Difference between Data Analytics vs Data Engineering vs Data Science

This is something still in the grey area which I believe it all overlaps together in different places (digital marketing + data analytics + Data Engineering + Data science). More than a few fields, I believe data science has been heavily applied in personalisation and improving the user experience of a visitor on a website or mobile application. Meaning even if you want to start data science as a career, getting into digital marketing and analytics will be a stepping stone to understand where you can apply this. Without any argument, data science is not just related only to digital marketing but application in digital marketing is easy to understand and apply. But how all this fit together?

Technologies, products, and services involved in Data

Disclaimer Source: Reference Architecture of Big Data Systems – ScienceDirect.com – Pekka Pääkkönen Daniel Pakkala

I know it’s going to be too much to look at this in this image. The purpose is here not to understand everything but getting an exposure of it. Let’s focus only on the number I have marked there.

1. After the data received from data sources or from processing (4) or from data analysis(3), data science can be applied there. For example: an analysis you did once found some substantial result but not a deep learning, you use machine to learn the deep results. This is the job of the Machine Learning expert to identify what models to use and how it needs to be done. Now a days lot of ready made tools has come in where you just need to give source data and use common sense to tell which model would be suitable for the data, they will provide deep learning
2. After every data is ready or want to be shown, you need to present in a way. You could present in MS Excel/GSheet in simple charts, or you build dashboards in Tableau or Periscope or inside GA or Adobe, etc. This part is played mostly by analyst but in different organisation, most people are equipped with knowledge on how to use this tool. So the data are transferred in to this tools, then everyone build their charts/dashboards/story in that.
3. Whether you are running machine learning(ML) or not, any data sources could be analysed yeah? Before ML, data analysts are the one who has done this job of analysing the data. And they are going to be the one who will do the first human touch of analyse. ML won’t be or replace 100% of manual analyze. And also for small data set you cannot expect ML do the job. That’s the part, the data analyst will be filling.
4. In organisation where you have multiple data sets and if they are complex manner. You will need to do a job of sourcing everything into single place. While you are bringing everything into one(as simple to say), it involves cleaning the data, merging and few more things. Some call it as ETL or ELT, it’s basically a data engineering job of bringing together in the expected model to facilitate for ML or Visualisation or Analysis.