Using A One-Stop Data Utility To Create A Complete User eXperience

In today’s digital environment, data is the lifeblood for organisations. Whether it is used for research to develop a vaccine or to drive growth strategy in a new market, data is a key enabler. But how do we unlock the value of data if it is inaccessible or unavailable to the wider users? Are there any alternatives to make data accessible to the right people at the right time and with the right controls?

These areas remain key challenges for many organisations, the reason why we would like to discuss the importance of data democratisation today.

Inculcating the right data culture and mindset is the first step when embarking on a data transformation journey. This can be challenging, because banks and financial institutions operate in a tightly regulated environment and handle large volumes of confidential data across a myriad of data store technologies, i.e., Relational Database Management System (RDMS), NOSQL Databases, Big Data and File based stores.

It is however important to balance accurate and speedy access to granular data with regulatory and compliance requirements. This will allow the bank’s internal users to make data-driven decisions that will improve the customer’s journey, making it seamless and invisible.

Data democratisation allows users appropriate access to the data digitally without having to rely on another person or unit, so the information becomes easily available with the requisite controls embedded in. In an organisation, data is owned by various departments and users are granted access to different types of data depending on their role.

Due to certain restrictions, the same data might not be readily available to other groups who might require the same data for insights. Allowing data sharing between groups, with the appropriate checks and balances in place, breaks down silos and empowers various teams to make informed decisions using the data at hand.

Importance of Data Access and tools

To be a truly data-driven organisation, it is imperative that internal users are equipped with the right tools that enhance the user experience, whilst maintaining data security and performance.

In today’s digital environment, technology has become the driving force of the world, where human to machine or machine to machine interactions is rapidly increasing. With this fast pace of change, most if not all businesses are heavily reliant on data, particularly banks and financial institutions where compliance to regulatory requirements and ensuring data security and integrity is important.

For the users of big data in the bank, access to data that is seamless, secure and via a single point of entry for data stored in different datastores have been long-standing challenges.

Users must access millions, if not billions, of data points while keeping the data integral and clean. While there are open source tools available that could aggregate results from different databases, these have limitations in performance, usability, security that are expected of a highly regulated environment such as banks and financial institutions.

Until a few years ago, big data platforms were fairly new to the organisation, and not many were familiar with accessing data from such data stores. While these platforms and tools enabled cost-efficient data storage at scale, it was inconvenient to attain ad-hoc access to the data and analyse the same to answer queries with the same sophistication that the erstwhile RDBMS’s provided.

At the same time there were two other key considerations:

a) To ensure that badly written or sub-optimal queries did not adversely impact the operations of a data store — a lot of which are key for timely regulatory reporting or internal reporting/ analytics. There was hence a need for a pre facto detection of bad or rogue queries that could have adverse impact on other users and processes in a multi-tenanted environment.

b) For the data accessed and shared to ensure that this was being done in a secure manner with requisite Data Owner approvals and oversight. In fact, there was also a need to define rules specific to Work From Home (WFH) conditions in the backdrop of the pandemic that saw a lot of the workforce needing to work remotely.

Celerity: Enabling Swift Access and Independence

In today’s world where data is an asset, speed and accuracy are of the essence. Having a solution that is scalable, resilient, secure, and fit-for-purpose was the catalyst for developing our in-house solution, Celerity.

We chose the name “Celerity” as it implies “swiftness of movement”, which was precisely what we needed to achieve in the process of creating a product that solves a multi-fold problem with speed and precision.

Celerity: offering seamless data access, data sharing and reconciliation

Apart from the above features, the tool suite encompasses a Self-Service Data Quality capability to profile the data accuracy. Users are also able to self-service and define the data quality criteria and reports. It also provides for automation of Data Lineage needs, that maps the data flow and helps to visualise the source to destination journey.

In summary, Celerity provides a complete and comprehensive set of fit-for-purpose utilities that equips the user in their journey on leveraging Data.

In the preceding paragraphs, we have expounded on why there was a need to build a tool like Celerity at DBS. We have briefly touched on the capabilities of Celerity. In our next post, we will be sharing more insight into the different modules of Celerity and how it is complementing the data journey and improving user experience within DBS.

Watch this space as we will bring you more insights about Celerity in part 2!

(This article is also translated in Traditional Chinese language.

資料驅動組織的關鍵，使用一站式資料工具創造完整的分析使用體驗 -Abhishek Pitliya -Medium)