Pricing Optimization
“For each product-customer pair, will they purchase at the current price in the next 14 days?”
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 product-customer pair, will they purchase at the current price in the next 14 days?
Traditional pricing optimization relies on broad A/B tests, segment-level elasticity estimates, or manual competitive analysis. These approaches treat all customers in a segment identically — but a price-insensitive enterprise buyer and a deal-seeking SMB customer respond to the same $50 price point very differently. Running A/B tests at the customer-product level is impractical at scale, and most teams default to one-size-fits-all pricing that leaves 15-25% of potential revenue on the table. For a $100M product line, that represents $15-25M in unrealized revenue annually.
How KumoRFM solves this
Relational intelligence for revenue growth
Kumo learns from the full relational graph — customer purchase history, product attributes, price sensitivity signals, and behavioral similarities across the customer-product network — to predict individual purchase probability at any price point. The model discovers that Customer C-5501 (segment: enterprise, low price sensitivity) has 91% purchase probability at $75 while Customer C-5502 (segment: SMB, high sensitivity) drops to 23% at the same price, enabling targeted discounting that maximizes revenue without blanket markdowns.
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 | segment | price_sensitivity |
|---|---|---|
| C-5501 | Enterprise | Low |
| C-5502 | SMB | High |
| C-5503 | Mid-Market | Medium |
PRODUCTS
| product_id | name | category | price |
|---|---|---|---|
| P-101 | Analytics Pro | Software | $75 |
| P-102 | Data Connector | Integration | $120 |
| P-103 | Starter Pack | Software | $25 |
PURCHASES
| purchase_id | customer_id | product_id | price_paid | timestamp |
|---|---|---|---|---|
| PUR-801 | C-5501 | P-101 | $75 | 2025-01-03 |
| PUR-802 | C-5502 | P-103 | $20 | 2025-01-05 |
| PUR-803 | C-5503 | P-102 | $108 | 2025-01-09 |
| PUR-804 | C-5501 | P-102 | $120 | 2025-01-12 |
Write your PQL query
Describe what to predict in 2–3 lines — Kumo handles the rest
PREDICT COUNT(PURCHASES.*, 0, 14, days) > 0 FOR EACH CUSTOMERS.CUSTOMER_ID WHERE PRODUCTS.PRICE > 50
Prediction output
Every entity gets a score, updated continuously
| CUSTOMER_ID | TIMESTAMP | TARGET_PRED | True_PROB |
|---|---|---|---|
| C-5501 | 2025-02-01 | True | 0.91 |
| C-5502 | 2025-02-01 | False | 0.23 |
| C-5503 | 2025-02-01 | True | 0.67 |
Understand why
Every prediction includes feature attributions — no black boxes
Customer C-5501 × Product P-101
Predicted: True (89% purchase probability)
Top contributing features
Historical purchase frequency (90d)
4 purchases
34% attribution
Price sensitivity segment
Low
26% attribution
Product category affinity
Software (3 prior)
20% attribution
Similar-customer conversion rate
88% at $75
14% attribution
Days since last purchase
19 days
6% 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 predicts individual purchase probability at any price point from relational customer, product, and transaction data — enabling precision pricing at scale without exhaustive A/B testing.
Related use cases
Explore more growth use cases
Topics covered
