The Cross-Industry Potential of Predictive Analytics

Several factors have led to the exponential rise of predictive analytics and its evolution into more sophisticated models such as prescriptive analytics. In a highly competitive business environment, enterprises are eager to stay a step ahead of the curve, carefully assessing future possibilities to understand the next-best action.

On the other hand, global data volumes are also rising. Both customers and employees (an enterprise’s internal customers in today’s experience-centric landscape) are ready to share their information if they perceive a direct value-add. There is also a large number of devices and machines gaining network capabilities, creating a highly connected and constantly data generating Internet of Things (IoT). It is estimated that there are over 10 billion connected things across the world, as well as 3.7 billion Internet users contributing to growing data volumes.

Enterprises can leverage this data to make better business decisions and innovate without fear of low adoption. Predictive analytics takes raw data (structured and unstructured), passing them through sophisticated analytical models. The insights are then converted into a human-readable format, represented on a dashboard so that business users could derive genuine action-points. Unlike historical analytics, predictive models could provide a glimpse into the future, identifying risk, new opportunities, and emerging problem areas. Prescriptive analytics takes this a step further, directly recommending the next best step without any need to draw inferences.

Globally, prescriptive and predictive analytics accounted for a modest USD 5.72 billion market in 2017. From this, it is expected to grow at a 19.6% CAGR, crossing USD 28.7 billion by 2026. This is primarily due to the industry-agnostic potential of these technologies. Instead of being focused on vertical-specific use cases, analytics could transform how every function across industries operate, much like the internet helped to build the foundation of the digital era. Actually, if data is the new oil, analytics is the machinery that makes the difference between idle-sitting data mines and lucrative business.

Despite the wide availability of data, effectively mobilizing, and monetizing in-house data repositories could be a challenge.

Young businesses are uncertain about the possible RoI from analytics investments, given that their data stores aren’t yet adequately sized. Even larger organizations struggle with fragmented data silos,historical datasets lying in legacy infrastructure, and the absence of skill sets to accurately pinpoint their data potential. As a result, several potential use cases across industries risk being overlooked.

This holds back incumbent organizations from innovation, even as digitally-born disruptors pivot their business on data to push their value proposition. A 2019 survey found that 69% of organizations have not created a data-driven organizational roadmap, and 53% do not even view data as a business asset. This prevents enterprises from realizing several opportunities and use cases that are already at hand.

Risk assessment has always been a problematic area for insurance carriers, given the myriad elements that contribute to insurer risk. Data collected via IoT can help make more informed decisions around risk assessment, accurately defining the insurance price point.

Further, predictive analytics could also help in claims verification processes. For example, data collected from drones, or image analytics via smartphone captured photographs to assess the actual degree of damage to property. Not only does this eliminate the need for manual inspection, it also significantly cuts down on inspection and approval timelines, leading to a better customer experience for the insurer.

In the long term, carriers could hope to garner significant savings as manual errors in risk assessment and claims verification are completely eliminated by analytics.

The banking sector has been always tech savvy and traditionally strong on collecting tons of data. There is tremendous potential to use predictive analytics to identify and arrest customer churn, advisor churn or determine credit rating for new customers where credit history is not available. Operations across the back office and middle office can be transformed by predictive analytics, speeding up task processing via automation, and ensuring accuracy via error-free analysis.

Finally, wealth management could gain significantly from predictive models, scanning individual customer risk appetites to recommend the best possible portfolio. Market data can be entered in real-time to dynamically update portfolio components as per the latest and most relevant risk analysis.

Anticipating demand in the Travel, Transportation and Hospitality (TTH) sector

With so many TTH consumers moving online, there are immense possibilities for leveraging their online data footprint to optimize service offerings.

Event and weather patterns, for instance, could indicate the most popular routes at specific times of the year, enabling airlines to roll out additional flights and tap into this latent demand. Hotels could also study booking patterns, to predict “off” and “peak” periods, aligning the marketing campaigns accordingly. Rate optimization is another use case in the TTH sector, where analytics engines can monitor competitor rates in real-time to preemptively optimize product/service pricing.

In other words, predictive analytics will help organizations go beyond staid industry best practices and adopt more agile business models.

Arguably, retail was among the earliest industries to embrace predictive analytics. This is perfectly captured in online recommendation engines, which are now a staple in e-commerce. Based on purchase history, browsing habits, wishlisting patterns, and past reviews (combined with personal data collected via social media and third-party databases) retailers can recommend the perfect product for every customer type.

Predictive analytics can also unlock bottom-line improvements in retail by eliminating inefficiencies in the physical product supply chain. By aligning supply to predicted demand levels, retailers can cut down on inventory waste and drive greater profitability.

Declining unemployment rates and high demand for specific skill sets have made recruitment a major challenge for leading organizations. HR is regularly under pressure to shorten the time-to-fill for critical job roles, without compromising on candidate quality. Predictive analytics can help assess candidate persona, experience, and skills, and accurately match these traits to available vacancies. The recruiter would receive an actionable notification, recommending possible candidates, while job seekers can quickly gain meaningful employment.

Also, predictive analytics can evaluate employee feedback to detect early warning signs of disengagement, helping HR to take course-corrective actions so that attrition is kept at a minimum. Together, this will lead to better candidate quality, more accurate employee-employer culture fitment, and sustained engagement, driving greater productivity.

It is important to remember that any new technology will bring a complete paradigmatic shift, giving rise to new challenges and concerns. For successful deployment organizations must ink responsible implementation roadmaps, factoring in several risks.

If historical data sets are biased, it may be perpetuated across future decisions. This makes it critical to constantly update and harmonize input data with an eye on objectivity and equitability. Further, data sharing across platforms and interfaces could open an enterprise to security vulnerabilities. This should be addressed via compliant and resilient analytics model creation. Also, every analytics project should be envisioned from an ethical standpoint so as not to encroach upon individual privacy, preferences, and rights.

Keeping these in mind, enterprises can explore three possibilities for deployment when embarking on their analytics journey:

In the future, every business across industries will be built on a bulwark of data. Analytics will be the key in driving this transformation, leapfrogging young companies into the digital era, even as large organizations pivot toward new and more promising terrain.