Householding

Householding

What is Householding?

Householding in LakeFusion is the process of identifying and grouping individuals that belong to the same household, helping organizations better understand relationships within their customer base. Whether in insurance, banking, or healthcare, Householding enables improved targeting, risk analysis, and cross-sell opportunities using Databricks-native capabilities.

Sources Integrated

LakeFusion uses structured and semi-structured data from:
  • CRM & Customer Portals

  • Billing Systems

  • Address Normalization Services

  • Subscription & Loyalty Programs

  • Public Demographic Databases

Use Case:


  1. Source Records
    Multiple customers from the same household appear separately in different systems:

  • CRM: Name: Sarah Thompson, Address: 12 Oak Street

  • Billing System: Name: S. Thompson, Address: 12 Oak St.

  • Loyalty Program: Name: Sam T., Address: 12 Oak Street

  1. Consolidated Entity Creation
    Using common attributes such as address, last name, and payment methods:

  • Sarah, S. Thompson, and Sam T. are grouped into a single household unit.

  1. Match Records

  • LLM Vector Search: Recognizes address and name variations (e.g., Oak Street = Oak St.)

  • Custom Matching Rules: Match records sharing addresses and surnames or linked financial accounts

  1. Survivorship Rules
    For conflicting data within the household:

  • Priority-Based: Trust verified billing address > CRM > loyalty program

  • Recency-Based: Use most recent household member addition

  1. Merged Household Record
    Household-level data includes:

  • Members: Sarah Thompson, Sam Thompson

  • Address: 12 Oak Street

  • Shared Subscriptions & Accounts

Bronze, Silver, and Gold Data Tiers

Provider 360 aligns with the Medallion Architecture for structured data processing and consumption:

Tier
Description
Bronze
Raw ingestion of provider data from all integrated sources
Silver
Standardized data with address normalization and deduplication
Gold
Final household profiles used for personalized marketing, underwriting, or care planning

This tiered architecture ensures data transparency, reusability, and performance across the provider lifecycle.

    • Related Articles

    • Key Features

      1. Golden Record Generation Automatically create the most accurate version of a record by consolidating data from multiple systems, applying survivorship rules, and resolving conflicts. These "golden records" ensure all downstream systems are ...