues months before they become public recalls. If you're analyzing pharmaceutical stocks, this is gold.

Municipal Bond Disclosures: City and state financials

State and local governments issue bonds to fund infrastructure, schools, and operations. Those bonds come with detailed financial disclosures—audited financials, debt schedules, pension obligations, revenue forecasts.

What you find:

  • State and city financial health (revenue trends, deficit projections)
  • Pension underfunding (massive risk for some municipalities)
  • Tax base erosion or growth (property values, business relocations)
  • Infrastructure investment plans

Why it matters:

  • If you're investing in munis, you need to know if the issuer can pay you back
  • If you're analyzing regional economic trends, this is ground truth
  • If you're in commercial real estate, this tells you where growth (or decline) is happening

How to access it:

Real example:

Puerto Rico's bond crisis was visible in their disclosures for years before the default. Pension liabilities, declining population, revenue shortfalls—all documented. The market just didn't care until it was too late.

Patent Filings: R&D in real-time

Patents are public. And they reveal what companies are working on years before products launch.

What you find:

  • Technology roadmaps (what they're inventing)
  • Competitive positioning (who's filing in the same space)
  • Talent signals (which engineers are named on cutting-edge patents)
  • Abandonment trends (when a company stops filing in a category, they're pivoting)

Why it matters:

  • Drug companies telegraph pipeline strength (or weakness) through patent activity
  • Tech companies reveal strategic priorities (AI, semiconductors, biotech)
  • You can identify emerging competitors before they're on anyone's radar

How to access it:

Real example:

Apple's patent filings around AR/VR headsets signaled Vision Pro development years before the announcement. The supply chain caught on early. The market didn't.

Why this matters (and why most people don't do it)

Here's the truth: this data is free, public, and valuable. But it's unstructured, messy, and boring to process.

That's the moat.

Bloomberg, FactSet, and S&P Capital IQ charge tens of thousands of dollars per year because they've already done the work of parsing, structuring, and organizing this public data. You're not paying for proprietary data—you're paying for convenience.

But if you can automate the extraction, you get the same insights for free.

That's what we're building at kibble.shop. We don't have proprietary datasets. We have structured access to public datasets. SEC filings, FRED data, FDIC reports, Census data—it's all out there. We just make it usable.

And here's the thing: the best insights often come from combining multiple public sources.

Cross-reference 13F institutional holdings with insider trading patterns. Compare earnings call sentiment with FOIA inspection reports. Track municipal bond health alongside regional employment trends. The data is there. Most people just don't connect the dots.

What's next for kibble.shop

Right now, we focus on structured APIs: FRED, SEC EDGAR, FDIC, Treasury, Census. That's 190+ financial datasets, all with real-time freshness tracking.

But the next frontier is unstructured data extraction:

  • Parsing 10-Ks and 10-Qs to extract specific risk factors, segment data, and financial tables
  • Analyzing investor relations decks to track KPI changes and language shifts
  • Sentiment analysis on earnings call transcripts
  • Cross-referencing FOIA disclosures with company financials

That's harder to build. But it's where the real edge is.

Stay tuned. 🐾

Want to explore public financial data yourself?

Browse our catalog of 190+ datasets at kibble.shop — all free while we build.