Effective utilization of payments’ data
The widespread adoption of digitalization has resulted in an exponential increase in data generation. Data associated with payment transactions is an extremely important asset that can be used to understand customer behavior, manage operational risks, and grow business in a sustainable manner within the financial institutions.
Key enterprise asks
This blog discusses the concept of payments operational data store, which is increasingly being adopted by progressive financial institutions to achieve process efficiency, gain additional insights, and provide customer-centric solutions.
An effective data management process must address the following questions.
- Data capture: Where is data available?
- Data sourcing: How to source the data?
- Data processing and storage: How to process, link, and store data efficiently?
- Data consumption and visualization: How to make use of the data? What are the various business use-cases?
- Data maintenance and archival: How to maintain the data?
Data capture
Payments’ data gets generated throughout the entire processing life cycle. The processing life cycle includes payment initiation, processing, exception management and post processing. The associated set of data that must be considered, falls into the category of reference data, which includes payment scheme related market data, customer agreements and preferences and routing rules. A typical challenge that must be addressed is the fact that there are usually multiple applications that are involved in the maintenance of reference data and in processing of payments
Data sourcing
Typically, data is sourced through end of day process or scheduled in batch jobs based on several factors such as transaction status and last extracted time. While that will continue to be a mode of extracting data, technologies such as Kafka, Solace, etc enable real-time streaming of relevant information, usually triggered by change in events. With the increased adoption of real-time payments processing, the capability to extract and utilize data on a real-time or near-real time basis enables a wide range of use cases. Depending on the source and based on the capabilities of the source system, multiple types of data ingestion mechanisms might be adopted.
Data processing and storage
The next stage is to ensure that the data is stored in a manner that enables a wide range of use cases. The ingested data must be converted and stored in a canonical data model, which normalizes the potential differences in nomenclature adopted by the source systems. A canonical model based on ISO20022 business components is increasingly preferred, given the widespread adoption of this messaging standard. An extremely important aspect that must be addressed at this stage is to ensure that the transaction data is appropriately enriched, wherever necessary, from existing sources and lifecycle events. From a technology standpoint, organizations are increasingly looking at NoSQL solutions such as MongoDB use cases to build quickly, adapt reliably, and scale quickly.
Data consumption and visualization
This sets the stage for implementing several use cases based on this data store. Some of the more common use cases include:
- Real-time operational and business dashboards and reports
- Real-time payments tracking and observability
- Consolidated data for billing purposes
- Additional data for case management investigation
- Data extracts for enterprise level consolidation and reporting
Data maintenance and archival
To maintain the effectiveness of the operational data store, it is important to have a well-defined maintenance and archival process that periodically exports data to a long-term archive and purges the data from the operational data store.
Conclusion
Payments’ data has always been considered an important asset because it encapsulates customer behaviors, transaction risks, operational practices, and much more. The exponential growth in technology capability, especially in the field of analytics, is enabling organizations to unlock insights that were so far not extractable easily. These insights are enabling organizations generate additional revenue streams, enhance customer experience, improve operational efficiency, and reduce operational risks. A well-structured payment operational data store is an extremely critical component that enables organizations realize these benefits.
