Every U.S. credit union's NCUA call report, ten years, one place.
CU Explorer turns the quarterly filings of 4,299 federally insured credit unions into something a board member, an executive or an analyst can read in minutes: one credit union's statements and history, how it compares with its peers, and the shape of the whole industry.
What is inside
- InSights: capital ratio — Every credit union at or below 9% net worth as one circle, packed into 1% bins, green at or above NCUA's 7.00% well-capitalized line and pink below, sized by members, assets or member growth. 460 credit unions sit at or below 9% in Q2 2026; 55 are below 7%. Open →
- Treemap of every credit union — All 4,299 credit unions as boxes sized by assets, grouped by asset band, sponsor type or state, and coloured green to red by loan growth or any other metric, as of Q2 2026. Open →
- Industry trends — The credit union loan book since 2018: the loan mix and how it has shifted, pricing by loan type, the real estate term structure, and aggregates by asset band, through Q2 2026. Open →
- Segments: who each credit union serves — The 4,299 credit unions sorted into 8 sponsor types, from community charters to military and educational, each opening into the credit unions that make it up. Open →
- CU AI Ready Report — An AI Score from one to five stars for every credit union, every star above the first linked to the public evidence behind it: 272 credit unions with public evidence as of 2026-09-23, from their own websites, vendor announcements and news coverage. Open →
- One credit union, compared with its peers — Any of the 4,299 credit unions: statements, ten-year charts, percentile rankings against asset-size peers or a peer set you build, leaders, branches and growth opportunities, Q1 2016 to Q2 2026. Open →
A search box on every page finds any credit union by name, charter number, city or state. Each credit union also has a plain page with its latest figures, one per state, for people who arrive from a search engine.
The data, and how it is treated
Everything comes from the NCUA 5300 Call Report, the quarterly filing every federally insured credit union makes, as published in NCUA's bulk quarterly data: Q1 2016 through Q2 2026, 42 quarters. Ratios, growth rates, peer percentiles, segment profiles and industry aggregates are the author's calculations on those filings. They are independent analysis, not an NCUA publication, and the site is not affiliated with the National Credit Union Administration.
Peer statistics use each credit union's asset-size group as of the latest quarter, and time-series bands use each quarter's own membership. Percentiles follow each metric's direction so that higher always reads as stronger where a direction is defined. Income-statement ratios are annualized from year-to-date filings. Sponsor types come from the filed field of membership for federal charters and, for state charters, from a text classifier trained on the federal ones; inferred assignments are marked as such.
Where the site shows a credit union's Google Maps listing, the listing is drawn live by Google's own component; the site keeps nothing from it but the listing's identifier.
The site is not a membership-eligibility finder, does not rate credit unions, and offers no financial advice.
Who built it
Jim Kyung-Soo Liew, PhD
Jim Kyung-Soo Liew, PhD, is the founder and president of SoKat, an AI consultancy in Baltimore that builds machine-learning and data products for financial institutions and U.S. federal agencies, and a professor of finance at Johns Hopkins University, where he teaches artificial intelligence, machine learning, blockchain and entrepreneurial finance.
Before academia he worked in quantitative investing at Morgan Stanley, Campbell & Company and the Carlyle Asset Management Group. He holds a BA in mathematics from the University of Chicago and a PhD in finance from Columbia University.
He built CU Explorer so that credit union leaders, boards and analysts could see their institution, its peers and the whole industry with the same rigor institutional investors expect, from public data anyone can check.
The quarterly brief
One email each quarter, when NCUA publishes the data this site is rebuilt from: what changed, in three charts, with links. Nothing else.
Using the site and its figures
The site is free to use. Figures may be quoted with attribution to CU Explorer (www.creditunionexplorer.com) and NCUA as the underlying source. For a briefing, a custom analysis or a data question, reach Jim Liew through SoKat or LinkedIn. The privacy policy explains what the site records, which is very little.