Big Data Analytics is a buzz word today and the skills in its related technologies such as artificial intelligence, machine learning, cloud computing and robotics for automation are driving the industry’s growth. As per the recent report, the IT companies are laying off their employees in thousands and at the same time looking for these skills. It says, “After thousands were fired by Infosys and Cognisant in the past six months, as many as 200,000 more may be at risk of losing their jobs in the next year as companies seek people with new skillsets, according to recruitment experts.” Two decades ago the people who face layoffs today had skills which were similarly valued. I lived through the IT revolution that began in the mid-1990s with the spread of the Internet into business and consequent changes in the way the business was done. In all the transitions that took place in these decades what survived is people who knew how to use technologies for business benefit.
There appears to be an uncanny parallel between what happened to the IT and what is happening to the Analytics today. Today Analytics skills are highly valued simply because they are in short supply. But will they remain the same in the coming years? The clichéd answer to such questions is to reskill. Of course, one needs to keep pace with rapid changes taking place in technological space, but it is not practicable to be competitive in changing technological skills. It is, however, possible to be competitive in using technology for the business benefit. If one is located in the business world, this is precisely the managerial skill that will always stand one in good demand.
Today, the typical way in which companies are organized to use analytics may be depicted by silos comprising Strategy, Data Science or analytics team and User Department teams. Many companies have a small in-house Analytics team representing these functions. Today, they are silos because there is a communication disconnect between them, each speaking its own language. The strategy typically conceives an analytics problem ideally in consultation with the user department and proposes it to the analytics team or data science team. The latter would take stock of data within the company and outside, formulate variables and try modelling the problem which, in process, undergoes a change from the original. There could be several such iterations and corresponding changes between the analytics team and strategy before it is cleared for execution. When it is executed and the solution is offered to the user department, it could be completely irrelevant and unusable, adding itself to worrying statistics of failures of analytics projects. It is, therefore, that Gartner reported in November 2017, that 60% of big data projects failed. A year later, Gartner analyst Nick Heudecker said his company was “too conservative” with its 60% estimate and put the failure rate at closer to 85%. Today, he says nothing has changed.
Gartner isn’t alone in that assessment. Reports of failure of analytics project is a legion: In July 2019, VentureBeat AI reported that 87% of data science projects never make it into production, in January 2019, NewVantage survey reported that 77% of “business adoption” of big data and AI initiatives continued to represent a big challenge for business, (which meant three-fourth of the software being built is apparently collecting dust), in January 2019, Gartner also came out saying 80% of analytics insights will not deliver business outcomes through 2022 and 80% of AI projects will “remain alchemy, run by wizards” through 2020, in November 2017, CIO.com lists seven sure-fire ways to fail at analytics. Tom Davenport, a senior advisor at Deloitte Analytics stated, “The biggest problem in the analysis process is having no idea what you are looking for in the data.” In May 2017, Cisco reported that only 26% of survey respondents were successful with IoT initiatives, which meant 74% of them just failed.
This takes me two decades back when we used to read exactly similar things about the IT projects. When I ventured in to implement ERP in my company, one of the biggest projects in the Asia Pacific the frightful dictum was in the air that over 80% ERP projects fail to deliver business benefits. We averted this aftermath by being extra conscious about the business objectives at every point and in the course even realised that it was not technology alone but adequate technology insight and lots of business acumen that is what is required in any manager. The same could be said of the analytics, which of course has become big data analytics. Analytics reflects the trend towards automation; embedding solution into business processes. The challenge would shift from technology or analytics skill to essentially managerial skill which embeds knowledge of what various technologies are capable of, and which one would best serve one’s business strategy. People who developed these skills have risen to head businesses and are not the potential victims of layoffs.
The future manager thus will have to be well-grounded in business, having sharp strategic acumen and knowing ‘enough’ of analytics. This ‘enough’ can never be specified enough. The possible approach is to know the analytics technologies in vogue thoroughly enough by working them in hands-on mode. It should necessarily include Stats and Maths as well as coding in R, Python and SAS as the foundation. It should also include ETL technologies, visualization tools, predictive and prescriptive analytics, AI and Machine Learning, Deep Learning, Cloud computing and security, IoT Analytics. In management domain, the approach should be to build strategic perspective with the help of capsuled courses in most functional areas—Economic, Accounting, Marketing, Finance, Operations, OB/HR—and integrative capstone course capped with an opportunity to work long enough as an intern in a company. It should be so designed as to stress the problem areas so as to generate use cases for analytics. The managers should also be formally trained in strategy and equipped with supplementary skills of design thinking, storytelling with data, data strategy, analytics project management and ethical aspects in managing data, etc. Depending upon their choice of industry, they should also have an opportunity of specialization in certain areas. The biggest anxiety of placement is taken care in this process of ensuring the output quality.
This is precisely what we offer in our Big Data Programme.
This kind of integrative training should make future managers break prevailing silos and insure analytics projects from failure as well as themselves from the risks of layoffs.
To apply for PGDM-Big Data Analytics programme at Goa Institute of Management, click here.
