TECH | Sep 14, 2017

Name of the medicine: Data Science

If Data Science were a medicine: read the instructions and warnings for use very carefully


Data Science is an interdisciplinary field of study useful to extract knowledge from the analysis of data of various nature, shape and size. The main active ingredients in the drug are therefore varied: Statistics, Data Mining, Machine Learning, Operational Research, Information Theory, Programming and Big Data.

Data Science has been defined by Jim Gray as a “fourth paradigm” of science that can be empirical, theoretical, computational, and now, with the advent of digital technology and the exponential growth of information available, driven by data analysis.

So much in vogue today – along with the new professional role that proposes its use in the company, the Data Scientist – it is often confused with Business Intelligence, Big Data Analysis, Data Mining, and anything that might contain the word Data. This happens, especially in the case of new, particularly fascinating drugs that are however still lacking experimentation. Just as it happens for emerging jobs that make predictions such as “a Data Scientist in every business”. Just like saying a doctor in every home, since it would be handy.


The Data Science medicine can be used in all those pathologies in which the need arises to decide company strategy based on data and not just perception or hearsay. Its use is recommended in situations where data is not collected and analyzed just to aid in better understanding certain phenomena, but to guide them in the best direction.


The drug should be taken when the time arrives that you need to make sense of the mountain of data you have collected and stored, which is often not even analyzed by businesses, but which instead, by building prediction models based on the data and on data-based machine learning, has an impact on strategic choices.

Data Science is not advised for anyone who collects data for the sole purpose of storing it, neither for those who gather it for the pleasure of analyzing it for their amazement or otherwise, nor for those who do not intend to modify corporate governance as a result of their “exploration” and correlation of said data.


The beneficial effect of Data Science can be amplified if combined with the ability to build forward-looking predictive models needed to create a business strategy that can leverage data and activate data-governance models. Its effect may be reduced or nullified where data quality logics are ignored and especially if care is not introduced into a structured and organic therapeutic plan.


How much

You must take the medicine by following the instructions of the Data Scientist to the letter and continuing for the entire prescribed time. The dose can be gradually increased over time, monitoring, along with the Data Scientist, any undesirable side-effects. It can also be taken in large doses as long as the patient is aware of the side-effects and is able to manage them.

When and for how long

Data Science can be taken regularly. Long-term use is recommended. Avoid unexpected and unplanned interruptions: if this should happen, it is recommended that you resume taking Data Science as soon as possible.
The benefits of Data Science may not be immediate and medium to long-term intake is required. The medicine may be addictive.


Data Science can be taken several times a day and at any time of day. It is ideal when taken with other medicines such as Big Data, Artificial Intelligence, Data Mining, Machine Learning, Predictive Analysis, and Statistical Analysis without which you will not notice the desired benefits.


Excessive doses of Data Science may lead to common-sense decisions based on data and its analysis when applied to the company’s strategic management. Excesses of Data Science may even lead to a change in business model and therefore strategic vision, which have probably been rooted and established in the company for many years. In these cases, you need to contact the company management, aiding the process with a dose of Data Storytelling useful to lowering the level of resistance to change.


Although usually well tolerated, Data Science can cause side-effects such as irritability in subjects who are not inclined to an in-depth reading of the present and future corporate situation; nausea and migraine in those who take note of the need to modify the business model and hysterical crises that may turn into genuine rejection crises in corporate staff involved in the changes.

Sonia Montegiove – Stefano Epifani