Methodology

How PRISM data is collected and structured.

PRISM is a benchmarking dataset built from active and historical rental listings that landlords advertised publicly, plus public HUD records. This page describes the pipeline so appraisers, lenders, and data teams can cite it in their workfiles.

Sources

Public rental listing sites (major landlord sites, MLS-hosted rentals, affordable and senior housing aggregators, regional listing sites) and HUD public records. PRISM does not use subscriber rent rolls or lease transaction data, and no subscription is conditioned on providing data.

Collection cadence

Most sources are collected weekly, with runs staggered across the week. Each observation is timestamped, and the source map records which listing contributed each value and when it was observed.

Deduplication

Listings for the same property across multiple sites are matched on normalized address, geocode, and unit attributes and collapsed into one property record with one PRISM property ID. More than 1.5 million listings resolve to more than one million properties (CRED iQ registry, September 2026).

Unit-type structuring

Asking rents are assigned to a unit type (studio, 1 bed, 2 bed, 3 bed, 4 bed) from the listing's bedroom count. For each property and unit type PRISM reports the minimum, maximum, and midpoint asking rent, square footage where published, and the count of listed units.

Market analytics

City and unit-type distributions report median, interquartile range (P25 to P75), P10 to P90, average square footage, and rent per square foot. Aggregates are anonymized and require a minimum listing count before they are shown.

What PRISM does not do

PRISM does not issue pricing recommendations, suggest rents, or tailor outputs to a subscriber. It reports what the market is publicly asking.

Exclusions and known limits

Last updated September 16, 2026 · Back to PRISM · www.cred-iq.com · Team@cred-iq.com