sf-agent-framework/sf-core/tasks/data-mapping.md

AI Agent Orchestration Framework for Salesforce Development - Two-phase architecture with 70% context reduction

# Data Mapping

## Purpose

Create detailed data mappings between source systems and Salesforce objects to
ensure accurate data migration and integration.

## Instructions

1. **Source System Analysis**
   - Document source data structures
   - Identify all data entities
   - Map relationships and hierarchies
   - Analyze data types and formats
   - Review business logic and calculations

2. **Target Model Design**
   - Map to Salesforce standard objects
   - Design custom object requirements
   - Define field-level mappings
   - Plan relationship structures
   - Consider platform limits

3. **Transformation Rules**
   - Define data type conversions
   - Document value transformations
   - Create picklist mappings
   - Design formula calculations
   - Plan data enrichment logic

4. **Relationship Mapping**
   - Map parent-child relationships
   - Design lookup relationships
   - Plan master-detail hierarchies
   - Handle many-to-many relationships
   - Resolve circular dependencies

5. **Data Quality Rules**
   - Define validation criteria
   - Create default value logic
   - Design duplicate handling
   - Plan error handling
   - Document exception processes

6. **Migration Sequencing**
   - Determine load order
   - Identify dependencies
   - Plan for reference data
   - Design rollback approach
   - Create validation checkpoints

## Input Requirements

- Source system data models
- Data dictionaries
- Business rules documentation
- Salesforce object model
- Integration requirements
- Data quality standards

## Output Format

- Data Mapping Specification including:
  - Object-level mapping matrix
  - Field-level mapping details
  - Transformation rules catalog
  - Relationship diagrams
  - Load sequence plan
  - Validation rules
  - Test scenarios

## Mapping Components

- **Field Mapping**: Source → Target field
- **Value Mapping**: Data transformations
- **Relationship Mapping**: Foreign key resolution
- **Business Logic**: Calculations and derivations
- **Quality Rules**: Validation and defaults

## Common Transformations

- Date format conversions
- Picklist value mappings
- Currency conversions
- Text field truncation
- Number format changes
- Boolean conversions
- Reference data lookups

## Best Practices

- Document all assumptions
- Validate with business users
- Consider performance impacts
- Plan for data volume
- Test edge cases thoroughly
- Maintain mapping versions
- Build reusable components