Omnichannel Journey Optimization
“For each customer, which fulfillment pathway (store pickup, delivery, ship-to-store) will they prefer for their next order?”
Book a demo and get a free trial of the full platform: research 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 customer, which fulfillment pathway (store pickup, delivery, ship-to-store) will they prefer for their next order?
Retailers offer multiple fulfillment options but present them generically — defaulting to shipping even when a customer lives 2 miles from a store. Mismatched fulfillment increases last-mile costs, delivery failures, and returns. Meanwhile, BOPIS (buy online, pick up in store) drives 30% higher in-store attach rates that retailers miss by not surfacing it to the right customers.
How KumoRFM solves this
Relational intelligence for optimal actions
Kumo connects customers, orders, and stores into a relational graph that captures fulfillment preferences, location proximity, order history, and cross-channel behavior. The multi-class prediction assigns each customer the most likely fulfillment type for their next order — enabling personalized checkout defaults that reduce friction and maximize attach revenue.
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 | location | segment |
|---|---|---|---|
| C-5001 | Maria Santos | Brooklyn, NY | urban-loyal |
| C-5002 | James Wright | Plano, TX | suburban-new |
| C-5003 | Priya Sharma | San Jose, CA | urban-occasional |
| C-5004 | Tom Baker | Rural, MT | rural-loyal |
| C-5005 | Lin Wei | Chicago, IL | urban-loyal |
ORDERS
| order_id | customer_id | fulfillment_type | amount | timestamp |
|---|---|---|---|---|
| ORD-701 | C-5001 | store_pickup | $89 | 2026-02-10 |
| ORD-702 | C-5001 | store_pickup | $145 | 2026-02-25 |
| ORD-703 | C-5002 | delivery | $62 | 2026-02-15 |
| ORD-704 | C-5003 | delivery | $210 | 2026-01-20 |
| ORD-705 | C-5004 | delivery | $95 | 2026-02-28 |
| ORD-706 | C-5005 | ship_to_store | $178 | 2026-03-01 |
STORES
| store_id | name | location | type |
|---|---|---|---|
| STR-01 | Brooklyn Heights | Brooklyn, NY | flagship |
| STR-02 | Legacy West | Plano, TX | standard |
| STR-03 | Santana Row | San Jose, CA | premium |
| STR-04 | Magnificent Mile | Chicago, IL | flagship |
Write your PQL query
Describe what to predict in 2–3 lines — Kumo handles the rest
PREDICT ORDERS.FULFILLMENT_TYPE FOR EACH CUSTOMERS.CUSTOMER_ID
Prediction output
Every entity gets a score, updated continuously
| CUSTOMER_ID | store_pickup | delivery | ship_to_store | PREDICTED |
|---|---|---|---|---|
| C-5001 | 0.84 | 0.11 | 0.05 | store_pickup |
| C-5002 | 0.22 | 0.68 | 0.10 | delivery |
| C-5003 | 0.35 | 0.52 | 0.13 | delivery |
| C-5004 | 0.03 | 0.91 | 0.06 | delivery |
| C-5005 | 0.28 | 0.15 | 0.57 | ship_to_store |
Understand why
Every prediction includes feature attributions — no black boxes
Customer C-5001 (Maria Santos)
Predicted: store_pickup (0.84)
Top contributing features
Last 2 orders were store pickup (ORDERS)
2 of 2 pickup
37% attribution
Lives 0.8 mi from Brooklyn Heights store (STORES)
0.8 miles
28% attribution
Segment = urban-loyal, 85% pickup preference (graph)
85% peer pickup
20% attribution
Avg order value aligns with in-store shoppers (ORDERS)
$117 avg
15% 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: Personalize fulfillment defaults at checkout. Increase BOPIS adoption 20-30%, reduce last-mile delivery costs by $3-6 per order, and drive 15% higher in-store attach revenue — worth $5-12M annually for a mid-size retailer.
Related use cases
Explore more next-best-action use cases
Topics covered
