Trends around adoption of data analytics across industries

Businesses are evolving and changing at an unprecedented pace partly due to traditional business models being disrupted by tech-enabled business models such as aggregators, operators etc., which cater to newer consumption trends like the rise of a sharing economy. The ease of reaching out to customers directly has increased manifold in recent times. In addition, uncertainties surrounding global policies, currency fluctuations, etc., have clearly necessitated companies to look very deeply into existing data pools that can help in any decision-making support.

While Big Data innovations have encouraged companies to churn large amounts of data using high computing powers to extract actionable insights, they have also led to a reduction in the marginal utility of existing data pools and have established the need for finding newer, more granular, richer and wider data pools to get deeper insights. For example, consumer product companies have moved away from traditional sales estimation based on aggregate sales data available at a distributor level to more advanced demand sensing methods that harness channels and, in some cases, even store-level data.

New start-ups are working on connecting retailers, brands and consumers by leveraging technology to empower the entire ecosystem. Social media sentiments, video content is increasingly mined to get more insights into buyer behaviour. Even traditional sectors such as automotive are embarking on unraveling hidden insights. For example, sourcing contracts for large commodity purchases such as steel earlier would be based on historical price trends, volume etc. But they are now looking and scanning the web for a host of very unstructured data such as global investor sentiment, policy changes etc., to build complex predictive models to embellish their sourcing strategies.

If you establish a plane of continuum of maturity of analytics adoption (low to high), most companies have clearly identified their top use cases for analytics leading to scouting for new and rich data sources and then building predictive models to help their agenda. The last five years or so have had a very different trajectory of adoption of analytics across different functions and industries compared to the past.

Companies have moved from dashboarding solutions to process automation to intelligent automation and in some cases to advanced real-time AI-based solutions very seamlessly. This has occurred primarily due to the rise of cloud-based platforms developed by industry big wigs. Google/Microsoft/Amazon have all developed frameworks and offer analytic services on the cloud thereby, allowing customers to focus on use-case definition and not worry about the heavy lifting involved in algorithm build and maintenance. For example, image recognition services of Google coupled with an RPA engine are very successfully used in scanning invoices and reducing the overall turn-around time for invoice processing by more than 40 percent.

With the continual increase in the complexity and size of the data pool, companies need to wrangle big data with increasing ease, have an assurance on the quality and have the right judgment on the use of the data. The gap between technologists who built the algorithms, the data engineers who are responsible for wrangling the data and the business stakeholders who rely on the insights has given rise to the need for Unified Analytics Platforms. These platforms have significantly reduced the effort of building and maintaining large data platforms, and this has meant that both structured and unstructured data can be easily handled and retrieved to run advanced simulations for decision support. Clearly the proliferation of data and analytics platforms on the cloud has significantly increased the adoption of analytics across industries.

Delivery methods of analytic solutions have also gone through a metamorphosis. Analytic solutions are no more viewed as independent of the workflows which they operate in. The clear trend is to have on-demand augmented decision making as a part of the workflow. For example, behavioural economics-based solutions to manage sales force performance is gaining popularity, in these kinds of solutions, short actionable insights are provided to the sales teams on the ground by churning through a universe of data such as sales trends, product assortment, stocks, competitor behaviour etc., and a crisp recommendation as a part of the day in the life of a sales manager is made.

Companies need to be cognizant about talent availability in the market and within the partner organizations. To be conscious of what talent is to be harnessed and retained within the company and what to outsource would be the key in gathering competitive advantage depending on the maturity of the business. Each industry has a different view on this. Traditional brick and mortar companies are leveraging expertise from outside the firm in defining the entire end-to-end analytic journey, however, more mature industries in retail, consumer products etc., are leveraging boutique skill sets from outside to augment their existing talent pool.

The establishment of the Center of Excellence (COE) by many companies has increased the adoption of analytics. Attracting the right talent and getting the employer branding right is very important to attract and retain top-notch talent. Most importantly there is a trend of proliferation of analytical skills in different business functions beyond just the CIO’s office. This democratization of analytical skills has vastly helped in the adoption of data analytics.

Organizational structure is also changing and the creation of a Chief Digital Officer who sometimes reports into the business heads or a transformation office is significantly helping the increase in adoption of analytics in industries.