Screening Alerts Configurations

Last updated: June 23, 2025

1. Overview: Enterprise-Grade Configurability

Greenlite's Alerts AI product is designed to be highly configurable to match the Standard Operating Procedures (SOPs) and standards of any compliance organization. Our system adapts to your specific requirements rather than forcing you to change your processes.

Proven Track Record

  • Successfully deployed at OCC-regulated banks

  • Trusted by large FinCEN-regulated fintechs

  • Auditor-tested and approved across multiple examinations

  • Configured to meet diverse regulatory requirements

Greenlite works by replicating the decision-making flow your analysts already follow, using a sophisticated tagging system that can be customized to match your organization's exact compliance requirements and risk appetite.

2. Default Match Tag/Mismatch Tag Logic

Greenlite's default logic evaluates the following types of factors:

Match Tags (Qualifying Factors):

  • Name matching (with handling for variations, nicknames, transliterations)

  • Date of birth matching (full and partial)

  • Geographic matching (full address, country, region, city)

  • Employment matching

Mismatch Tags (Disqualifying Factors):

  • Name mismatch

  • Date of birth mismatch

  • Geographic mismatches (country, region, city)

  • Employment mismatch

  • Entity type mismatch

All tag weights are fully configurable to meet each organization's specific requirements and risk appetite. The scoring starts with a base score of 5 and calculates a final True Positive likelihood on a 0-10 scale.

3. Custom Match Tag/Mismatch Logic

Screening Alert Type Logic

Configure handling based on alert type:

  • PEP-Specific: Example: PEP Out of Scope - mark specific types of PEPs as false positives or low risk

  • Adverse Media: Custom handling for negative news alerts

  • Sanctions: Specialized rules for sanctions list matches

Source-Based Logic

  • Perform false positive whitelisting for highly noisy sources

  • Example: Automatically mark alerts from the Chinese Credit Bureau as false positives

Entity Type Based Logic

  • Different handling for Company vs Person alerts

  • Separate scoring and thresholds by entity type

4. Evaluation Summary Configuration

Greenlite allows teams to configure the evaluation summary output of alert cases.

Example Summary:

Customer John Doe (DOB: 05/15/1985) flagged as a possible match against 
John Doe - PEP List (YOB: 1985). The investigation resulted in a determination of Name Match, DOB Match, and Country Mismatch. 
The investigation resulted in a False Positive disposition.

Customization Options

  1. Regulatory formatting: Match specific regulatory report requirements

  2. Language customization: Adjust terminology to match your SOPs

  3. Dynamic content: Include/exclude sections based on match type

5. Analyst Workflow Customization

Greenlite adapts to your existing analyst workflows:

Closure Options

  • Auto-Closure: Greenlite closes false positives directly in your case management tool/screening alerts provider

  • Hybrid: Greenlite closes specific alerts that are considered lower risk (e.g., just PEPs)

  • Manual Review: Human closes out all the hits individually

  • Batch Closure: Human closes out all the hits for a specific case at once

Integration Flexibility

  • API-based closure: Direct integration with your screening system

  • Escalation paths: Define review hierarchies

6. Research Relevancy Thresholds

Configure how external research is incorporated into investigations:

OSINT Relevance Scoring

A prompt evaluates the relevancy of OSINT information on a scale from 1 to 5:

  • 1. Irrelevant or Misleading: Incorrect, outdated, or unrelated information. These are filtered out by default.

  • 2. Low Relevance: Adds little value, only confirming well-established facts.

  • 3. Moderately Relevant: Contains useful information supporting known facts or introducing tangentially related information.

  • 4. Highly Relevant: Provides significant, directly applicable information to the investigation.

  • 5. Critical Source: Contains crucial information that could significantly impact or redirect the investigation.

Threshold Configuration

  • Global threshold: Set minimum relevancy for all research findings

    • Default Threshold of >= 4 for all customers

AI Findings
Source: www.lawfact.co.kr
Relevance: 4/5 (Highly Relevant)
- Served in Constitutional Court as legislative researcher in 2017
- Ministry of Government Legislation positions since 1991

Implementation Note

Available configurations depend on your screening system's capabilities and API functionality. Not all screening systems support all configuration options.

To customize your Greenlite implementation:

  1. Review this documentation with your compliance team

  2. Identify desired configurations matching your SOPs

  3. Discuss technical feasibility with your deployment strategist

Contact your Greenlite deployment strategist to begin the configuration process tailored to your organization's specific needs.