Effective data architecture key to supporting analytics initiatives: Rahul Mehta, CitiusTech
Companies are realising that they can benefit tremendously from data and analytics to drive positive outcomes for themselves as well as customers. For this, organizations need to have a clear vision that focuses on driving efficiency, productivity and profitability. However, there are several challenges associated with data analytics which managers find difficult to handle.
Rahul Mehta, Sr Vice President and Head of Data Management Proficiency, CitiusTech talks about how tech leaders can effectively manage and deploy analytics function.
What Investments are needed to set up analytics functions in an organisation? How would one do the cost-benefit analysis?
An effective data architecture and technology infrastructure are prerequisites that must be in place to support any analytics initiative. They must also prioritize investments in analytics and start out on a small scale. Once a use case has been successfully piloted, organizations can expand its application across various enterprise functions to add more value, save costs, improve customer experience, market share impact, etc.
Organizations that have moved ahead from the pilot stage need to then focus on operationalizing and scaling their pilot models – some investment in training, process optimization and automation is anticipated at this stage. Overall, for the successful implementation of any major project, leadership buy-in is of utmost importance. Communication and direction from the top can go a long way in articulating the importance of initiatives. Senior-management involvement along with organisational structure play a decisive role in determining effectiveness of any initiative, including analytics.
What should tech leaders keep in mind while setting up analytics functions and while deploying it?
They need to understand and build aligned capabilities like data integration, data mining, enterprise data warehouse management, data lakes, data security, user experience management, etc. There is a strong interplay of these areas with core analytics functions like reporting / dashboard development, rules management and performance management.
Additionally, if we talk in healthcare context, healthcare, data models are lot more complex and adds to the overall implementation difficulty and complexity of performing analytics on the data. Hence, to get maximum value from the enormous volumes of data available to organizations, analytics teams will need to build a long-term approach that includes next-generation paradigms like AI, machine learning and deep learning. Identifying, grooming and retaining talent is likely to be an area of focus in establishing an analytics function.
Further, when moving from the pilot stage to the production stage, organisations will need to overcome challenges like data accessibility and availability, building, operationalizing and scaling models in a standardized way and making informed decisions around the right platforms and deployment options (on-cloud vs on-premise).
Lastly, organizations need to ensure that the quality of the data being analysed is sanitized. Gaps in data quality can impact data analysis and thereby lead to erroneous results. In the era where valuable, unstructured data pours in from various sources, fine-tuning data quality is an essential step to succeed with analytics.
How does analytics impact the top line and /or bottom line of an enterprise?
Investment of analytics can help exponentially increase efficiencies across business functions once implemented. Technology intervention can automate simple but time-consuming manual tasks, freeing up the individual’s time to work on more complex functions. It ensures accuracy and eliminates the probability of human error, ensuring a lot more accuracy in reporting and other such manual activities. Specifically, in the context of healthcare system, analytics has a huge impact on the providers’ and payers’ bottom line by identifying problem areas early on and use of artificial intelligence and machine learning algorithms to enable early detection and addressing of the issue promptly. It also helps organizations discover newer business opportunities and discern possible problems that enterprises could encounter in the long-term.
Though India boasts a large tech workforce there is a lack of data scientists and AI engineers who can create algorithms. How should enterprises deal with this shortage of tech talent?
Analytics today has become a multi-dimensional space, ranging from data mining and building data lakes at one end, extending to AI and Machine learning at the other. The range of skills required within analytics teams, therefore has become extremely diverse and demonstrating an organizational edge in analytics involves attracting, retaining, and sourcing the right talent. Analytics professionals need to have a strong understanding of specific domains and verticals in which they operate. It is tough both to obtain and retain employees with core analytical skills, even more than engineers and data scientists. We believe that one of the biggest challenges for companies engaging and keeping pace with technological changes, is acquiring the right people, with the necessary skillset and getting them trained to solve client-specific use cases.
Dealing with a complex and fast changing technology landscape, including new paradigms like RPA, AI/ML, Cloud, etc., means that analytics companies and the industry overall are under continuous pressure to stay ahead of the technology curve. Organizations that are limited in their ability to train and develop competencies are likely to face a bigger hiring challenge. At CitiusTech, we are addressing this through a mix of focused hiring, college/university partnership and internal learning and competency development via cross-skilling, up-skilling and multi-skilling programs across multiple areas such as data management, performance management, AI/ML, healthcare domain, etc.