The Due Diligence Problem

When a deal comes across my desk, the clock starts ticking. Partners want a memo by end of week. The standard process:

1. Read the last 4 quarters of earnings calls (2-3 hours)
2. Pull SEC filings and scan for risk factors (1-2 hours)
3. Check insider trading activity (30 min on EDGAR)
4. Read industry news coverage (1-2 hours)
5. Patent landscape analysis (2-3 hours)
6. Check if anyone in Congress is trading the stock (30 min on Senate.gov)

That's 8-12 hours of manual research before I write a single word of analysis.

I cut it to 2 hours with one API.

Step 1: Ticker Intelligence — Everything in One Call

This is the killer feature. One endpoint returns a cross-source intelligence briefing: earnings, insider trades, congressional trading, SEC filings, and news — all correlated.

curl "https://api.gettrawl.com/api/intelligence/AAPL?sections=earnings%2Cfilings%2Cform4%2Ccongress%2Cnews"

Step 2: Earnings Call Deep Dive

I need to hear what management is actually saying — not the analyst summary. Speaker-segmented transcripts let me search for specific themes across quarters.

curl "https://api.gettrawl.com/api/earnings/search?ticker=AAPL"

Step 3: Patent Landscape

Patent filings reveal R&D direction 2-3 years before products ship. For a tech company, this is critical for understanding competitive moat.

curl "https://api.gettrawl.com/api/patents/search?assignee=Apple+Inc&q="

Step 4: Insider Trading — Are Executives Buying or Selling?

Form 4 filings reveal what management actually does with their money. Words are cheap — insider buys are not.

curl "https://api.gettrawl.com/api/filings/form4/AAPL?max_results=10"

Step 5: Congressional Trading — Smart Money Signals

Members of Congress trade stocks with access to non-public briefings. Follow the money.

curl "https://api.gettrawl.com/api/congress-trading/search?ticker=AAPL&chamber="

The Due Diligence Pipeline

from trawl import TrawlClient

client = TrawlClient()
ticker = "AAPL"

# 1. Cross-source intelligence briefing
intel = client.intelligence.get(
    ticker,
    sections="earnings,filings,form4,congress,news"
)
print(f"Signals found: {len(intel.signals)}")
for signal in intel.signals:
    print(f"  [{signal.severity}] {signal.source}: {signal.summary}")

# 2. Last 4 quarters of earnings
calls = client.earnings.search(ticker=ticker)
for call in calls.results[:4]:
    transcript = client.earnings.get_transcript(
        ticker, call.year, call.quarter
    )
    # Extract CEO/CFO commentary on growth and risks
    for seg in transcript.segments:
        if seg.speaker_role in ["CEO", "CFO"]:
            if any(kw in seg.text.lower() for kw in [
                "guidance", "outlook", "risk", "growth", "margin"
            ]):
                print(f"\nQ{call.quarter} — {seg.speaker}:")
                print(f"  {seg.text[:300]}")

# 3. Insider trading (Form 4)
form4 = client.filings.form4(ticker, date_after="2025-01-01")
net_insider = sum(
    t.shares * (1 if t.transaction_type == "P" else -1)
    for t in form4.results
)
print(f"\nNet insider activity (2025): {net_insider:+,} shares")

# 4. Congressional interest
congress = client.congress_trading.search(ticker=ticker)
print(f"\nCongressional trades: {len(congress.results)}")

# 5. Patent pipeline
patents = client.patents.search(assignee="Apple Inc")
print(f"\nRecent patents: {len(patents.results)}")

What This Changes

The Memo Template

After running the pipeline, I have structured data for every section of the investment memo:

1. Management Quality — Extracted from CEO/CFO commentary themes
2. Financial Trajectory — Earnings call guidance and forward-looking statements
3. Insider Conviction — Net Form 4 buying/selling signals
4. Regulatory Risk — Congressional trading patterns + lobbying activity
5. Innovation Pipeline — Patent filing velocity and focus areas
6. Market Sentiment — News coverage tone and volume

The analysis still requires judgment. But the data gathering is automated.