Loyalty with structural equation modelling
How structural equation modelling shows which guest experiences actually drive loyalty, with a hotel path model, total effects and an investment decision.
Statistical Methods
Statistical thinking and analysis
Media mix modelling: adstock, saturation and where the next pound goes
How media mix modelling uses adstock and saturation curves to find the marginal ROI per channel and where the next pound should go, with a hamper shop example.
Data Roles in Practice: How Data Engineers, Scientists, and Analysts Actually Grow SaaS and eCommerce Businesses
A deeply practical guide to what data engineers, data scientists, and data analysts actually do to grow SaaS and eCommerce businesses. Each use case includes why it matters, exactly how it is done, the measurable impact with real revenue and profit examples, and an importance rating from 1 to 10. Covers everything from building unified customer views and calculating true customer lifetime value to predicting churn, optimising pricing, running incrementality tests, and building recommendation engines. No theory, just tactics with numbers.
Modelling the value of marketing communications
How to put an honest number on one more email, webinar or retargeting ad: response curves, marginal value, holdout groups and the frequency that pays.
Tools of segmentation: k means, latent class analysis and going beyond RFM
k means, latent class analysis and RFM compared on a real hamper shop: how to choose k, profile and name segments, check stability and activate them.
The Complete Guide to Digital Marketing Testing Methods
A comprehensive deep dive into every major testing methodology used in digital marketing. Covers running without tests, A/B testing with two variants, multi-split testing with three or more variants, multivariate testing for element combinations, holdout control groups for measuring incremental lift, geographic split testing for regional campaigns, audience split testing for segment-specific optimisation, and sequential time-based testing for low-traffic scenarios. Each method includes detailed explanations of when to use it, statistical requirements, implementation considerations, advantages and disadvantages, and real-world examples across e-commerce, SaaS, fintech, travel, and retail industries.
Principal components and factor analysis for marketers
How PCA and factor analysis turn 24 survey items or 18 behavioural variables into a few dimensions you can name, score and act on, with hotel and hamper examples.
Simultaneous equations: when marketing causes sales and sales cause marketing
Budget rules make sales drive ad spend while ad spend drives sales. Why one regression overstates the ad effect, and how two stage least squares corrects it.
Forecasting demand: autocorrelation, seasonality and honest error bands
How to forecast weekly demand with seasonal dummies, handle autocorrelation, judge models on a holdout and plan against an 80 percent error band.
Conversion and UX Psychology: The Science of Why Users Click or Bounce
A comprehensive deep dive into the psychological laws that govern user behaviour and conversion. From Hick's law and Fitts's law to the peak-end rule, serial position effect, Miller's law, cognitive load theory, friction frameworks, and attribute framing. Includes research data, Python code for measuring UX impact, and practical implementation strategies backed by Baymard Institute research.
The Math Behind It All: Growth Marketing Mathematics Explained
A comprehensive deep dive into the mathematics that powers growth marketing and data-driven decision making. From LTV:CAC ratio and payback period to cohort analysis, survival analysis with Kaplan-Meier curves, Bayesian A/B testing, multi-armed bandits with Thompson sampling, marketing mix modelling vs attribution, power law distributions, price elasticity with Van Westendorp's price sensitivity meter, and the compounding math of retention. Includes LaTeX formulas, Python code, visual explanations, and practical implementation guidance for both beginners and advanced practitioners.
Panel regression and same store sales
How panel regression separates a good store from a good quarter, why pooled data lies about marketing, and how fixed effects power like for like sales.
Survival analysis, churn and customer lifetime value
How survival analysis tells you when customers leave, not just whether, and how to turn Kaplan Meier curves and a Cox model into a lifetime value you can defend.
AI for Reducing Cart Abandonment and Returns
The average cart abandonment rate is 70%. Ecommerce returns cost US retailers $890 billion in 2024, with fashion return rates hitting 24-30%. AI reduces cart abandonment by 18% and size-related returns by 27%. This post shows how to build predictive models that spot drop-off patterns and sizing issues before they cost you money - with Python scripts for abandonment risk scoring and return prediction, plus the Solidus/Rails implementation that wires them into your checkout and product pages.
Logistic regression, lift charts and market basket analysis
How logistic regression ranks customers, where the marginal brochure stops paying for itself, and how the same model turns baskets into cross sell recommendations.
Poisson regression: modelling how many times customers do things
Orders, tickets, visits: when the outcome is a count, ordinary regression misbehaves. Poisson regression, rate ratios and overdispersion, shown on a hamper shop.
Polynomial distributed lags: how long does marketing keep working?
Why an email keeps producing trials for weeks, how the Almon polynomial lag model measures that curve, and how Werkbank used it to pick a three weekly send cadence.
Modelling demand and price elasticity with regression
A year of hamper sales, one regression and an elasticity of minus 1.6: how to model demand with OLS, read the output and find the price that pays.
What is an insight, and why most reports do not contain one
A finding says what happened. An insight names the decision it changes, prices it and states the risk. How to tell them apart, score them and write one.
The statistics every marketer actually needs
Mean, median, standard deviation, confidence intervals and correlation, explained on a hamper shop's orders so you can read marketing numbers honestly.
The Importance of Statistical Significance and Error Margins in Performance and Growth Marketing
Understanding statistical significance, error margins, and sample sizes is crucial for data driven decision making in marketing. In this post, we explain these concepts and how they impact your marketing campaigns, especially in ecommerce.
Statistical Thinking for Growth Hackers: Why Averages Are Lying to You
Learn how I use statistical thinking to uncover hidden profit in skewed distributions, segment high value customers, and make better growth decisions than those who rely on misleading averages.