TechStack Data

Public-record intelligence

Who controls it, what changed, and how we know.

TechStack LLC builds data products that resolve ownership, control and counterparty risk out of United States government filings. We publish the answer, the filing it came from, the date of that filing, and how confident the evidence lets us be. Where the sources genuinely disagree, we say so rather than choosing.

Products

Each product answers one question for one kind of decision. They share a single evidence standard, a single confidence vocabulary and a single corrections policy, so a finding means the same thing wherever you encounter it.

WhoOwnsCare

Who actually controls this Medicare-certified provider, and what changed?

Post-acute ownership, transaction and counterparty intelligence for skilled nursing, home health and hospice.

live

Business Buyer Check

Does this business-for-sale listing hold together well enough to spend diligence money on?

Evidence-led screening for small-business acquisitions: what the listing claims, what the deal math says, and what still needs proof.

live

Full product directory, with coverage and pricing

One evidence standard across every product

Most public-records tools present a retrieved value and a calculated value in the same typeface, which quietly invites you to trust both equally. We don't. Every value we publish is assigned one of five evidence classes, and the class travels with the value wherever it appears.

Class What it means
Primary filing Taken directly from a government file as submitted by the filer. We reproduce it; we do not verify that the filer was truthful.
Resolved Derived by linking two or more primary filings through shared identifiers, shared officers, declared parent relationships or portfolio containment. The linking rule is published and the inputs are cited.
Calculated Computed from primary or resolved values using a published formula. The inputs and assumptions are shown alongside the result.
Claimed Asserted by an interested party and not independently supported. It stays labelled as a claim until a document supports it. A claim never silently becomes a fact.
Unresolved Missing, stale, or in genuine conflict between sources. We say so rather than choosing. Unresolved evidence lowers coverage and generates a specific follow-up question.

How the standard is applied · What "Verified" actually requires

The sources behind the products

Everything we publish is built from public government files. We catalogue each one with its publisher, its licence, the edition currently in use, the grain of a row, and the limits that make it easy to misread. The limits matter as much as the fields: most errors made with these files come from treating a lender as an operator, or an enrolled entity as an owner.

Full source catalogue

Why this is published at all

A finding you cannot trace is a rumour with a citation style. Publishing the method, the source list and the failure modes is what separates an intelligence product from an aggregator, and it is the only way a reader can disagree with us specifically rather than generally.

If you believe something we have published is wrong, our corrections policy commits us to a logged, answered response within two business days.

Research

Explainers on reading government filings correctly, and analysis that spans more than one product. This is also where we test a question before it becomes a product.

All research

Data licensing

Resolved datasets are available for licensing where the underlying sources permit redistribution. Licensing terms and what is available, or email data@techstackllc.io.