Who is a data analyst ?





The term "business analyst" and "data analyst" is used interchangeably quite often by organizations. Depending on the size, function, and domain of the organization, the definition keeps on changing. But, here we will talk about "data analysts" only.

A data analyst is an individual who investigates, cleanses, manipulates, and models data with the goal of discovering useful insights, conveying conclusions, and thus, supporting decision-making.

What skills does he possess ?

Largely , a data analyst possesses these abstract skills -

  • Business domain knowledge - I can't stress it enough - KNOWING THE BUSINESS DOMAIN IS INDISPENSABLE! It is the understanding of a particular industry. I think domain knowledge for an analyst is as important, if not more, as the technical skills. That is to say, it can not be ignored at all.
  • Technical skills - These are the primary skills for a data analyst. These include the techniques and tools required to cleanse, process and visualize data. Certainly, these are primarily looked for in a candidate.
  • Communication skills - It is very important for an analyst to be able to communicate their insights and findings with the senior members and/or with the clients properly.

A deep dive into the technical part

  • Analytical skills - Although these are not necessarily the tools or techniques for analyzing data, but within the 3 broader categories, they fall under the technical skills. These consist of logic forming aptitude, problem-solving ability, and critical thinking.
  • A spreadsheet tool or a database language - e.g. Excel or SQL. This is a weird comparison. However, I think they should know at least one of these as a foundation of their data analytics technical skills. This is because it is essential for them to understand how data is stored and organized (rows, columns, tables). On top of that, they would then understand basic data crunching methods like grouping ( Pivot / GROUP BY), sorting ( Excel sort / ORDER BY), filtering ( Excel filter / WHERE), subsetting ( Deleting unimportant columns / SELECT) etc.
  • A data visualization tool - After hard-fought days of data preparation and number crunching, one needs to convey insights from the data to (usually) non-technical people. Consequently, he needs to be able to represent the trends, patterns, correlations among different fields visually. Some hands-on on one of the popular visualization tools come in handy. The most popular viz tools today are - Tableau, Microsoft Power Bi, QlikView, and Amazon Quicksight.
  • A statistical tool - Advanced tools like SAS, R, and Python. These tools have great inbuilt functionality for statistical computing as well as data processing. Also, they can handle huge data. Hence, they are suitable for both descriptive analytics and predictive analytics.
  • Programming knowledge - For working with either of the above mentioned advanced tools, one must know how to program. In addition, a programmer is likely to have good logic building.
  • Applied statistics - Applied statistics is mandatory when it comes to hypothesis testing and predictive modeling.

A day in a life of a data analyst

Contrary to popular opinion, a data analyst doesn't just "create charts" or "build models" all the time. Most people would be surprised to know that 80% of the times, an analyst spends his time in data preparation for the "model building" (or other similar end goals). This is what a data analyst does throughout the day -

  • Identifies the problem from the business context.
  • Collects and prepares the data for analysis.
  • Exploratory data analysis (EDA).
  • Data munging, data mining, data wrangling, data manipulation etc.
  • Visualizes the trends and patterns, and creates final reports.
  • Conveys the insights with seniors and clients.