Would you be a data analyst or data scientist, instead?

• Read original article here

Having worked as both a professional data analyst and a data scientist, I thought it would be insightful to highlight each position‘s experience along with some key differences in how they feel every day. In the end, I hope my article will help you decide which role fits you best. If you’re still in one of those roles, you may want to move to another.

Some people start as data analysts and then go on to becoming a data scientist, while, as a less common but still very prominent path, they go from a position of a non-senior data scientist to a senior data analyst. There are many definitions and overall experiences for and a job that is crucial to know as you move your next big step.

I will illustrate below how he thinks he is both a data analyst and a data scientist. I’ll raise common questions about each position and address them according to what I’ve experienced — besides some close peers in each sector.

If you want to explain past or current data while presenting key observations, changes, and patterns, and finally visualizing data to stakeholders, then the role of a data analyst is ideally suited to you. Although there is some overlap between the two positions that I pointed out in another article (linked at the end of this article) covering the discrepancies and similarities between the skills of these two roles, I wanted to take some time now to go over how it feels to be a data analyst versus a data scientist.

Knowing what to expect from your everyday routine in this field is important. You should expect to work with different people, have different (more) contact, and move faster than a traditional data scientist.

Below I will pose some common questions along with their corresponding answers — shed some light on the experience of the data analyst.

— You’ll deal with mainly company stakeholders asking for data to be pulled, perspective visualizations, and studies. Communication by using tools like email, Slack, and Jira can be expected to be both verbal and digital. You’ll concentrate on the people and the analytical side of the market, not on your company’s an engineering and product portion (from my experience).

— You’ll most certainly share your results with the same people from above. If you have a boss, however, you will often report to them, and they will relay your findings and discuss them with the relevant stakeholders. You can also have a phase where you are gathering specifications, producing a report, and presenting it to stakeholders. For reporting, you can use tools such as Tableau, Google Data Studio, Power BI, and Salesforce. Often these tools can be linked to easy access to data sources such as a CSV file, although others need more technical work by advanced querying a SQL database.

— You’ll be working on projects much faster than a data scientist. On a weekly basis, you can have several data pulls (queries) or reports per day, and bigger visualizations and perspectives. Since you don’t (usually) construct a model and anticipate, you’ll turn around results quickly as they’re more descriptive and ad-hoc.

Data scientists are vastly different from data analysts. While some methods and languages can overlap between the two jobs, you may expect to work with various individuals and spend more time exploring bigger projects such as developing and implementing machine learning models. Data analysts can prefer to work alone on their projects; for example, one person may be involved in working with a tableau dashboard to present results, but a data scientist should include many other engineers and product managers to ensure that the model solves the business issue and that the code is accurate, strong, and efficient.

— Unlike a data analyst, for some projects, you can collaborate with stakeholders but turn to data developers, software engineers, and product managers on other aspects of the model and its performance.

— You should expect to share your results with stakeholders, but also with some developers who need to know what the end result is to create a UI ( user interface) around your predictions, for example.

— Maybe the greatest difference in how these tasks feel and work is the amount of time you assign for each project. Although data analytics is more efficient, it can take weeks or months for data scientists to complete a project. Since there are processes such as data collection, exploratory data analysis, base model development, iterations, model tuning and results in production, data science models, and projects which take longer.

You should expect to share popular resources like Tableau, SQL, and even Python as a data analyst and data scientist, but the experience from each role can be vastly different. For a data analyst, each day’s job requires more meetings, more face-to-face interactions, soft skills, and faster project turnaround.

Data scientist work may include longer processes, interactions with engineers and product managers, and, ultimately, a predictive model that looks at classifying new findings or events in time, while data analytics focuses on the past and current state.