Skip to main content
Autobound logo

Financial Intelligence

10-K Annual Report Signals

Published Updated

82 AI-extracted signal subtypes from SEC annual filings. Detect cybersecurity investment, digital transformation, leadership changes, and cost reduction initiatives across 4,500 public companies.

Read the API docs
Companies Covered
4,500
Signal Subtypes
82
Signals / Month
~11K
Historical Depth
5+ years
Delivery
REST API, GCS Push, Generate Insights API, Flat File
Format
JSONL or Parquet
Bucket
gs://autobound-10k-v1/

What Are 10-K Filing Signals?

Weekly refresh

10-K annual reports are the most comprehensive disclosures public companies file with the SEC, covering business operations, financial condition, risk factors, and strategic direction across 100-300 pages. Autobound's AI models extract structured signals from these filings within hours of them hitting EDGAR, so sales teams get actionable intelligence without reading hundreds of pages.

Extraction covers 82 distinct subtypes across categories like technology investment, organizational changes, regulatory concerns, growth initiatives, and risk factors. Each signal includes a natural-language summary, a detailed explanation with evidence, a relevance score, confidence level, and a direct link to the source filing. Because 10-Ks contain forward-looking statements about the next 12-18 months, disclosures like increased cybersecurity or AI infrastructure spending are strong buying signals.

Example subtypes

Compliance burdencomplianceBurden

From a real record

Ciscocisco.comSep 2, 2026

Cisco shifting from point solutions to integrated platforms to combat 'increasingly sophisticated cyberattacks'.

Form
10-K
Period
FY2026
Fiscal year end
07/25
Category
Compliance burden
Tone
Negative
CIK
858877

Source: sec.gov

From a 10-K filing to your next move

What we detect, what you get, and what you do with it. Pick the path that fits your team.

Sales, marketing and RevOps: Slack, CRM fields and plays.

  1. 0103

    Companies tell the SEC their plans

    Each 10-K lays out strategy, risks and spending priorities. We read every filing and pull out the parts that matter to sellers.

    1. SEC EDGAR
    SEC EDGARJan 11, 2026

    Elastic's 10-K discloses accelerating R&D investment in AI-driven security analytics, citing competitive pressure to embed ML-native detection into their Elastic Security platform...

    Extracted
    Elastic N.V.elastic.cocompanyCybersecurity investmentsignal_subtype

  2. 0203

    You get one clean record

    Each record names the company and its domain, labels what happened, and links back to the source.

    2. Signal record

    Elastic N.V.

    elastic.coEnterprise Search

    Cybersecurity investmentcybersecurityInvestment

    AI-powered threat detection investment increasing as Elastic embeds ML across SIEM and endpoint security.

    “Elastic's 10-K discloses accelerating R&D investment in AI-driven security analytics, citing competitive pressure to embed ML-native detection into their Elastic Security platform...”
    Filing year
    2025
    Category
    Technology
    Detected
    Jan 11, 2026
    Source
    sec.gov

    signal_id bfb35bca…

  3. 0303

    Act on it where you work

    Send it to Slack, write it to your CRM, or give it to an AI agent.

    When
    signal_subtype = cybersecurityInvestment, relevance ≥ 0.8
    Do
    AE task plus an exec email that cites the 10-K line
    3. Your team acts
    or your own code
    #account-signals

    Signal alert from your workflow

    Elastic N.V.· Cybersecurity investment

    AI-powered threat detection investment increasing as Elastic embeds ML across SIEM and endpoint security.

    Play: AE task plus an exec email that cites the 10-K line

    Start the playOpen the source

    Illustration. The record fields are real.

Try it on your own accounts

1,000 free credits. Query 10k in the Signal API or from Claude.

Get a free API keyUse it in Claude via MCP

How we build this signal

How we build each 10-K signal

Every week a job reads the previous week's 10-K filings and runs each through five stages. Code checks every model answer before it ships.

Relevance floor
0.7
Model reads per filing
3
Refresh
Weekly
  1. 01

    Collection

    Each Tuesday a job queries an SEC filings index for every company in our public-company list and collects the 10-K filings dated in the previous week.

  2. 02

    Section extraction

    We fetch the filing and keep the parts that carry business detail. For a 10-K, that is Item 1 (Business), Item 1A (Risk Factors) and Item 7 (MD&A). If none is found, a trimmed copy of the full text is used.

  3. 03

    Three focused reads

    A model reads each filing three times, once per theme: leadership and strategy, technology and operations, financial health and risk. Each read picks subtypes from a closed list, uses each once and must quote the filing.

  4. 04

    Quality gates

    Code drops any signal with an unlisted subtype, a relevance score under 0.7, or evidence that is under 50 characters, a table fragment or boilerplate such as safe-harbor text. A filing whose company has no website ships nothing.

  5. 05

    Dedup and delivery

    One filing yields at most one signal per subtype. Each signal is fingerprinted on form, subtype, ticker and filing date, and a shipped fingerprint is never delivered again. Records are schema-checked before upload.

10-K Signal Data Schema

Every 10-K signal follows a normalized schema with company resolution, relevance scoring, and detailed evidence. Here is a real example from our production database.

Request, response and price →
{
  "signal_id": "bfb35bca-818c-4f60-b69c-c119cb24eca2",
  "batch_id": "2026-05-01-00-00-00",
  "signal_type": "10k",
  "signal_subtype": "cybersecurityInvestment",
  "detected_at": "2026-01-11T08:16:43Z",
  "association": "company",
  "company": {
    "name": "Elastic N.V.",
    "domain": "www.elastic.co",  // match on domain
    "linkedin_url": "linkedin.com/company/elastic-co",  // or match on LinkedIn URL
    "industries": ["Enterprise Search", "Observability", "Security Analytics"],
    "employee_count_low": 3000,
    "employee_count_high": 5000,
    "description": "Search-powered observability and security analytics..."
  },
  "contact": [],
  "data": {
    "summary": "AI-powered threat detection investment increasing as Elastic embeds ML across SIEM and endpoint security.",
    "detail": "Elastic's 10-K discloses accelerating R&D investment in AI-driven security analytics, citing competitive pressure to embed ML-native detection into their Elastic Security platform...",
    "relevance": 0.88,
    "confidence": "high",
    "sentiment": "positive",
    "source_url": "https://www.sec.gov/Archives/edgar/data/1707753/...",
    "technologies_mentioned": ["artificial intelligence", "machine learning", "SIEM"],
    "fiscal_year_end": "04/30",
    "filing_year": 2025,
    "sales_relevance": "AI infrastructure and security tooling opportunity",
    "signal_category": "technology"
  }
}

Bulk data

Need this signal for 10,000+ companies?

Get a free sample file or a bulk feed (API, GCS push, flat file). The sample is real 10-K Filings records, as a CSV.

Talk to sales

Measured coverage

What's inside 10-K Annual Report Signals

A dated snapshot counted in our production database, not a live count. Minimums, not ceilings.

Coverage for all 35 signal types →
1,204
Companies with fresh data per month
Published baseline, records through Jul 21, 2026
11,203
Records created per month
6-month average, records through Jul 21, 2026

Top subtypes

82 total
Compliance Burden3.7K · 5.4%
Market Expansion3.6K · 5.4%
Platform Strategy3.5K · 5.1%
Restructuring3.2K · 4.8%
Product Launch3.2K · 4.8%
Capex Increase2.9K · 4.3%
Margin Pressure2.3K · 3.5%
Competitor Named2.3K · 3.4%

Geography

all 38 countries →
🇺🇸United States2.9K · 68%
🇮🇳India42 · 1.0%
🇨🇦Canada33 · 0.8%
🇬🇧United Kingdom18 · 0.4%
🇮🇱Israel11 · 0.3%
🇲🇽Mexico9 · 0.2%
🇪🇸Spain6 · 0.1%
CA 778TX 376NY 371MA 257FL 165PA 153

Top industries

all industries →
Biotechnology Research436 · 10%
Financial Services262 · 6.3%
Banking211 · 5.1%
Leasing Real Estate179 · 4.3%
Software Development168 · 4.0%
Pharmaceutical Manufacturing151 · 3.6%
Oil and Gas149 · 3.6%

Join keys

fill rate
Domain100%
LinkedIn URL95%
Company name100%
Records measured 67,494+Companies reached 4,177+

Company pages

See this data on real companies

Each company page shows the latest rows from this data for free, with the date and a link to the source.

See for yourself

Check out sample data: real 10-K Annual Report records with the same fields the API returns.

Read the API docs
“They provide insight and triggers about accounts we would otherwise have trouble finding.”
Rob Consoli, Chief Revenue Officer, Skuid

Frequently asked questions

How quickly are 10-K signals available after a company files?
10-K signals are typically available within 24-48 hours of a filing appearing on SEC EDGAR. Our weekly refresh cycle processes all new filings, extracts 82 signal subtypes, and delivers them via API, GCS, or flat file.
What's the difference between 10-K and 10-Q signals?
10-K signals come from annual reports and cover a full year of strategic direction, risk factors, and business overview. 10-Q signals come from quarterly reports and focus on quarter-over-quarter changes, making them better for tracking momentum and timing outreach.
How many subtypes does the 10-K signal extraction cover?
We extract 82 distinct subtypes organized into categories like technology investment (AI adoption, cloud migration, cybersecurity), organizational changes (leadership, restructuring), regulatory concerns (compliance, ESG), and growth initiatives (market expansion, M&A).
Can I filter 10-K signals by industry or company size?
Yes. The API supports filtering by industry, sector, employee count, ticker, and company domain. GCS deliveries include all fields so you can filter downstream in your data warehouse.
Do 10-K signals include the source filing URL?
Every signal includes a direct source_url linking to the specific SEC EDGAR filing page, allowing your team to verify the context and read the full disclosure when needed.