Data Analyst or Data Scientist – who would you like to be?
Photo from the site Unsplash. Author: Christina @ wocintechchat.com
I had the opportunity to work as a professional data analyst (Data Analyst), and data researcher (Data Scientist). I think it would be useful to share experience in each position, indicating the key differences in everyday tasks. I hope that my article will help determine what is right for you. And for those who are already working, perhaps after reading they will want to change their position. Some start with data analysts and then move on to researchers. Not so popular, but no less interesting is the path from a researcher in low positions to an analyst in a senior position. Both posts have their own characteristics and require certain skills that you need to know about before taking the next big step in professional development.
Below, based on my experience, I will tell you what it means to be a data analyst and data researcher, and I will answer in detail the most common questions about each position.
Data analyst
If you want to describe data for a past period or current moment and present key results of search, full visualization of changes and trends to stakeholders, then the position of a data analyst is suitable for you. The posts mentioned have common features that I described in another articlecovering the similarities and differences between the skills needed for these positions. Now I want to show how the role of the data analyst in comparison with the role of the data researcher is felt. It is very important to understand what to expect for these specialists in their daily work. The analyst will interact with different people, communicate a lot and maintain a high pace of task execution – higher than what is required from the data researcher.
Therefore, the impressions received in each of the posts can vary greatly.
Below you will find answers to the most common questions about what data analysts face.
- Who will have to work with?
Mostly with stakeholders of companies that request data compilation, visualization of findings and reports on the results. Communication is usually oral or via digital channels: email, Slack and Jira. In my experience, you have to closely interact with the human and analytical components of the business, and not engineering and production.
- Who is provided with the results?
Most likely, the aforementioned stakeholders. However, if you have a manager, you report to him, and he already transmits data to stakeholders. The option is not ruled out when you collect a pool of requests, compile a report on them and present it to stakeholders. For reporting, you may have tools such as Tableau, Google Data Studio, Power BI, and Salesforce that provide easy access to data, such as CSV files. Other tools require more technical effort – making advanced database queries using SQL.
- What will be the pace of work on the project?
Significantly higher than that of data researchers. You can prepare several pools of data (queries) or reports daily and large presentations with conclusions weekly. Since you do not build models and do not make forecasts (usually), and the results are more descriptive and situational, work goes faster.
Data scientist
Data researchers are quite different from data analysts. They can use the same tools and languages, but the researcher has to work with other people on larger projects (such as creating and implementing a machine learning model) and spend more time on it. Data analysts usually work on their projects on their own: for example, one person can use the Tableau panel to present the results. Data researchers are entitled to engage several engineers and product managers to efficiently complete business tasks using the right tools and quality solutions.
- Who will have to work with?
Unlike a data analyst, you have to interact with stakeholders only on some issues; for other issues related to models and the results of their use, you will contact data engineers, software engineers, and product managers.
- Who is provided with the results?
You can share them with stakeholders, as well as with engineers who need to have an idea of the finished product, for example, to develop a UI (user interface) in accordance with your forecasts.
- What will be the pace of work on the project?
Probably the biggest difference in the perception and functioning of these posts is the amount of time for each project. The speed of data analysts is quite high, and data researchers may take weeks or even months to complete a project. Model development and preparation of data researcher projects are long processes, as they include data collection, exploratory data analysis, creation of a basic model, iteration, model tuning and extraction of results.
Conclusion
Photo from the site Unsplash. Author: Markus winkler
Analysts and data researchers use the same tools, such as Tableau, SQL and even Python, but their professional tasks can be very different. The daily activities of a data analyst include more meetings and personal interaction, requires pumped soft skills and quick project execution. The researcher’s work involves longer processes, communication with engineers and product managers, as well as the construction of predictive models that comprehend new data or phenomena in their development, while analysts focus on the past and current state.
I hope the article was interesting and useful. Thanks for your attention!
