Current Modules
3
Business, Nonprofit, and Schools all run on the same platform shell.
Siviq
About Siviq
The platform combines multiple public-record domains into one fast, searchable system with stable URLs, shared search, and reusable normalization infrastructure.
Shared Search
Cross-Domain
One search entry point can find companies, organizations, schools, EINs, and related entities.
Normalization Model
Backend-First
Canonical facts are normalized and stored in the backend before they are rendered on the web.
URL Strategy
Stable
Pages are intended to be permanent, readable, and usable for search, linking, and future compare features.
What The Platform Is
Siviq is a modular system for turning hard-to-use public datasets into fast, searchable, structured web pages.
Each domain keeps its own ingest and normalization logic, but the platform shares core infrastructure for storage, search, related-profile linking, and page rendering.
How The Modules Fit Together
The Business module focuses on retirement-plan and sponsor data from Form 5500 filings.
The Nonprofit module focuses on organizations and returns built from Form 990 and related IRS status datasets.
The Schools module focuses on school-profile PDFs and school-level facts that can later link back to nonprofit financials and other domains.
- Shared search can surface records across modules from one query.
- Related-profile links connect entities when a stable identifier like EIN is available.
- Compare and ranking features are designed to become reusable across multiple modules.
How Normalization Works
The platform stores original source artifacts and then produces canonical fields from those sources so pages can be sorted, filtered, graphed, and compared without relying on raw prose.
When a value cannot be normalized reliably, the platform prefers explicit gaps or review flags over pretending the source is cleaner than it is.
FAQ
What makes this different from a simple dataset mirror?
The platform keeps the original source artifacts, but it also creates stable entity pages, canonical facts, shared search documents, and cross-domain links so people can use the data directly.
Does every module use the same data model?
No. Each module has its own domain-specific schema, but shared platform tables handle ingest runs, source files, document assets, and cross-domain search.
Is the platform trying to replace source documents?
No. The goal is to make the source material easier to discover, compare, and analyze while preserving the original artifacts and provenance.
Will more modules be added?
Yes. The current implementation is intentionally modular so new domains can reuse the same search, URL, SEO, and normalization patterns.