87% OF LARGE COMPANIES HAVE ALREADY INVESTED IN MACHINE LEARNING OR PLAN TO DO SO IN THE NEXT TWO YEARS

  • 58% of companies see data analysis as a strategic asset for their business, a figure that rises to 70% in the case of Spain
  • The companies interviewed agree on two barriers that prevent the adoption of machine learning: the lack of resources and talent. In addition, three out of four see it as a cost
  • 89% of managers have a basic knowledge of machine learning, although only half know how it works or remember success stories

Cloudera, the machine learning and data analysis platform optimized for the cloud, has presented a report on the status of machine learning adoption in Spain, France, Germany and the United Kingdom. The report indicates that 35% of large companies have fully developed this technology, while another 31% have done so in some specific departments for specific functions.

One third of the companies that have implemented machine learning in their processes affirm that they have already seen a return on investment. However, companies agree that there is a major barrier: resources and talent are lacking. In fact, 51% hesitate to implement machine learning, since they do not have the ability to do so.

Half of the large companies have indicated that the data is completely embedded in their organization, and that they would no longer be able to work without them, while another 36% indicate that they do use the data, but not in a way that is transversal to the company. In addition, 49% of businesses state that the information technology department is responsible for maximizing the value of the data, and in the rest of the cases the responsibility lies with the data scientists or the users themselves.

“Machine learning should be the next step for data-driven organizations. At Cloudera, we plan to accelerate our roadmap to help customers capitalize faster on the value of their data with machine learning, “said Romain Picard, vice president of Cloudera in South EMEA . In addition, he added that “this kind of solutions have the cost of developing internal skills and differentiation, in addition to making leaders aware of how machine learning is capable of driving companies”.

The barriers to machine learning development

Picard has indicated that “the first problem is the lack of knowledge among the leaders, and that the main cause that makes the machine learning to resist is an aspect of perceptions, since the businesses still do not trust that in the capacity of the machines of replace the human being in decision-making “. In this sense, 79% believe that people make better decisions.

On the other hand, there is no clear vision of what it consists of and what benefits machine learning offers. Decision makers associate machine learning with elements of artificial intelligence and automation: 44% have indicated that it has to do with computational algorithms, others that it is the process of creating systems independent of human intervention, and a third block associates it with artificial intelligence.

Finally, there is an essential lack of talent. Less than one in three companies has a stable team of data scientists with machine learning skills. In addition, only 52% of them stay updated with relevant and specific news from the information technology sector.

Adoption and use of machine learning

The adoption of machine learning is in full expansion, since 47% of companies are already investing and another 40% expect to do so in the next two years. In fact, the only technology that is more widespread than machine learning is data analysis, with a current investment of 54% of large companies.

69% of IT directors expect machine learning to have a relevant influence in their departments, and 31% expect it to be a transformational phenomenon. However, only 20% of managers think that it will have a significant impact on the sales divisions, even less in the human resources department: 17%.

The spokesman of Cloudera has indicated that “the use of algorithms allows to extract useful knowledge of the data. In this way, machine learning allows teams to work more intelligently, do things faster and routinely convert previously impossible tasks. ” However, 65% find problems in prioritizing where it will have the greatest impact, although apparently it will be in the information technology and R & D divisions

“At the time of implementing machine learning in the company, new problems arise such as the lack of talent capable of developing it, which is one of the main barriers that prevents this technology from spreading at a higher speed,” concludes Picard.

The Cloudera report, which analyzes the situation of machine learning in Germany, Spain, France and the United Kingdom, has been created on the interviews of technicians and innovators from 200 large corporations in sectors such as finance, services, retail, health, energy, industry or telecommunications, as well as the public sector. 81 of these companies have between 1,000 and 5,000 employees, 61 more than 5,000 and the rest more than 500.