In today’s digital world, data silos are a massive challenge for organisations trying to improve efficiency and make better decisions. Nearly 85% of enterprises see data silos as a significant challenge to their digital transformation.
The second edition of the DECODE webinar, hosted by DBS Bank in collaboration with AIM Media House, brought together industry experts to explore the issue. The speakers, Luis Carlos Cruz Huertas, executive director at DBS Bank, and Sriram Venkateswaran, APAC head – data & analytics services, Google, shared their views on the causes of data silos, their impact and the role of cloud technology in overcoming this challenge. The webinar was moderated by AIM editor Amit Raja Naik.
The Problem with Data Silos
Data silos are isolated collections of data that exist within different departments or systems in an organisation. These silos typically arise when systems are not designed to integrate or communicate effectively, leading to inefficiencies.
According to Huertas, “Data silos are an unavoidable problem waiting to happen.” He said that companies often rely on a range of legacy systems, which are difficult to integrate due to their age, technical complexity, and lack of compatibility with modern platforms.
Huertas explained how these data silos create incomplete views of business operations, hinder collaboration, and lead to inefficient decision-making. “When you don’t have a comprehensive view of your data, you end up making decisions based on fragmented or outdated information. This can result in duplicate efforts, conflicting data, and inconsistent reporting, which ultimately reduces productivity,” he said.
Echoing similar sentiments, Venkateswaran said that data silos limit an organisation’s ability to fully realise the benefits of artificial intelligence (AI) and machine learning (ML). “Many organisations jump into AI and ML without having a proper data foundation in place. When data remains in silos, it’s challenging to derive meaningful insights, which is essential to drive business performance,” he explained.
How Cloud Can Help Break Data Silos
One of the key solutions to breaking down data silos, as discussed during the webinar, was cloud integration. Cloud platforms provide an end-to-end data lifecycle management solution, from data ingestion to transformation, storage, and analysis. Venkateswaran explained the capabilities of cloud data warehousing solutions and how they simplify the process of consolidating data from various sources.
“Cloud-based solutions offer native connectivity with popular ERP systems and the ability to build custom integrations with homegrown solutions. This facilitates real-time data access, which is crucial for decision-making and collaboration across teams,” he said.
Venkateswaran also discussed the scalability of cloud solutions, allowing organisations to grow their data capabilities without heavy investments in on-premise infrastructure. “The ability to scale up or down depending on your needs makes cloud a very healthy proposition for data integration,” he added.
He further explained that cloud technologies provide robust governance frameworks that ensure data security and compliance. “Cloud platforms allow for sophisticated role-based access controls, which means data can be shared securely across departments while ensuring compliance with regulations like GDPR,” said Venkateswaran.
Navigating Data Ownership and Governance in the Cloud Era
As organisations shift to cloud-based data strategies, it becomes crucial to clarify who owns the data, especially when the data moves across different platforms. “The ultimate ownership of data lies with the organisation as a whole. The board of directors and the CEO must drive the vision for breaking data silos, making it a core part of their business strategy,” said Huertas.
According to him, the responsibility isn’t confined to one individual or role but distributed across several layers within the company.
“Leaders must prioritise data literacy and governance at the C-suite level. Having a CDO to champion data initiatives, in collaboration with the CIO, ensures that the organisation can leverage data effectively to drive business decisions,” said Venkateswaran. He said that in most cases, the CDO’s role centres around maintaining data quality and governance, and ensuring that the data is used in ways that benefit the business.
Meanwhile, Huertas explained that a CEO doesn’t have the functional ownership of the data. He noted that different departments and units of an organisation might have the ownership of specific data assets or data sets related to their functions.
Citing an example, he said that the legal team often has a say when it comes to sensitive data. “Legal and compliance departments typically play a critical role in determining data ownership, particularly when handling regulated or sensitive information.”
This layered responsibility ensures that no single role has absolute control. Instead, it involves a concerted effort from different areas of the business.
“One of the most important elements of governance is establishing clear access controls and ensuring that data security is managed at the most granular level. Whether it’s role-based access or data retention policies, having the right governance guardrails in place ensures both compliance and data quality,” Venkateswaran said.
Huertas believes that effective data governance is not just about control, but about ensuring that the right access and security protocols are in place. “You need a comprehensive data governance framework that spans across the entire organisation, defining roles, responsibilities, and policies for how data is managed and shared,” he explained.
Real-Time Data Processing
Talking about real-time data processing, Venkateswaran expressed a nuanced view: “This is an age-old debate about real-time data versus batch processing. It’s fashionable to try real-time data processing, but that presents its own set of challenges.”
He urged data teams to consider if real-time data is truly necessary. “A significant portion of reporting analytics use cases could potentially be solved through traditional batch-processing methods, which can provide clean data with reasonable latency.”
However, he acknowledged that some use cases require real-time solutions. “For those use cases, cloud solutions provide capabilities like PubSub for streaming data to your data layer. Minimal transformations will ensure that data flows with minimal latency.”
Huertas added that a robust infrastructure is required for real-time data processing. “Cloud solutions are scalable and flexible, allowing them to handle high volumes and high velocities,” he said.
Measuring RoI with Data Practices
As they wrapped up the session, Venkateswaran shared insights on measuring the success of data-integration efforts. “The success or ROI of data integration can be measured in various dimensions, much like a balanced scorecard,” he said. He highlighted two key metrics for measuring ROI: the cost savings from transitioning to a unified warehouse and the significant improvements in performance, particularly in reducing latency.
“Data quality, compliance, and user satisfaction with integrated efforts are also crucial. Surveys and audits can provide valuable insights into these areas,” he added.
In conclusion, Venkateswaran provided valuable advice for IT leaders and developers. “The most important aspect is gaining executive buy-in for unifying data silos into a data warehouse. Technical details will follow, but driving change management and stakeholder engagement is key.”
Huertas echoed this sentiment, saying, “Breaking data silos is an ongoing process that requires commitment, collaboration, and continuous improvement. Prioritise data governance, adopt an API-first approach, use cloud technologies, and focus on data quality and security. Encouraging a data-driven culture across the organisation is essential.”
About DBS Tech India
DBS Tech India, formerly known as DBS Asia Hub 2 Pvt Ltd, was established in 2016 as the bank’s first offshore technology development centre in Hyderabad, India. As an integral part of DBS’ long-term strategy and operating as a development centre of excellence and innovation, it
In line with DBS Bank’s vision of being a #differentkindofbank, DBS Tech India is more than just a technology hub, always believing that technology is business and business is technology. With over 4,000 techies collaborating as one team, it is stepping upwards to a new future and fueling the next
The post The DECODE Webinar: Leveraging Cloud to Break Data Silos for Seamless Integration appeared first on AIM.