The task aimed at analysing the performance of a marketing campaign that was produced on different channels using common metrics such as conversion rate and retention rate and later on performing A/B testing.
The task started with exploratory data analysis through determining how many users are seeing the marketing assets each day. This is crucial to understand how effective the marketing campaign was over past months.
Next was the visualization of the daily marketing reach which revealed a high peak in times when an email get sent to the target audience.
After visualization, it was time to calculate some marketing metrics such as conversion rate and retention rate.
Conversion and retention rate were calculated across language displayed of the marketing content and channels used respectively. The results showed conversion rate was much lower for English and Spanish speakers. Moreover in January while visualizing conversion rate proved to be steady with exception of a single day where there was a high peak.
To investigate whether the marketing channels reach all users, a visualization was done across different age groups. The results showed that Emails were not reaching older age groups and Facebook was not reaching the younger population.
An error was discovered which revealed that the under performance of House Ads was due to the fact that the ads were all served in English language rather than the user's preferred language.
Finally the project ended with the A/B test results analysis of the email portion of the campaign. Half the emails sent out were generic upsells to the product while the other half contained personalized messaging around the users’ usage of the site. The results showed personalization was more effective for younger groups than older groups.