<p>The client runs an extensive e-commerce portal that receives a huge amount of user, transactional, and behavioral data. They were using an in-house relational database system that contained 7 years of data of all online and offline marketing campaigns from different marketing platforms.</p>
<p>Each campaign was expected to have at least 1 million products with 48K – 72K customer interactions, resulting in 48 – 72 billion data points. It includes the responses received, types of responses, date, time, minimum/maximum temperature, product information, customer information, and more via social media platforms, SMS, push notifications, and emails. The in-house database was incapable of analyzing such huge amounts of permutations and combinations.</p>
<p>The client wanted to get key insights from the database—such as campaign entry/exit touchpoints, performance statistics of various channels, etc.—to better target future campaigns. They wanted to ensure that only the intended users receive the discount coupons via the marketing campaigns.</p>