One search field. Plain English, a ticker, a CIK, a date.
No modes, no tabs. Type anything — NexusAlert figures out what you mean, shows you how it understood the question as chips you can edit, and answers from 280,000+ filings with cited sources.
Try It Free
The Limitations of Keyword Search
Traditional keyword search works well when you know exactly what you're looking for — a specific ticker, a form type, a date. But investment research often starts with a question, not a keyword. You might want to know which companies recently disclosed supply chain disruptions, or which firms are facing regulatory investigations, or where management expressed concern about rising interest rates.
Keyword search can't handle these queries effectively. A company might describe supply chain issues as "logistics constraints," "vendor disruptions," or "procurement challenges" — none of which would match a search for "supply chain risk." You'd need to try dozens of keyword variations and still miss relevant results.
How it works under the hood
Under the ask-anything surface, NexusAlert uses AI vector embeddings to understand the meaning behind your query, not just the words. When you search for "companies facing bankruptcy risk," the AI finds filings that discuss going concern warnings, debt covenant violations, liquidity crises, and credit downgrades — even if the word "bankruptcy" never appears in the filing.
Under the hood, every AI-analyzed filing in our database has been converted into a mathematical representation (a vector embedding) that captures its semantic meaning. When you enter a query, it's converted into the same vector space and matched against all filing embeddings to find the most conceptually similar results.
- Natural language queries: Ask questions the way you'd ask a research analyst — "tech companies with declining margins" or "recent executive departures at financial firms."
- Concept matching: Results are ranked by semantic relevance, not keyword frequency. A filing that thoroughly discusses your concept ranks higher than one that mentions your keywords in passing.
- Find similar filings: Found an interesting filing? Click "Find Similar" to discover other filings with comparable themes, risks, or events.
Use Cases for Investors
Semantic search opens up research workflows that were previously impossible without a team of analysts:
- Thematic screening: Search for "companies investing in artificial intelligence" to find filings that discuss AI investments, regardless of the specific terminology used.
- Risk discovery: Query "revenue concentration risk" to find companies that depend heavily on a small number of customers — a common risk factor in smaller companies.
- Event-driven research: Search for "management changes after poor earnings" to find companies where leadership transitions followed disappointing financial results.
- Competitive analysis: Search for themes in one company's filing and find competitors discussing similar challenges or opportunities.
What's free and what's Pro
The search field is one box for everyone. Identifier searches — by ticker, CIK, form type, or date — are free, so anyone can pull up a company's filings without guessing keywords. The AI layer on top — plain-English interpretation and cited answers synthesized across 280,000+ filings — is a Professional feature, designed for investors who want to discover opportunities and risks across the entire SEC filing universe. Combined with NexusAlert's risk scoring and daily alerts, it provides a research toolkit that rivals what institutional desks use.