Most companies do not manage to fully manage the potential value of data and analytics to improve business and struggle to reach 30% of the estimated value in some sectors.
In 2011 McKinsey drew up a report where companies’ incapacity to fully exploit the value of data, especially in relation to five essential production sectors, is highlighted: public services in Europe, health services in the United States, manufacturing sectors, retail and location based data. Five years later and in spite of the fact that some progress has been made in the location and retail sales services, production, public sector and health services continue to use less than 30% of the potential estimated value of data and analytics. This is what emerges from the new McKinsey report “The Age of Analytics: competing in a data-driven world“.
The critical factors to be overcome
The most obvious obstacles encountered by companies for the right optimisation of data and analytics are mainly organisational in nature and linked to the complexity in the choice of technological solutions and in attracting professionals and skilled workers required to implement efficient business processes.
Also the introduction of new types of sets of data (“orthogonal data”) can destabilise companies and make the ability to integrate difficult: however, the added value of data analysis lies in the very fact of its capacity to overcome the organisational and technological silos, allowing the discovery of new intuitions and the implementation of new models. For example digital hyperscale platforms manage to match purchasers and vendors in real time, transforming market inefficiencies; granular data can be used to personalise products and services as, for example, in the healthcare sector. New analytical techniques bring about discoveries and innovation and can also speed up decision-making processes.
A further critical factor is represented by the matters linked to legislation that places further obstacles in the way of optimising data, especially in the public and healthcare sectors. Lastly, in many cases, the lack of system interoperability in exchanging data determines the failure by complex, public and private organisations in the fully satisfying use of data governance strategies.
Recent progress in machine learning can be used to solve a large number of problems: for example, automated systems can provide customer services, logistics management, medical record analysis, and even news writing. The potential value of data is being performed everywhere, even in those sectors that are become digitalised more slowly. In the 2011 Report, the survey had shown that 45% of work activities associated with 14.6 trillion dollars of total salaries could be automated thanks to some technologies that could today be machine learning. According to a recent McKinsey report this enabling technology could redevelop and automate even 80% of working activities. One example to refer to is the translation sector. In the USA it is estimated that this sector could be a possible market of about 4 million dollars within the next decade. The estimate was made by calculating the number of graduates in the next decade, i.e. Circa 9.5 million individuals in the STEM sectors, who, in 40% of cases, will need platforms, services and professional translation systems to satisfy their own business needs that could be implemented via machine learning.
A further interesting aspect to look at is linked to the +50% growth in analytics processes starting from 2010. In this context, the most stimulating challenge that companies face is the complexity of data evaluation that is heterogeneous in relation to different categories of data, such as, for example, the ones that include behavioural data such as the acquisition of actions in digital and physical environments, transactional data in business relations, environmental data, or conditions in the physical world, monitored and captured by sensors, geo-spatial data etc.
As far as demand is concerned, data can provide ideas for various uses: some industrial sectors are changing completely as are some characteristics of a given market. In personal transport, ride-sharing services use geo-spatial mapping technology to collect vital data on the exact position of passengers and drivers available in real time. The introduction of this new type of data enables efficient, immediate correspondence that brings with it important economic advantages if analysed adequately. One tangible case is given by the Uber, Lyft and the Chinese ride-sharing Didi Chuxing platforms that can intercept the client’s demand with supply, positively rebalancing management of an unused resource in relation to the use of personally-owned vehicles (that in most cases are unused for 85-90% of the time), offering a quality service at a low cost and not having to make important investments to obtain a car. It is estimated in this sector that by 2030, mobility services, such as driving and car sharing could represent more than 20% of the total of the thousands of passenger vehicles worldwide. Consumers could save on car, fuel and parking purchases with a consequent economic impact that could go as high as 845 million million dollars, 2.5 trillion dollars of which worldwide by 2025.
Retail Banking is also a sector rich in data on customer transactions, financial situations and demographics; with efficient data governance, a potential economic impact from 110 to 170 billion dollars is estimated in the banking sector alone of developed markets and from 60 to 90 billions dollars for emerging markets.