Today, organisations across sectors deal with vast quantities of data, ranging from customers’ personal details, browsing habits, and purchasing histories. By leveraging analytics, businesses are able to get insights into the business. They are also able to predict what is likely to happen in the future, by looking for patterns in the information they already have. But moving from descriptive analytics to prescriptive analytics is not easy. It involves new roles, skills and strategies.“The most critical step while starting the journey is choosing the right problem – this involves identifying an issue which is business-critical, and setting boundary conditions in such a way so that the problem is solvable in 8-12 weeks. Ensuring success in the first couple of problems is important – and for the same, getting in external expert help is the right strategy,” said Nishant Nishchal, Partner, Kearney.
Organizations can move up in their Analytics maturity level by automating and building self-service solutions for reports/dashboards. The automated and self-service solutions will reduce the effort spent by Analytics teams in building/managing descriptive analytics solutions and allow them to focus on predictive and prescriptive analytics solutions.
“Business users should be able to subscribe to and receive reports in an automated manner (ex. email) at required frequencies. For ad-hoc requests, a self-service tool that allows users to download reports on demand or view dashboards which allow users to drill down and analyse root causes will help organizations move from descriptive/diagnostic analytics to predictive analytics,” said Subramanian M S, Vice President and Head of Category Marketing and Analytics, bigbasket.
By design descriptive and diagnostic analytics are lookback solutions. They answer the ‘What’ (ex. what is my sales in the Eastern region) and the ‘Why’ (ex. why is my customer attrition % increasing) questions.
“Spending effort/time of the Analytics team in answering these questions limits their ability to build and deliver value using forward looking predictive and prescriptive analytical solutions. Answers to the ‘what will happen’ predictive questions (ex. what will be the default rate of my customers) and the ‘what should we do’ prescriptive questions (ex. how can I improve my customer retention) deliver more impact to the organization and its stakeholders,” Subramanian further added.
For this upgrade to the new and advanced level of analytics, businesses need to build new skills and roles internally while taking help from external partners. Organizations will need to build ‘trilingual operators’ over time – people who understand the business, the technical side of the problem, and the basics of data science & system integration. Such operators will be the pivots for driving analytics in the organization. The start will be by bringing together these capabilities in the form of a team followed by an amalgamation of capabilities in individuals.
“As the Analytics team moves to deliver more predictive analytics, the team members will need to enhance their business knowledge, functional knowledge (finance, marketing, sales, etc), their analytical skills (operations research, machine learning, statistical skills), and technical skills (cloud technologies, advanced programming using Python, Scala, parallel processing using map-reduce, etc) to build, deliver and maintain predictive analytical solutions,” Subramaniam explained.
Deepak Pargaonkar, Vice President, Solution Engineering, Salesforce believes that managers need to train and empower employees in a digital-first world. And executives and boards need to be digitally savvy enough to navigate pressing issues on an ongoing basis.
“In an IDC survey, 83% of CEOs said they expect to be working in a data-driven organization, but a mere 33% of employees are comfortable using data analytics to support their decision-making,” Pargaonkar said.
Investing in employees’ learning and development for this upgrade helps an organisation scale up quickly.
