Guides
API Reference

Watchlists & Sanctions: Standalone Checks

Standalone Sanctions & PEP Checks

Baselayer offers dedicated endpoints for running sanctions and PEP screening independently from Business Search. These are useful when you need to check individual names, run batch screening, or build custom compliance workflows outside of the standard business verification flow.

For watchlist screening as part of Business Search, see the Watchlists & Sanctions: Business Search guide. For a deeper understanding of sanctions compliance concepts, see the Watchlists, Sanctions & Compliance Screening guide.

When to Use Standalone Checks

Standalone sanctions and PEP endpoints are designed for workflows where you need screening outside of a full Business Search:

  • Individual KYC — Screening a person's name (e.g., a beneficial owner or authorized signer) without running a full business verification
  • Batch processing — Screening many names at once by uploading a CSV file, such as during periodic rescreening of an existing customer base
  • Re-screening existing customers — Running updated checks against current sanctions lists for customers already onboarded
  • Custom compliance workflows — Integrating sanctions checks at specific points in your application flow (e.g., before disbursing funds, during periodic reviews)
  • Pre-screening — Quick name checks before initiating a full Business Search

Available Endpoints

OFAC Sanctions Check

POST /ofac/searches

Checks a name against OFAC sanctions lists (SDN, SSI, and other OFAC-administered lists).

POST /ofac/searches
{
  "name": "John Smith",
  "similarity_threshold": 0.85
}

PEP Check

POST /pep/searches

Checks a name against Politically Exposed Persons databases to identify individuals holding or having held prominent public positions.

POST /pep/searches
{
  "name": "Jane Doe",
  "similarity_threshold": 0.85
}

The Similarity Threshold

Both standalone endpoints accept a similarity_threshold parameter that controls how closely a name must match a watchlist entry to be flagged.

How It Works

The threshold is a value between 0.0 and 1.0 (0% to 100% similarity). A name must meet or exceed the threshold to be returned as a match.

  • 0.9 (90%) — High precision. Only very close name matches are returned. Fewer false positives, but may miss matches with significant name variations.
  • 0.85 (85%) — Balanced. A good starting point for most use cases.
  • 0.7–0.8 (70–80%) — High recall. Catches more name variations (transliterations, abbreviations, typos) but generates more false positives requiring manual review.

Choosing a Threshold

The default for standalone checks is 0.9 (90%) if no similarity_threshold is specified. This provides high precision with fewer false positives. Adjust based on your results:

  • If you're seeing too many false positives (common names flagging unrelated watchlist entries), raise the threshold toward 0.95.
  • If you're in a highly regulated industry and want maximum coverage, lower it toward 0.7 — but expect more manual review.

Note: Business Search uses lower default thresholds (0.7 for standard watchlists, 0.8 for PEP) to ensure comprehensive coverage during business verification. Standalone checks default to 0.9 but give you explicit control over this parameter per request, allowing you to tune sensitivity for different workflows.

Reading the Response

Standalone check responses return match results directly.

Key Fields

  • total_matches — Number of matches found. A value of 0 means no matches.
  • matches — Array of individual match records, each containing name, type, sanctions program, addresses, and alternative names.

Identifying Matches

Any response where total_matches > 0 indicates a positive match requiring review. Examine the matches array to understand what matched and assess whether it's a true positive or false positive.

Implementation Examples

OFAC Sanctions Check

import requests

def check_ofac(name, similarity_threshold=0.85):
    """
    Perform an independent OFAC sanctions check.
    """
    url = "https://api.baselayer.com/ofac/searches"
    headers = {
        "X-API-Key": "your-api-key",
        "Content-Type": "application/json"
    }
    payload = {
        "name": name,
        "similarity_threshold": similarity_threshold
    }
    
    response = requests.post(url, json=payload, headers=headers)
    response.raise_for_status()
    
    data = response.json()
    total_matches = data.get('total_matches', 0)
    
    return {
        'has_hit': total_matches > 0,
        'hit_count': total_matches,
        'hits': data.get('matches', [])
    }

# Usage
result = check_ofac("John Doe", similarity_threshold=0.85)
if result['has_hit']:
    print(f"⚠️ OFAC hit: {result['hit_count']} match(es)")
    for hit in result['hits']:
        print(f"  - {hit['name']} (Program: {hit.get('program')})")
else:
    print("✅ No OFAC matches")

PEP Check

def check_pep(name, similarity_threshold=0.85):
    """
    Perform an independent PEP check.
    """
    url = "https://api.baselayer.com/pep/searches"
    headers = {
        "X-API-Key": "your-api-key",
        "Content-Type": "application/json"
    }
    payload = {
        "name": name,
        "similarity_threshold": similarity_threshold
    }
    
    response = requests.post(url, json=payload, headers=headers)
    response.raise_for_status()
    
    data = response.json()
    total_matches = data.get('total_matches', 0)
    
    return {
        'has_hit': total_matches > 0,
        'hit_count': total_matches,
        'hits': data.get('matches', [])
    }

# Usage
result = check_pep("Jane Doe", similarity_threshold=0.85)
if result['has_hit']:
    print(f"⚠️ PEP hit: {result['hit_count']} match(es)")
    for hit in result['hits']:
        print(f"  - {hit['name']} (Position: {hit.get('title')})")
else:
    print("✅ No PEP matches")

Batch OFAC Screening

For screening many names at once — such as periodic rescreening of an existing customer base — Baselayer provides a dedicated batch endpoint that accepts a CSV file.

POST /ofac/searches/batches

Submit a CSV file (max 5MB) with two columns: entity_name and reference_id. The reference_id is your internal identifier for each entity, allowing you to match results back to your records.

Sample CSV: here

entity_name,reference_id
Alice Johnson,CUST-001
Bob Chen,CUST-002
Carlos Rivera,CUST-003

Request:

import requests

def batch_ofac_screen(csv_file_path, similarity_threshold=0.9):
    """
    Submit a CSV for batch OFAC screening.
    """
    url = "https://api.baselayer.com/ofac/searches/batches"
    headers = {
        "X-API-Key": "your-api-key",
        "accept": "application/json"
    }
    params = {"similarity_threshold": similarity_threshold}
    
    with open(csv_file_path, "rb") as f:
        files = {"file": f}
        response = requests.post(url, headers=headers, params=params, files=files)
    
    response.raise_for_status()
    return response.json()

# Usage
result = batch_ofac_screen("customers.csv", similarity_threshold=0.9)
print(f"Batch ID: {result['id']}")
print(f"State: {result['state']}")
print(f"Total entities: {result['entities_count']}")
print(f"Entities with matches: {result['entities_with_match_count']}")

Response:

The API returns a summary of the batch analysis:

{
  "id": "fce8c776-163d-4c28-bd3d-cc2797b4f9af",
  "name": "customers.csv",
  "similarity_threshold": 0.9,
  "state": "COMPLETED",
  "entities_count": 100,
  "valid_entities_count": 100,
  "entities_with_match_count": 10,
  "invalid_entities_count": 0,
  "created_at": "2026-03-31T23:48:20.569048",
  "updated_at": "2026-03-31T23:48:20.569059"
}

The similarity_threshold parameter defaults to 0.9 and can be adjusted per batch via query parameter.

Best Practices

Use standalone checks for individual screening. If you're verifying a person (not a business), standalone endpoints are the right tool. Don't run a full Business Search just to check a name.

Match your threshold to your risk tolerance. Start at 0.85 and adjust. Highly regulated industries may want 0.7; low-risk use cases may prefer 0.9.

Always review match details. A positive match doesn't always mean a true hit. Examine names, addresses, and sanctions programs to distinguish true positives from false positives caused by common names.

Log everything. Record the name checked, threshold used, timestamp, and full response. Regulators may require proof of screening.

Combine with Business Search. Standalone checks are complementary to — not a replacement for — Business Search watchlist screening. Use standalone checks for individual names and edge cases; use Business Search for full business verification workflows.

Set up ongoing monitoring. Standalone checks are point-in-time. For continuous screening, use Portfolio Monitoring to automatically rescreen your customer base when watchlists are updated.

Next Steps

Updated 5 days ago