Predictive Customer Analytics

Explore regression and linear regression to predict a continuous outcome from factors like car age, mileage, and brand, and store revenue from location, parking, income, and density.

Explore linear regression in predictive customer analytics by modeling store revenue from size and other predictors, and use Excel to predict revenue and interpret the slope and intercept.

Interpret how the r squared and adjusted r squared reveal how well the regression explains sales variation. Use the significance f value and p value, and Excel to predict sales.

Prepare data for linear regression by selecting sales as the dependent variable, encoding 11 store features with one-hot methods for text fields while dropping high-cardinality columns, and training in Excel.

Preprocess data for linear regression by converting location type and weather into numerical features via one-hot encoding, creating rural and urban columns and weather indicators.

Learn to build and interpret a linear regression model in Excel using the Analysis Toolpak, including data preprocessing, range selection, and coefficients, intercept, f-value, and p-values.

Utilize linear regression to predict sales from features like urban, population density, competitors, weather, store size, and parking space, and assess deviations to identify optimal store locations.

Explore how Excel stat automates encoding of categorical data to enable linear regression and other models in Excel, with coefficients, correlations, predictions, and charts.

Learn how to extend customer retention by predicting churn with logistic regression and targeting high-risk customers with personalized offers, improved delivery, and restocking alerts to reduce attrition.

Explore logistic regression for churn prediction using historical customer data; apply a function that turns results into a 0–1 probability, then use a threshold to drive retention actions.

Learn to assess the accuracy of a logistic regression model using the churn confusion matrix. Explore true positives, false positives, true negatives, and false negatives, and compute accuracy from totals.

Explore a logistic regression case study to predict customer churn at a large retailer, using 1000 customer features and Excel stat for encoding and churn probability prediction.

Implement logistic regression in Excel using the XLSTAT plugin to model churn with a binary target, interpret coefficients, ROC AUC, and confusion matrix for actionable predictive analytics.

Learn to perform logistic regression in Excel for predictive customer analytics using macros and Python scripts, including data encoding with VBA, and returning results in a new sheet.

Learn clustering as an unsupervised learning technique to group customers by purchase frequency, average transaction value, and product preferences. Use k-means to form three clusters and tailor marketing strategies.

Explore how k-means clustering groups data into three clusters by assigning each point to the nearest centroid, then updating centroids by averaging and repeating until stable.

Explore a case study on using k-means clustering in Excel to group customers by spending scores and income, enabling cluster-specific marketing strategies and long-term profitability for a mall.

Perform k-means clustering in Excel using Excel stats to segment 200 customers into three groups based on gender, age, income, and spending score, and interpret centroids and cluster profiles.

Forecasting in predictive customer analytics uses historical data to predict future events such as customer retention, sales, customer lifetime value, and engagement, with examples from Netflix, Amazon, Starbucks, and Spotify.

Forecast store sales across a retail chain using time series and machine learning models. Use historical data and factors like business growth, consumer confidence, and market trend.

Apply an additive time series model in excel by estimating base value, trend with month numbers, and seasonal effects to forecast air miles, using solver to minimize squared errors.

Learn the multiplicative time series model in Excel, multiplying the base by trend^t and the geometric mean of seasonality for forecasts, and compare it to the additive model via solver.

Celebrate crossing the final milestone and join the top 5% of students as you complete all lectures, check for missing lectures, and download your certificate of completion.