Architecting data-driven success: Challenges, solutions and best practices
When it comes to business success, data matters. Just ask the McKinsey Global Institute. The analysts firm noted that data-driven organizations are 23 times more likely to acquire customers, six times as likely to retain those customers, and 19 times as likely to be profitable as a result.
Becoming data-driven is not easy, however. It is more than just deploying business intelligence products—a reason why a report from EY noted that whilst 81% of organizations agreed data should be at the heart of everything they do, the vast majority continue to keep it locked away in silos.
At a roundtable, organized by Questex Asia together with Tableau Software, key senior executives representing a cross section of public and private industries in Singapore discussed about the challenges, key solutions and proven strategies that they are employing to become data-driven.
Where is the starting point to become data-driven? Apparently, there is no hard fast rule and depends on the industry and firm. According to representatives, it could be top-down, bottom-up or both.
“For some organizations, it is an initiative that is coming from top down. In others, it is bottom up where people are driving innovation and is a really organic approach,” Francois Ajenstat (photo left), Chief Product Officer, Tableau Software said.
For example, Singapore Post is becoming data-driven using a bottom up approach. “We are bottom up because we deal a lot with customer data and this data gives us insight to helps us understand which are the areas within a particular geography where there is a certain demand and how we can optimize our operations or process flow to service this demand,” Amit Dhupkar (photo right), Head of Group Technology, Singapore Post said.
For JLL, data-driven strategies began at the top and over time reverses to become bottom-up. “It definitely started top down with a strong strategic drive as part of CS JLL 2020 strategy over 3 years ago. We created two structures for BI and data governance. And now [our strategy] is evolving to also be bottom up as we find pain points and drive innovation around data,” Nabila Hadjkali (photo right), Head of Data Governance, Corporate Solutions, Asia Pacific, JLL said.
In some organizations it happens both ways at the same time. One of the participants to the roundtable noted that at their organization innovation is bottom up process while CRM is top down. What is important is developing a framework where data is viewed in different ways.
Both Jean-Marc Henaff (photo left), CIO, Direct Asia Management Services and Peter Ton-That (photo right), Head of APAC IT Delivery (Investment), Schroders agreed.
For Ton-That, it was a matter of deploying the right data governance. Meanwhile, Henaff highlighted that while his organization looked to become data-driven, it became increasingly important to have a holistic data strategy first in place. “We now think about the whole data strategy; no longer just from an actuarial view,” he added.
One big advantage of using BI tools is allowing business users to quickly identify opportunities and examine areas for improvements.
Here, Wilson Tan (photo left), Head, Virtual Banking & Payments, Maybank identified a gap. “Clearly between the need for data and the ease of access to data—there is a gap,” he said.
One key reason for this gap is skillset. One participant noted that self-service is a necessary direction for the company to take if they want to compete with big companies. This requires reviewing the resources and capabilities of those at the business level looking to get to the real core data set.
“Lack of internal expertise and resources, whether from IT or business who want to take the lead, is a major challenge,” observed Roy Lim (Photo right), Regional Associate Director, IS & EUC BRM, Zimmer Pte Ltd.
JLL’s Hadjkali recommended firms to take a step back and understand the role of data quality and definition in the SSBI equation. “We need to have self-service BI to be as simple as possible to use and clearly this is the technology challenge. We also need to ensure that the data provided to self-service users is sound and clearly understood. These aspects form the data governance challenge. So, we should take a step back from solving SSBI equation only with technology solution and in parallel focus on developing the skills, process and strategy to ensure technology is fed with sound and commonly understood data,” she explained.
Previous tools can also hinder moving forward. Atish Mukhopadhyaya (photo left), Senior Manager, Customer Care APAC, BASF SEA noted that his company’s reliance on a previous ERP tool concentrated the skillset and tool knowledge to a very few. They were becoming a bottleneck for innovation. In this regard, visualization became important for BI as it allows “the need for specialized skilled people to come down.”
It may also be time to change how we operate and run companies to enable a culture of analytics, said Tableau Software’s Ajenstat. “We can’t operate companies the same way we did years ago. We need to train people. At Tableau, we have made it easier to [securely] analyze the data. We also give training for free, and we built a culture of sharing best practices through activities such as hackathons. We are also investing in analytical skills in schools through free software and curriculum so future interns have the basic of understanding [of data management and analysis].”
Sometimes, the skillsets need to be changed to get value from the capabilities of the tool. JLL’s Hadjkali, for example, saw that the deployment of a dashboard did not immediately improve the BI capabilities of business users. “We found that business [users] were not in the habit of questioning the data but rather spent time preparing the data. Once [this need to prepare was removed], data analysis became a new skillset [for business users] to learn.”
Vishal Sharma (Photo right), Lead Digital Architect for a global bank argued that business users need to change their mindsets. “Data is owned by the business; it is their data. The question we are always asking is what we want to do with this data. So, we are getting our business managers to think a bit more differently to see what they will do with these insights and how it is going to affect their bottom lines,” Sharma said.
While data-driven strategies are becoming increasingly important, participants noted that firms need to overcome key misconceptions.
A key one is the idea that there is one tool for all data analysis workloads. “This is a major one for us: the idea that there is one tool to solve everything,” Ramesh Munamarty (Photo left), Group CIO, International SOS said.
Another is the sudden realization by users that they have to learn a new skill. “They come in thinking the tools are more flexible and faster, but then they realize that they need new skillsets to use it more effectively,” Chin Sien Phang (Photo right), Media Insights Specialist, StarHub said.
Meanwhile, Maybank’s Tan sees BI more than just data analytics. “There are many things that require business intelligence to work. First there is getting the data source and then data preparation before we can get to the insights. I do not think that it is that simple.”
JLL’s Hadjkali echoed a similar viewpoint. “A lot of companies invest in BI visualization and not on data. Then they face frustration and disappointment when the [right] data is not there.”
“Data is still biggest challenge. But you do not have to wait till you have to have the perfect data set, which is a misconception, as there is no single version of the truth,” Ajenstat added.
In addition, organizational structure and differing data privacy regulations across the region were cited as major concerns.
Despite these challenges, all agreed that becoming data-driven and BI-savvy is important. Most, like Direct Asia Management Services’ Henaff, sees this as a means to compete better.
Singapore Post’s Dhupkar sees data driven strategies helping make for better decision making.
More importantly, becoming data-driven allows firms to discover the unknown. This is where Maybank’s Tan sees the main advantage lies. “BI allows you to know what you do not know with a good data set,” he said.
Tableau Software’s Ajenstat summarized this as “visibility.” “You cannot know what you cannot measure. Having the visibility can ultimately drive problem resolution. It allows you to develop questions that you may not have asked and there is a ton of innovation there waiting to happen.”