Record Linking
“For each transaction, which customer account does it belong to?”
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 transaction, which customer account does it belong to?
10-15% of transactions arrive without proper customer linkage — payment processor mismatches, guest checkouts, system migration artifacts. These "orphan" records distort LTV calculations, break attribution, and waste marketing spend. A retailer with $1B in annual revenue may have $100-150M in transactions that cannot be attributed to any customer, making segmentation, personalization, and forecasting unreliable.
How KumoRFM solves this
Relational intelligence for identity resolution
Kumo connects orphan transactions to the full relational graph of known customer transactions, merchants, card details, and timing patterns. Instead of relying on exact card-number matches, Kumo learns that Orphan Transaction T-901 shares merchant patterns, amount ranges, and timing cadence with Customer CUST-442's known transactions. The link prediction model identifies which customer account each orphan transaction most likely belongs to — resolving attribution at scale.
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
ORPHAN_TRANSACTIONS
| transaction_id | amount | merchant | card_last4 | timestamp |
|---|---|---|---|---|
| T-901 | $87.42 | Whole Foods #127 | 4821 | 2025-09-14 |
| T-902 | $234.99 | BestBuy.com | 7753 | 2025-09-15 |
| T-903 | $42.10 | Shell Station #89 | 4821 | 2025-09-15 |
TRANSACTIONS
| txn_id | customer_id | amount | merchant | timestamp |
|---|---|---|---|---|
| TXN-6001 | CUST-442 | $92.18 | Whole Foods #127 | 2025-09-10 |
| TXN-6002 | CUST-442 | $45.00 | Shell Station #89 | 2025-09-11 |
| TXN-6003 | CUST-781 | $199.99 | BestBuy.com | 2025-09-12 |
CUSTOMERS
| customer_id | name | card_last4 | |
|---|---|---|---|
| CUST-442 | Sarah Chen | sarah@email.com | 4821 |
| CUST-781 | David Park | dpark@corp.com | 7753 |
| CUST-195 | Lisa Wong | lwong@mail.com | 3390 |
Write your PQL query
Describe what to predict in 2–3 lines — Kumo handles the rest
PREDICT LIST_DISTINCT(TRANSACTIONS.CUSTOMER_ID, 0, 30, days) FOR EACH ORPHAN_TRANSACTIONS.TRANSACTION_ID
Prediction output
Every entity gets a score, updated continuously
| TRANSACTION_ID | MATCHED_CUSTOMER_ID | SCORE | TIMESTAMP |
|---|---|---|---|
| T-901 | CUST-442 | 0.93 | 2025-10-01 |
| T-902 | CUST-781 | 0.89 | 2025-10-01 |
| T-903 | CUST-442 | 0.91 | 2025-10-01 |
Understand why
Every prediction includes feature attributions — no black boxes
Orphan Transaction T-901 ($87.42, Whole Foods #127)
Predicted: 93% match with CUST-442 (Sarah Chen)
Top contributing features
Card last-4 match
4821 = 4821
30% attribution
Merchant overlap with known txns
92%
28% attribution
Transaction timing pattern
Weekly grocery
20% attribution
Amount range consistency
$40-100
13% attribution
Geographic proximity
Same ZIP
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: Resolve 10-15% of orphan transactions back to customer accounts — recovering $100M+ in attributable revenue and fixing downstream LTV and segmentation models.
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.
