Selpal’s quality customer experiences enabled by Keyrus and Datapult

The implementation of Datapult by Keyrus improves Selpal’s complex relational database model and ensures it has access to up-to-the-minute client information, significantly improving customer service.

Today, it is widely recognised that simple ‘customer service’ is no longer enough to keep clients happy. Instead, the total ‘customer experience’ (CX) has to be taken into account, meaning it is more important than ever for businesses to be able to leverage their clients’ data more effectively. After all, it is only by knowing more about your customers that you will be in a position to deliver faster, better or more targeted services.

As a rapidly growing organisation, Selpal’s business model entails creating a network of point of sale (POS) devices in the informal sector and leveraging this network for an array of different business propositions. Selpal is an ecosystem that connects store traders to customers and suppliers – delivering everything from airtime and data to electricity, DStv, gaming and cashless orders and payments to suppliers.

This business model has meant the company has been able to collect a vast dataset of transactions from its various devices in these informal markets. However, the key challenge the company faced in making use of this information was that its application had a complex relational database model. This made it difficult to report on, due to query complexity, and unsuited for reading large amounts of data.

Inadequate data warehousing practices along with difficulties in finding the necessary human resources led Selpal to contract Datapult, to fast track its data warehousing journey. The company also sought expertise that could help it avoid the pain points involved with large data loads from the application’s database in the cloud.

An Extract, Transform and Load (ETL) solution, Datapult was implemented by a team of experts at Keyrus led by Evan Barke, principal consultant. Datapult is a solution that enables the rapid building of a data warehouse and the centralising of the entirety of a company’s business data. This, in turn, helps to improve time to market and significantly reduces the costs of preparing data for analysis,” Barke says.

“The beauty of Datapult is that it can build a data warehouse that transforms data optimised for a heavy transactional system, and does so in a way that facilitates swift analytics and ensures data-inspired insights for all business users. It therefore makes it much simpler for Selpal to deliver the required level of CX to its customers.”

According to Barke, Datapult’s solution is a revolutionary enterprise data warehouse builder that significantly improves time to market by removing all custom programming, SQL coding and ETL package writing.

“Thanks to an innovative, programmatical approach, Datapult can also leverage the metadata of the underlying source information, enabling it to build a state-of-the-art data warehouse within days, as opposed to months. The benefit of this is that it significantly improves time to market while at the same time curtailing the costs surrounding the preparation of data for analysis, mining, machine learning and reporting.”

He adds that perhaps, most crucially, it also operates in real-time, meaning that a Datapult data warehouse is in sync to the minute, enabling Selpal to report on real, current data while also eliminating the load on the business-critical application database.

Stephen Goldberg, Joint CEO at Selpal, indicates the company is pleased with the Datapult implementation, which has delivered a dynamic warehouse that transforms data optimised for a heavy transactional system into information better suited for analytic querying.

“This means that business users can query a read-only optimised database and return business intelligence rich results for extremely large datasets, all delivered with incredibly fast response times,” says Goldberg.

“We were very impressed with both Keyrus and Datapult, as the implementation occurred just as fast as promised. In fact, within days, a star schema data warehouse was designed, backloaded and in sync with production. Because the entire process has been simplified and the data model is much easier to understand, our business intelligence (BI) department is now able to spend its time writing visualisations and reports that offer us tangible business value,” Goldberg concludes.

Barke points out that the most successful companies are able to deliver next-level CX. “Ultimately, Datapult helps Selpal achieve this by delivering a central source of truth, while helping to reduce both time to market and cost, while delivering increased stability and trust.”