---
slug: "commerce-industry"
name: "Commerce & Industry"
tagline: "Customer behaviour · sales forecasting · inventory replenishment · pricing"
description: "How tabular and time-series machine learning supports customer retention, sales forecasting, inventory replenishment, and wholesale power pricing."
datasetIds: ["telco-customer-churn", "fitness-club", "spambase", "abalone", "rossmann-1d", "rossmann-1w", "rohlik-orders-1d", "epf-np", "m-dense-1d"]
canonicalUrl: "https://carlahq.eu/machine-learning/commerce-industry"
---

# Machine Learning for Commerce & Industry

> Customer behaviour · sales forecasting · inventory replenishment · pricing

## Domain Overview
Commercial machine learning should answer a concrete operating question. Who may leave? Who is likely to attend? Which message needs review? What quantity should a team plan for? Customer, subscription, booking, product, and interaction systems store these signals in structured tables.

Many customer problems use classification with a yes or no target. Pricing, lifetime value, demand, and inventory often use regression or multi-step time-series forecasting. Choosing an algorithm isn't the hard part. Teams must define an outcome they can act on, prevent future information from leaking into training, and measure whether the action creates value.

## Business Decision Use Cases
| Decision Problem | Formulation & Operational Impact |
| :--- | :--- |
| **Customer retention** | Identify accounts with patterns associated with churn so service teams can address a real problem before offering an incentive. |
| **Sales and demand forecasting** | Forecast daily store-level sales across 1,000+ branches, weekly revenue, and e-grocery warehouse fulfillment volumes to plan logistics and inventory. |
| **Energy and commodity pricing** | Forecast wholesale day-ahead electricity prices (Nord Pool) and spot energy costs to optimize industrial procurement and power trading. |
| **Attendance and capacity** | Predict bookings, arrivals, or no-shows to manage waitlists, staffing, and constrained service capacity. |
| **Pricing, yield and replenishment** | Estimate value, product yield, demand, or replenishment quantities across complex multi-site retail supply chains. |

## Recommended Data Attributes
- Tenure, contract, product mix, service usage, and billing history
- Campaign exposure, channel, timing, and observed response
- Booking lead time, capacity, attendance, and cancellation behavior
- Store footfall, calendar holiday indicators, local weather, and historic sales sequence telemetry
- Product measurements, category, price, stock movement, and realized outcomes

## Evaluation Split & Benchmark Considerations
- Churn and response labels need a precise time window. Without one, teams may train on information that appears after the moment a prediction would be used.
- A strong classifier doesn't prove an intervention works. Retention and marketing actions should be evaluated against an appropriate control or experiment.
- In multi-site retail forecasting (e.g. Rossmann, Rohlik), zero-shot foundation models generalize across store hierarchies without requiring separate manual model fitting per SKU.

## Privacy, Governance & Regulatory Compliance
- Customer behavior can become intrusive when data is collected or combined beyond reasonable expectations. Use purpose limitation, consent where required, and clear retention and access policies.
- Running analysis locally can reduce spreadsheet uploads and third-party processing. Teams should still remove unnecessary identifiers, protect shared Sheets, and ensure predictions are used consistently with customer commitments.

## Representative Benchmark Datasets
- [telco-customer-churn](/datasets/telco-customer-churn.md) — Customer churn
- [fitness-club](/datasets/fitness-club.md) — Attendance and no-shows
- [spambase](/datasets/spambase.md) — Message filtering
- [abalone](/datasets/abalone.md) — Age and yield estimation
- [rossmann-1d](/datasets/rossmann-1d.md) — Daily store sales & promotional lift
- [rossmann-1w](/datasets/rossmann-1w.md) — Weekly retail sales & inventory replenishment
- [rohlik-orders-1d](/datasets/rohlik-orders-1d.md) — E-grocery fulfillment order volumes
- [epf-np](/datasets/epf-np.md) — Day-ahead electricity price forecasting
- [m-dense-1d](/datasets/m-dense-1d.md) — Multi-site operational demand planning

## Canonical Resource
- **HTML Guide:** [https://carlahq.eu/machine-learning/commerce-industry](https://carlahq.eu/machine-learning/commerce-industry)