Customer conversion rate is a metric used to identify the number of users who have bought products from a site compared to the total number of visitors to the site. Tracking marketing campaign conversion rates is essential to fine tune the marketing strategy, or even change it if it proves to be ineffective. It can also be used to identify the areas with the most conversions and areas with lesser conversions, allowing companies to identify focus groups and fine tune their advertising strategy.
This project aims to visualise and analyse the effect of certain promotions on customer conversion. Using this information, the website can identify users with the most conversion potential and can understand the effects of different types of offers.
From the graphs and matrices plotted, we can see that the conversion rate was highest in users who received discounts, compared to users receiving 'Buy one get one' offers, and among web users compared to mobile users. The conversion rate was also higher in users who had used the platform within the past 2 months. Furthermore, the customer conversion is highest in suburban areas, and among users who have used discounts previously.
A Qini curve used to evaluate the performance of a model that predicts the incremental impact of a marketing or treatment campaign on individuals. A steeper Qini curve indicates that the uplift model is better at identifying individuals who are most likely to respond positively to the treatment compared to a random model or a less effective uplift model. In this notebook, the second section attempts to use uplift modeling to identify groups which are more likely to respond to the offers, i.e. BOGO and discounts. The Qini curve plotted is used to evaluate the accuracy of the models.
- Visualization using PowerBI: Visualise the data in PowerBI.