Overview

Red Planet Data built and operates Atlas, a data engine that sources, normalizes, verifies and continuously improves its own record of property, transactions, distress, jurisdictional decisions and infrastructure. Every Red Planet product runs on it, and firms can license the data directly.

Why We Built It

Real estate and the infrastructure being built on top of it are among the largest asset classes in the world, and the data underneath them is fragmented, expensive and sourced the way it was twenty years ago. Parcel records, transaction history, distress filings, entity ownership, permits and utility plans live in separate systems sold by separate vendors in formats that do not join. The largest providers price for the largest institutions, deliver fixed formats for fixed questions, and refresh monthly or quarterly.

There is a second gap underneath that one. The decisions that determine what a piece of land is worth are made in a county commission meeting, an ordinance amendment, a comprehensive plan update or a utility's capital plan. That record exists in the jurisdiction's own documents, but it is scattered across hundreds of agenda portals and minutes archives and reconciled by nobody. So the market runs on secondhand reports: a tracker says a county passed a moratorium when the county passed a first reading of a draft that later died in committee. Anyone underwriting against that is underwriting against hearsay.

Atlas closes both gaps. It holds 306,408,066 verified records drawn from 319 active sources, and every consequential claim in the jurisdictional record is tied to the document it came from.

What Atlas Is

Atlas is an engine rather than a database. A database holds what you put in it. Atlas decides what it needs, goes and gets it, checks it against the document it came from, and makes it queryable. It identifies gaps in its own coverage, queues new sources to fill them, and monitors its own health. It runs every night whether or not anyone is watching.

Built Directly, by Rule

We do not buy from ATTOM, CoreLogic, DataTree, Black Knight or any aggregator that resells someone else's collection, and we do not buy third-party vendor datasets. This is a standing rule rather than a budget decision. A data asset assembled from the same vendors as everyone else's is a subscription, and a subscription can be cancelled. What Atlas holds instead comes from going to each jurisdiction and system directly: hundreds of source relationships, thousands of normalization decisions, and a verification layer that ties consequential claims back to the supporting document.

Much of It Cannot Be Collected Again

Most county and city systems show only the current version of a record. When a filing is satisfied, withdrawn or superseded, the earlier version is overwritten rather than archived, because administration only needs to know what is true today. Atlas captures records every night and keeps each version, so the asset holds a history of what a record said over time, including states that no longer exist anywhere a buyer can reach.

Take a single Hartford property. A lien, then a foreclosure, then a withdrawal, then a transfer. Several of those filings are no longer visible on the county's own systems, and the sequence is what tells you what happened to the owner. A competitor with more capital can buy a parcel file tomorrow. Nobody can buy the last two years of overwritten filings, because they were not retained. That gap widens every night we run.