Sources and licences
All five sources are works of the United States government and are in the public domain (17 USC §105). None requires an attribution notice; they are credited here and in the footer of every page.
| Source | What is used | Version in this build |
|---|---|---|
| CFPB Consumer Complaint Database, bulk CSV export | Date received, product, sub-product, issue, sub-issue, company, state, tags, company response, timeliness, company public response category. 18,091,520 complaints. The narrative column is never read. | Complaints through 30 Sep 2026; downloaded 30 Sep 2026 |
| FDIC BankFind Suite, institutions API | Name, certificate number, total assets and holding-company identifier of every active insured bank | Report date 30 Jun 2026 |
| NCUA quarterly call report data | Name, charter number, total assets and members of every federally insured credit union | Cycle 30 Jun 2026 |
| FFIEC HMDA transmittal sheet | Respondent name and the number of loan and application records each lender filed | Activity year 2025 |
| US Census Bureau population estimates (downloaded file, not the API) | Resident population of each state and the nation | Vintage 2025 |
What the CFPB says about its data
When looking at complaint volume about a company or product, consider company size and/or market share. For example, companies with more customers may have more complaints than companies with fewer customers.
On the same page the Bureau states that the database “is not a statistical sample of consumers' experiences in the marketplace”, that complaints are published “after the company responds, confirming a commercial relationship with the consumer, or after 15 days, whichever comes first”, and that it does not publish complaints referred to other regulators, “such as complaints about depository institutions with less than $10 billion in assets”. The Bureau does not verify every allegation in a complaint before publishing it. See the Bureau's database page for its full description.
Sources considered and not used
The CFPB's search API returns the same records as the bulk file; the bulk file is used because a single daily download is lighter on the Bureau's servers than thousands of queries. No commercial, scraped or licence-restricted data is used anywhere on the site.