This project aimed to create a predictive model to estimate the additional profits generated by two loyalty programs at a major retailer. The primary goal was to separate the profits stemming from day-to-day operations from those driven directly by loyalty initiatives. The model provided insights into whether loyalty customers inherently generated higher profits or if those profits resulted from the retailer’s loyalty campaigns.
-
Data Collection and Analysis:
- Analyzed customer data, including growth rates, revenues, active customer base, and EBITDA.
- Analyzed multiple datasets containing historical customer data, loyalty enrollment figures, revenues, growth rates, and EBITDA.
- Evaluated the average ticket size of loyalty customers versus non-loyalty customers, identifying significantly higher values among loyalty customers.
- Evaluated historical data to identify patterns in loyalty customer behavior versus regular customer behavior.
- Compared growth in revenues between regular and loyalty customers to assess the source of increased spending.
- Conducted analysis on customer activity patterns and how quickly loyalty customers scaled in numbers over time.
-
Regression Analysis:
- Plotted the historical growth in loyalty customers.
- Performed regression analysis to predict the expected number of loyalty customers for 2016. Tested several regression types to find the most accurate.
- Used regression results to forecast expected revenue uplift and define the impact of loyalty-related efforts versus organic growth.
-
Collaboration with Business Stakeholders:
- Engaged with the business to evaluate the impact of loyalty campaigns, customized offers, and targeted marketing on customer behavior.
- Segmented profits into two categories: organic growth and revenue generated from loyalty-driven marketing efforts.
- Identified the fraction of growth driven organically versus through marketing efforts.
-
Savings Prediction and Evaluation:
- Estimated the savings (additional profits) expected from loyalty customers in 2016.
- Set a target of approximately 10.5 million in predicted savings for the year.
- Analyzed the contribution of specific loyalty campaigns, such as personalized offers, bonuses, and targeted advertisements, to profits.
-
Outcome Tracking and Iterative Improvements:
- The model continues to be used to evaluate the effectiveness of loyalty programs and the profits they generate.
- Data Analysis Tools: Excel for data consolidation and regression analysis.
- Collaboration: Engaged business stakeholders for insight into loyalty campaigns and marketing initiatives.
- Statistical Techniques: Regression analysis for predicting customer behavior and loyalty program impact.
- Accurate Predictions: The model accurately forecasted the number of loyalty customers for 2016, aligning closely with actual figures by the end of the year.
- Savings Achieved: The company achieved the predicted 10.5 million savings target, demonstrating the effectiveness of the loyalty program.
- Business Insights: The analysis showed that the majority of customer growth was organic—customers naturally increased spending without requiring aggressive campaign efforts. Only a smaller portion of the uplift was attributed to loyalty-specific campaigns.
- Continuous Use: The model is now integrated into routine business practices, supporting future analysis on whether loyalty programs continue delivering expected profits.
- Operational Insights: This model provides real-time monitoring of customer behavior, allowing the business to adjust marketing strategies and incentives effectively.
- Predictive Modeling: Developed expertise in using regression models to forecast customer behavior and financial outcomes.
- Data Analysis and Reporting: Gained experience in working with large datasets, identifying patterns, and communicating findings effectively by use of graphs.
- Business Collaboration: Improved skills in working with business stakeholders to align data insights with strategic goals.
- Impact Assessment: Learned how to objectively measure the effectiveness of loyalty programs and their contribution to profitability.
- Model Integration: Gained experience in embedding predictive models into business processes, ensuring ongoing relevance and value for stakeholders.
-
Clone the Repository git clone https://github.com/realdanizilla/Loyalty-Savings.git
-
Open the main Excel File and start at sheet "Premises" to understand basic model premises
-
Analyse "Model" to understand the basic data and how model is applied to each month
-
Visualize actual and projected data related to loyalty customers
-
Change parameters in the "Saving Parameters" section to change how the model works