Loading...

'We're expanding our base to hire skilled people from smaller town in workforce': Sunil Senan, Infosys

'We're expanding our base to hire skilled people from smaller town in workforce': Sunil Senan, Infosys
22 Aug, 2022
Loading...

Data is growing at an extraordinary rate and as every company today is focusing on digital transformation initiatives, they are leveraging data analytics to get important business insights that can be used to make better business decisions, drive new levels of agility and improve bottomline.   

In an interaction with Tech Circle, Sunil Senan, Senior Vice President and Business Head, Data and Analytics, Infosys, discusses important trends in data analytics and challenges in keeping up with modern business requirements. He also explains how Infosys is tackling skill shortage or high attrition in the areas of data science, artificial intelligence (AI) and analytics. 

What are some of the key trends you see in data analytics today?   

Loading...

Today, data is no longer limited to the enterprise. Organisations are partnering to build ‘data ecosystems’ where data is being consolidated and shared through marketplaces from various sources. Companies are increasingly turning to industry data platforms with pre-built models that offer insights and recommendations to address industry-specific business requirements. 

Further, 5G, with its ability to support ultra-low latency in transmitting humungous volumes of data, combined with the processing power of edge computing, would be the new drivers for AI-based business transactions and operations in telecom as well as several other industries.   

Also, organisations are using AI and analytics for driving sustainability initiatives. For example, AI can help retailers improve container space utilisation on long-distance shipments or suggest load optimization strategies to reduce freight.  

Loading...

How do you differentiate between modern and traditional analytics?  

Traditional analytics uses statistical models and are typically designed for structured data. They most often need human intervention for validation and interpretation of data. Very often, such analytics are limited to a particular department, environment, or team, and hence become relevant only to a particular situation. Also, such models often help you focus only on the ‘what’ of things and don’t shed much light on the cause or reason behind the trends or changes. For instance, it can suggest what price to set for a product without explaining why that price is being recommended.   

In contrast, modern data analytics are robust and intelligent, being powered by AI and machine learning (ML). They are dynamic in nature, using an ensemble approach to combine information from various sources, both structured and unstructured, on a real-time basis to answer questions. They can combine different kinds of models to arrive at holistic insights and decision-making. They support natural language queries such as, “How did the new campaign impact our brand awareness?” instead of asking users to define their questions in technical terms. AI models can help organisations choose variables which are most relevant and have the maximum impact to the organisation reducing the data scientist’s burden of identifying and selecting the right data variable. Explainability and transparency are an integral part of modern analytics that tells you why a particular recommendation is being made.   

Loading...

Data analytics is evolving and today, with emphasis on intelligent automation, ML, augmented analytics, NLP, and adoption of cloud. Can you tell us about the new opportunities and challenges in keeping up with the contemporary business requirements? 

The biggest opportunity in data analytics is to empower people to use more data in their daily work and decision-making process. Secondly, the prevalence of open banking and commerce is making it feasible to create business ecosystems comprising of customers, suppliers, partners and even competitors to share data, such as the UPI (Unified Payments Interface) in India. The coming of the metaverse and web 3.0 is the future opportunity where data will again be a dominating factor in creating an experience-driven economy. 

To fully realise these benefits, organisations need to make investments. However, long gestation periods and changing dynamics of the market hold back enterprises from venturing headlong into data analytics. 

Loading...

Also, data analytics is as good as the quality of data. Ensuring consistency and integrity in data is a challenge in many industries. Explainability is another large concern, where confidence in an output is often missing because there is no view into how that output or recommendation was generated. 

What are the skills Infosys looks for in data experts? What kind of projects do they handle?  

Data analytics requires numerous skills across technology, business consulting and industry expertise. At Infosys, we look for expertise in data engineering, AI/advanced analytics, and data consulting and data management. We focus on niche and emerging skills through hiring digital specialists and full stack programmers that deepens our core capabilities across data on cloud, visualisation, AI/ML ops engineers and data scientist roles. Our data strategists with deep domain, technology and business understanding front end strategic conversations with client CXO’s and orchestrate offerings across data consulting, domain, architecture and engineering functions. We are driving multiple large data-led digital transformation programs for our clients worldwide which provide very exciting opportunities for these experts. Many of them are involved in consulting and asset-led data transformation programs that drive business outcomes for clients.  

Loading...

How are you tackling skill shortage or high attrition in the areas of data science, AI and analytics?   

In the last few years, the industry has witnessed high attrition, however we have taken several measures that has helped us reduce the phenomenon. At Infosys, we enable our employees to reskill and upskill themselves by leveraging our online learning platform, Lex, which offers industry-standard learning programs. Infosys Springboard is another digital learning platform. We have also accelerated automation in the organisation for efficiency and productivity and we are constantly re-calibrating our productivity baseline to stay competitive in the market. As part of our strategy, we are expanding our geographical base to include skilled people from smaller cities in the workforce. We recently set up four new offices in tier-II cities in India. We have also made large investments in expanding our local workforce in the US, the UK, Europe, Japan, China, and Australia.  

What are the most critical challenges involved in the data democratisation within an organisation and how can companies overcome the challenges?   

Loading...

Data democratisation is when an organisation makes data accessible to all employees and stakeholders, and educates them on how to work with data. Needless to say, data in an organisation needs to be mobile and available to all in order to be valuable. However, to collate, store, consolidate, manage, and distribute data from various sources with the required safety and ease is challenging both in terms of cost and effort. 

For data to be available to all, companies should establish the infrastructure needed to ensure that the right data is captured, is not corrupted and easily accessible, preferably on real-time basis. This requires integration of systems across departments and safety protocols that could be complex to achieve. They should also create the capability that enables people to use this data for analysis and insights. Even today, MS Excel is the most used analytical tool across organisations because of its simplicity. For more advanced tools, people lack the skill or the comfort required to make use of their full potential.   

On the other hand, democratisation of data could also mean high risks of data leakage or corruption or inaccurate insights if employees are not trained well or they lack the awareness or the culture to be responsible for the data they handle.