There is no reason for Data Science projects to be complicated

There is no reason for Data Science projects to be complicated

Understand our method to work with Data Science as an agile project

It has been estimated that by 2020, each person on earth would create 1.7MB of data every second, according to the annual report “Data Never Sleeps”. This volume makes data the new source of global capital – data is being considered the new oil – as reported by The Economist magazine.

Today, Data Science projects do not have to be complex. And what makes this difference? Among all the factors and premises, we highlight:

  • An appropriate work methodology;
  • An agile approach without reliance on proprietary platforms;
  • The correct definition of the problems to be solved, and to solve them one at a time;
  • A team with experience in the most diverse disciplines of Data Science: analysts, engineers, and data scientists.


Data or information?

Here at CINQ, we work with Data Analytics Sprint, a specific methodology for Data Science projects that has been helping our clients to get faster answers and insights from their data. The Data Analytics Sprint was designed to be run in 10 days.

There is no reason for Data Science projects to be complicated

During this period, our team works together with our clients’ teams to:

  • Understand the problem and preferably, define one or more KPIs (performance indicators). Here, by working together, we prioritize the problems we want to solve;
  • Minimally understand the structure and data that were made available;
  • Raise the hypotheses, that is, ask the right questions;
  • Perform an exploratory analysis of these data (without torturing them) and, finally;
  • Validate and present the results to our client.

This process can be performed several times. At CINQ, we use an agile and lean approach, as we already do in the development of digital product projects. That is:

  • Short delivery cycles (2-week sprint);  
  • Prioritization of the value to be generated for the business; 
  • Focus on the hypothesis’ analysis and formulation, and not on the configuration of proprietary tools and platforms for data analysis.

In short

  1. Which problem do we want to solve?
  2. Understanding and preparing data
  3. Raising and validating hypotheses
  4. Exploratory analysis and modeling
  5. Displaying the results

As you can see, a Data Science initiative does not have to be complicated or time-consuming. CINQ can help you ask the right questions and generate relevant information from your data, get to know our Data Science and Machine Learning service: Data Agility.