To paraphrase an age-old poem, “Data, data everywhere–nor any insight to act.” In today’s consumer-choice driven market, especially post pandemic, profound changes both in business models and cost structures have made forecasting the future and creating scenarios while being agile, extremely critical. Gone are the days when rear-view mirror decisions by gut feel or at best analyzing siloed data would help organizations to stay competitive. In today’s scenario, with demanding consumers, increasing direct and indirect costs, ever growing competition, etc., it has become imperative for businesses to be forward-looking in decision making. Hence, business leaders are looking for tools, technologies, people and solutions which can help them make informed, futuristic, scenario-based, fact-based decisions. With a rising variety of data sources and reduction in computing costs, predictive analytics have risen to become one of the top agendas of business functions to drive and achieve business outcomes.
Predictive analytics is a subset of advanced analytics which analyzes the current and historical data and identifies trends and patterns to make meaningful and insightful predictions about the future using different techniques like advanced machine learning and deep learning. Many businesses are embracing predictive analytics to increase their efficiency and efficacy–while the pandemic has caused some discontinuities in historical data needing careful interpretation and handling. Enterprises are using predictive analytics across the value chain–viz. marketing, sales, operations, customer service, supply chain, talent, IT, etc., to achieve superior business value.
While there are myriad applications of predictive analytics, we intend to highlight some experiences where it has helped businesses in achieving superior outcomes–higher topline, profitability, reduced cost to serve or enhancing customer experience, etc. –where clients have leveraged predictive analytics to differentiate their offerings leading to a competitive advantage in the market.
A leading financial company had a historical data-based view while targeting their large customer base for 30-plus offerings which led to multiple offers to the same customer over multiple channels like IVR, voice calls, emails, SMS, etc., eventually putting the customer off. Predictive analytics offered a customer-centric view which helped identify the need of a customer and offer products and services basis their need through their preferred channel and message. Behavioral segmentation was done to classify customers into micro-segments and for each sub-segment a next best product recommendation engine designed to cross-sell and up-sell services/products basis customer affinity. The result was a significant improvement in customer experience on account of this personalization as well as customer engagement and topline as a result of positioning of offers/products relevant to the customer’s propensity based on predictive analytics.
A leading global manufacturing company was struggling with variability in the costs of raw materials, leading to margin leakage. After analyzing demand-side and supply-side factors, various machine learning models were trained to predict raw material prices several weeks in advance for both domestic and international markets. It helped the business teams take decisions on what to buy, when to buy and from where to buy. In terms of outcomes, the business saved significant costs in procurement and provided forward assurance on gross margins to the enterprise.
A media house was pricing advertising space across their different editions using a rate card and inputs from the sales team, largely driven by gut. Using predictive analytics, price optimization and recommendation models were built, to give an optimal price for an advertiser for different inventory positions within a newspaper. Predictive analytics brought consistency in sales pricing and increased the revenue by around 10 per cent, while bringing a small increase in the market share too, and empowering the sales team with analytics on their fingertips to close deals with clients.
An oil distribution company was struggling to manage margins due to its crude oil pipeline capacity and fuel run rate. Through an understanding of the supply chain and factors affecting it, an iterative process was followed to zero down on the deep learning models to forecast the pipeline capacity. The predictive algorithms aided the company in reducing the spot market trade and helped in gaining $0.5 per barrel, adding up to a monthly savings of around $10 million.
A leading technology brand is now using predictive analytics across their marketing function for data driven insights to invest in marketing campaigns. Traditional way of budget allocation across media channels was basis intuition and experience. Now, with advanced marketing mix models, they understand the ROI of historical campaigns and the extent of sales generated due to these marketing activities. So, it is possible now for the CMO to reasonably predict success while planning for next set of campaigns and allocate future budget to media channels optimized to achieve higher ROI.
With new innovations in technology, businesses are harnessing the power of ever-growing amounts of data and use of predictive analytics to achieve superior business outcomes. It helps them gain sustainable competitive advantage with tangible insights which lead to action across the business value chain.