Since the beginning of the last decade, when access to high-performance computing became affordable, a double race related to a data-based strategy began: that of generation, and that of exploitation. The first has to do with the production of a continuous flow of information from the sources, that is, the activity of the clients during the consumption of the products and the use of the services. For example, financial transactions such as the purchase and sale of securities on the stock market, or banking operations of individuals. The second of the elements, exploitation, has to do with the development of algorithms from the analysis of the source data. In the financial example, machine learning models allow the construction of algorithms for the identification of fraudulent operations, or the detection of credit non-payment risks. Also, the detection of “good clients” to whom loans are granted without a previous solvency study, since the bank has done it, prevents it from the data available to it.
Barriers to growth: technology and human resources
Both facets of data, generation and exploitation, have been and continue to be two handicaps not only for growth, but also for the survival of many companies. The generation is relatively simple compared to the second within the complexity involved in setting up a scalable computing infrastructure for a growing number of clients. However, exploitation imposes a whole series of requirements to be able to put it into practice. The first of these is the collection of data in a structured way, something that must be done from the generation stage, as the data is received from the sources. The second requirement is to have qualified personnel, analysts, specialized in information processing. Together with the data engineers, they design the recommendation algorithms, sales projections, etc. And also the dashboards for decision-making in the boards of directors of companies.
Survival and competitiveness
From this perspective, there are various challenges for companies, small and large, new and well-established. One of them is that of survival, due to the rapid change in business models driven by digital.
In a data-driven strategy, exploitation imposes a whole series of requirements to be able to put it into practice
The second is not only to maintain, but also to increase competitiveness against new players that build their infrastructures from the Big Data and Cloud Computing paradigm. These new companies are much more agile due to their ability to immediately adopt new technologies, while existing ones have to face the cost of reconversion and recycling, which affects both the technological infrastructure and the training of human resources.
Big data to the rescue
Coping with this situation involves designing a data-based strategy that enables the exploitation of the large volume of information generated by customers. Big Data arises from the moment the technology is available to squeeze data in real time. The origin is in the exponential increase in computing capacity: much more powerful computers at much lower prices. In computing jargon, this behavior obeys the well-known Moore’s law, by virtue of which computing capacity doubles every 2 years for the same cost of computing resources. This has been going on for fifty years, and we know that it will continue for at least ten more. Which means that in the last two decades the capacity of computers and laptops has multiplied by more than 1000 without the need to increase economic investment, which has meant that the SME and small business sector finally has the opportunity to compete. at the level of the great. The reality is that a large part of our business fabric is at an early stage of developing a data-based strategy, which is conditioning its competitiveness, especially in the global market. The emergence of the new generation of companies derived from the startup model is what is most driving the rapid renewal of the business fabric. It is a model that allows rapid growth with a small financial investment. And this is not only because of the use of technology, but because they have placed the value of data at the center of business strategy. This is learning that pre-existing companies can and should apply to reinvent themselves and create new business opportunities.
By Bernardo Ronquilo, director of the Master in Data Analytics at Loyola University and Principal Software Architect and Senior Data Scientist and Software at Universal DX.