Household Mapping
“For each individual customer, which other customers belong to the same household?”
Book a demo and get a free trial of the full platform: data science agent, fine-tune capabilities, and forward-deployed engineer support.
By submitting, you accept the Terms and Privacy Policy.
Loved by data scientists, ML engineers & CXOs at
A real-world example
For each individual customer, which other customers belong to the same household?
Household-level targeting can increase campaign ROI 2-3x, but identifying households requires more than address matching. Multiple family members may have different addresses, last names, or use different devices. Kumo identifies household relationships through shared shipping addresses, devices, payment methods, and behavioral patterns. Without household mapping, marketers send redundant offers to the same household, inflating acquisition costs by 20-40%.
How KumoRFM solves this
Relational intelligence for identity resolution
Kumo builds a relational graph connecting customers to their orders, shipping addresses, shared devices, and payment methods. Instead of grouping by address hash alone, Kumo learns that Customer C-110 and Customer C-445 share a shipping address, use the same device fingerprint on weekends, and have complementary purchasing patterns (children's clothing + adult clothing). The graph captures household signals that address-only matching misses — different last names, P.O. boxes, and multi-generational households all become visible.
From data to predictions
See the full pipeline in action
Connect your tables, write a PQL query, and get predictions with built-in explainability — all in minutes, not months.
Your data
The relational tables Kumo learns from
CUSTOMERS
| customer_id | name | address_hash | |
|---|---|---|---|
| C-110 | James Rivera | ADDR-7F2A | jrivera@email.com |
| C-445 | Ana Rivera-Lopez | ADDR-7F2A | ana.rl@gmail.com |
| C-820 | Tom Baker | ADDR-3B91 | tbaker@corp.com |
ORDERS
| order_id | customer_id | shipping_address | amount | timestamp |
|---|---|---|---|---|
| ORD-3001 | C-110 | 789 Pine Rd, Apt 4B | $142.50 | 2025-09-12 |
| ORD-3002 | C-445 | 789 Pine Rd, Apt 4B | $89.99 | 2025-09-13 |
| ORD-3003 | C-820 | 55 Maple Dr | $215.00 | 2025-09-14 |
SHARED_DEVICES
| device_id | customer_id | device_fingerprint | timestamp |
|---|---|---|---|
| SD-001 | C-110 | FP-8A2C | 2025-09-12 20:15 |
| SD-002 | C-445 | FP-8A2C | 2025-09-13 09:30 |
| SD-003 | C-820 | FP-D41E | 2025-09-14 11:45 |
Write your PQL query
Describe what to predict in 2–3 lines — Kumo handles the rest
PREDICT LIST_DISTINCT(ORDERS.CUSTOMER_ID, 0, 30, days) FOR EACH CUSTOMERS.CUSTOMER_ID WHERE CUSTOMERS.ADDRESS_HASH = CUSTOMERS.ADDRESS_HASH
Prediction output
Every entity gets a score, updated continuously
| CUSTOMER_ID | HOUSEHOLD_MEMBER | SCORE | TIMESTAMP |
|---|---|---|---|
| C-110 | C-445 | 0.96 | 2025-10-01 |
| C-820 | C-821 | 0.88 | 2025-10-01 |
| C-302 | C-519 | 0.74 | 2025-10-01 |
Understand why
Every prediction includes feature attributions — no black boxes
Customer C-110 (James Rivera)
Predicted: 96% household match with C-445 (Ana Rivera-Lopez)
Top contributing features
Shared shipping address
789 Pine Rd, Apt 4B
30% attribution
Shared device fingerprint (FP-8A2C)
Same device
28% attribution
Complementary purchase categories
Adult + Kids
20% attribution
Order timing correlation
Same weekend
13% attribution
Payment method proximity
Same bank
9% attribution
Feature attributions are computed automatically for every prediction. No separate tooling required. Learn more about Kumo explainability
PQL Documentation
Learn the Predictive Query Language — SQL-like syntax for defining any prediction task in 2–3 lines.
Python SDK
Integrate Kumo predictions into your pipelines. Train, evaluate, and deploy models programmatically.
Explainability Docs
Understand feature attributions, model evaluation metrics, and how to build trust with stakeholders.
Bottom line: Identify household relationships that address matching misses — increasing campaign ROI 2-3x and eliminating 20-40% redundant acquisition spend.
Related use cases
Explore more entity resolution use cases
Topics covered
One Platform. One Model. Predict Instantly.
KumoRFM
Relational Foundation Model
Turn structured relational data into predictions in seconds. KumoRFM delivers zero-shot predictions that rival months of traditional data science. No training, feature engineering, or infrastructure required. Just connect your data and start predicting.
For critical use cases, fine-tune KumoRFM on your data using the Kumo platform and Data Science Agent for 30%+ higher accuracy than traditional models.
Book a demo and get a free trial of the full platform: data science agent, fine-tune capabilities, and forward-deployed engineer support.
