Marketing Intelligence Volume 2: Developing the Models and Analysis, by Dr. Kyle Allison
Unlock the full potential of data in shaping innovative and effective strategies with this second installment of a three-volume series. Transform your marketing data into confident, decision-ready models that finally show you what’s working and where to invest next.
Starbucks and the morning rush
From data-driven to decision-ready
First define the real problem
The foundation for everything that follows
Chapter takeaways
The role of descriptive analytics in marketing
Data summaries and aggregations
Visual storytelling
Turning observations into narratives
Avoiding vanity metrics and misleading averages
Mini case: The weekday spike
Chapter takeaways
Moving from description to diagnosis
Correlation versus causation in marketing data
Attribution as diagnostic logic
Identifying leading and lagging indicators
Leading vs. Lagging Indicators
Using diagnostic insight to frame predictive models
Mini case: The ROI drop in ad group alpha
Chapter takeaways
Data collection and integration across platforms
Feature engineering
Encoding and scaling techniques
Splitting data into training, testing, and validation sets
Auditing data quality and bias
Mini case: Preparing multi-channel campaign data for predictive modeling
Chapter takeaways
The anatomy of a model
Linear and multiple regression in marketing contexts
Coefficients and interpreting relationships
Evaluating model fit
Dealing with multicollinearity and variable selection
Translating statistical output into marketing meaning
Mini case: Relationships between ad spend, price promotions, and sales
Chapter takeaways
Clustering techniques
Overall clustering perspective
Evaluating segments for actionability
Rule-based attribution models
Algorithmic attribution
Integrating segmentation and attribution for strategic insight
Mini case: Segmenting online customers and identifying touchpoints
Bringing it all together
The evolving future of segmentation and attribution
Chapter takeaways
Regression versus classification models
Regression versus classification in the real world
Machine learning basics
Avoiding overfitting
Building trust in predictive models through validation
Mini case: Predicting lead conversion with logistic regression
The future of predictive analytics
Chapter takeaways
Marketing mix modeling fundamentals
Bringing it all together
Mini case: Beauty brand launch
Using MMM for budget allocation and scenario planning
MMM meets predictive and attribution models
Out-of-sample forecasting: The final exam
Communication: Where trust becomes action
Mini case: Measuring ROI of social, search, and TV spend
Insight from the field
The future of marketing mix modeling
Chapter takeaways
Why forecasting matters strategically
Understanding time series components
Mini case: Forecasting in action
Forecasting techniques
Forecasting models in Excel, Python, or Power BI
Evaluating forecast accuracy
Using forecasts for planning and inventory decisions
Comparing forecasting to marketing mix approaches
Mini case: Forecasting holiday sales using ARIMA
Insight from the field
The future of forecasting in marketing analytics
Chapter takeaways
Evaluation metrics
Model comparison and selection techniques
Analytics is a strategic choice, not a beauty contest
Best practices for presenting results
Turning analytical findings into marketing recommendations
Creating feedback loops for continuous model improvement
Bridging to prescriptive analytics: Volume 3 preview
Mini case: Evaluating and presenting churn models to leadership
Chapter takeaways
Journey step by step from “we have a lot of data” to “we have a clear plan.” Start by nailing the real marketing problem you’re trying to solve, then move through descriptive analytics, diagnostic analytics, and data preparation so your dashboards, reports, and SQL/Excel/Python work actually line up with the questions the business is asking. Along the way, you’ll see how brands like Starbucks, Amazon, and modern e-commerce retailers use data storytelling, segmentation, and attribution to turn scattered metrics into sharp customer insights and smarter campaign decisions.
Next, the book dives into the core models every modern marketer needs: regression and correlation for understanding relationships between ad spend, pricing, and sales; clustering for customer segmentation; attribution models that give search, social, email, and offline channels proper credit; and predictive analytics that estimate churn, conversion, and lifetime value before they happen. See how to structure your datasets, engineer features, handle bias, split training and test sets, and avoid overfitting so your machine learning and marketing analytics hold up in the real world.
Finally, get a practical tour of marketing mix modeling (MMM) and time-series forecasting, showing how to quantify the impact of TV, paid search, social, retail promotions, and email on revenue and profit. Clear examples, mini-cases, and visuals illustrate how to build scenarios, reallocate budgets, forecast demand, and connect your marketing mix models to inventory planning, pricing strategy, and brand growth. Whether you work primarily in spreadsheets, BI tools like Power BI, or coding environments like Python, you’ll find concrete starting points you can lift straight into your own workflow.
By the time you’ll reach the last page, you’ll have a toolkit for segmentation, attribution, predictive modeling, marketing mix modeling, forecasting, and model evaluation that fits together into one coherent system for data-driven marketing. More importantly, you’ll know how to turn those models into decisions people actually follow, and be able to bridge the gap between marketing analytics, marketing strategy, and real business impact, setting you up perfectly for Volume 3’s focus on delivering strategic, enterprise-wide influence.
Dr. Kyle Allison, known as The Doctor of Digital Strategy, blends over two decades of industry and academic expertise in digital marketing, analytics, e-commerce, and merchandising. He has held various retail roles up to senior executive levels at renowned organizations such as Best Buy, Dick’s Sporting Goods, Dickies, and the Army and Air Force Exchange Service, leading data-driven, transformative initiatives. As a distinguished professor and mentor, he has shaped future professionals by teaching digital marketing, analytics, e-commerce, and business strategy at top institutions while also contributing to curriculum design and doctoral mentorship. A prolific author, Dr. Allison has published Quick Study Guides, textbooks, journal articles, and trade books, bridging academic theory with practical application.
Please complete all fields.
