CFPB Consumer Complaint Delta Monitor

Monitors the public CFPB Consumer Complaint Database API for a user-maintained watchlist of companies / products / issues and emits one dataset item (alert) per newly filed complaint matching the watchlist since the last run — "free when nothing changes."

Who it is for

Compliance, risk, market-research and consumer-advocacy teams that need to track new complaints filed against specific financial companies, products or issues. Instead of manually re-polling the CFPB complaint database on a schedule, run this Actor on a cadence and it only alerts on what is new.

What makes it different

Existing complaint monitors typically bill you on every scan even when nothing changed. This Actor keeps a watchlist plus a per-watchlist checkpoint (last checked dateReceived high-water mark plus a content fingerprint per complaint ID) in the Apify default key-value store, so repeat runs emit only genuinely new complaints matching the watchlist — a true recurring delta monitor, free when nothing changes.

Data source

Public CFPB Consumer Complaint Database — no API key, no login, no proxies. CFPB states all complaint data it publishes is freely available for anyone to use, analyze, and build on:

• https://www.consumerfinance.gov/data-research/consumer-complaints/search/api/v1/

No credentials, no proxies, no scraping of anti-bot pages, no commercial inference (no LLMs) inside the product.

Pricing (pay-per-event framing, Apify Store)

This is a FREE listing until the owner enables monetization and completes payout/KYC. Pricing shown here is README framing only; no monetization is flipped in this release candidate.

• New-complaint alert (dataset item): per-alert price — you pay only when the Actor emits a new complaint matching your watchlist.
• Nothing-changed runs: free (zero dataset items emitted).
• Actor start: standard Apify Actor start fee.

Apify retains a 20% commission on revenue; payouts are monthly and require completed identity verification (KYC).

Input

See .actor/input_schema.json. All fields optional; the default input is:

{
  "watchlist": ["Wells Fargo", "Chase", "Citibank", "Bank of America", "Capital One"],
  "lookbackDays": 30,
  "maxResults": 25
}

• watchlist: optional array of company/product/issue strings matched case-insensitively against the company, product, sub-product, issue and sub-issue of each complaint. Empty means every complaint is in scope.
• lookbackDays (1-365, default 30): how far back to look on the first run; repeat runs only emit complaints newer than the last checked state.
• maxResults (1-200, default 25): cap on alerts emitted in a single run.

Output

One dataset item per new complaint:

{
  "complaintId": "31568254",
  "dateReceived": "2026-11-01T03:56:29.000Z",
  "product": "Credit reporting or other personal consumer reports",
  "subProduct": "Credit reporting",
  "issue": "Incorrect information on your report",
  "subIssue": "Information belongs to someone else",
  "company": "TRANSUNION INTERMEDIATE HOLDINGS, INC.",
  "state": "TX",
  "zipCode": "78237",
  "timely": "Yes",
  "companyResponse": "In progress",
  "newSinceLastRun": true,
  "changeType": "new",
  "watchlistMatched": null,
  "source": "https://www.consumerfinance.gov/data-research/consumer-complaints/search/api/v1/"
}

Every field is null when it could not be parsed from the CFPB API — the Actor never guesses. changeType is "new" for a previously unseen complaint ID.

Checkpoint / delta behaviour

State (the last-checked dateReceived high-water mark plus a content fingerprint per complaint ID) is kept in the Apify default key-value store. On each run the Actor queries the CFPB API for complaints received on or after the checkpoint, sorted newest first, deduplicates by complaint ID, and only pushes previously unseen complaints matching the watchlist. Resetting the store re-emits complaints within the lookback window.

Limits

• Max 50 watchlist terms, 200 alerts per run, 365-day lookback.
• Up to 1,000 CFPB records scanned per run (per watchlist term, sorted newest first, stops at the checkpoint).
• Checkpoints persist per watchlist in the Actor key-value store.

Verification

• pytest actor/tests — deterministic unit tests (no network): input normalization, watchlist matching, fingerprinting, delta scan, dedup.
• ruff check actor/ — must be clean (0 errors).
• Default-input smoke run against the public CFPB API returns more than 0 items in under 5 minutes (publish precondition).