Basket Size Optimization
“Which customers will either spend over $500 or place more than 15 orders in the next 30 days?”
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
Which customers will either spend over $500 or place more than 15 orders in the next 30 days?
"High-value" means different things — some customers spend big on a few orders, others make many smaller purchases. Traditional models predict either total spend or order frequency, but never both in a single pass. This forces teams to maintain two separate models, reconcile conflicting signals, and manually define thresholds that miss edge cases. A customer who places 20 orders at $30 each ($600 total) and one who makes a single $800 purchase both deserve premium treatment, but single-metric models catch only one pattern. The OR condition captures both in a unified prediction.
How KumoRFM solves this
Relational intelligence for revenue growth
Kumo's PQL supports compound conditions natively — PREDICT SUM > threshold OR COUNT > threshold — so both high-spend and high-frequency patterns are captured in a single query. The graph transformer learns from the full relational structure of customers and orders, discovering that Customer C-1042 is trending toward both thresholds (True_PROB 0.94) while C-3391 is unlikely to hit either (True_PROB 0.12). No need to build, maintain, or reconcile two separate models.
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 | tier | signup_date |
|---|---|---|---|
| C-1042 | Meridian Holdings | Platinum | 2023-03-15 |
| C-2187 | J. Vasquez | Gold | 2024-01-08 |
| C-3391 | NovaTech Inc. | Silver | 2024-06-22 |
ORDERS
| order_id | customer_id | value | timestamp |
|---|---|---|---|
| ORD-4401 | C-1042 | $285 | 2025-01-03 |
| ORD-4402 | C-1042 | $142 | 2025-01-05 |
| ORD-4403 | C-2187 | $38 | 2025-01-04 |
| ORD-4404 | C-2187 | $22 | 2025-01-06 |
| ORD-4405 | C-3391 | $15 | 2025-01-10 |
Write your PQL query
Describe what to predict in 2–3 lines — Kumo handles the rest
PREDICT SUM(ORDERS.VALUE, 0, 30, days) > 500 OR COUNT(ORDERS.*, 0, 30, days) > 15 FOR EACH CUSTOMERS.CUSTOMER_ID
Prediction output
Every entity gets a score, updated continuously
| CUSTOMER_ID | TIMESTAMP | TARGET_PRED | True_PROB |
|---|---|---|---|
| C-1042 | 2025-02-01 | True | 0.94 |
| C-2187 | 2025-02-01 | True | 0.71 |
| C-3391 | 2025-02-01 | False | 0.12 |
Understand why
Every prediction includes feature attributions — no black boxes
Customer C-1042
Predicted: True (94% high-value probability)
Top contributing features
Rolling 30-day order value
$427 (trending up)
36% attribution
Order frequency (30d)
11 orders
28% attribution
Average order value trend
+18% MoM
17% attribution
Account tier
Platinum
12% attribution
Similar-customer spend pattern
Top 3% cohort
7% 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: Kumo's compound OR prediction captures both high-spend and high-frequency customers in a single query — no separate models, no manual threshold reconciliation. Marketing teams can target all high-value patterns with one unified prediction.
Related use cases
Explore more growth 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.




