Data Platform and Data Science

For more than 20 years we have BI (Business Intelligence). In the early days it was Business Objects, Cognos, Panorama, Hyperion, MicroStrategy, SAS, SSAS and ProClarity. Nowadays it is Tableau, QlikSense, Power BI, Looker and SiSense. And many others. Until today, BI tools are still dominating the business landscape.

People who develop Tableau dashboards, Qlik dashboards, Power BI dashboards, etc. are called BI Developer. They develop dashboard using BI tools.

A few weeks ago I wrote that it’s not called BI any more, it’s called Analytics (link). BI only covers the past, whereas Analytics covers the past and the future.

There are 4 types of analytics: descriptive, diagnostic, predictive and prescriptive. Also known as 2D+2P. The 2 Ds are BI, whereas the 2 Ps are AI. So, to move from a BI Developer to an Analytics Engineer, one needs to be capable of doing AI. Meaning machine learning. Selecting the right algorithms, building the models, training the models and deploying them.

If you are a Tableau developer (or Power BI, Looker or Qlik for that matter), are you willing to train yourself in AI? I find that this gap is too big to get across. Only a small percentage of BI Developers are willing to retrain themselves in AI and machine learning. After all, not everybody is able do AI. AI requires certain background knowledge. Strong programming is required (Python), and some foundation in mathematics are required too (statistics and calculus).

What is an Analytics Engineer?

Claire Carroll defines Analytics Engineer as follows (link): Analytics engineers provide clean data sets to end users, modelling data in a way that empowers end users to answer their own questions.

I am not sure if this definition covers AI (ML models). It doesn’t seem so. There is no mention of AI or ML here. “Modelling data” usually does not mean ML – it means data modelling in its usual sense.

So there are 2 different definitions here. I define Analytics Engineer as BI + AI developer, Claire defines it as providers of clean data sets. Well if we use Claire definition (please read her article on Ref #1 below), then the shift is towards data engineering/analyst. If you are doing Tableau dashboards today, you’ll do more of sourcing and cleaning data, and make them readily accessible for the business users.

If that is the definition (providers of clean data) then there is no need to train yourself in AI. Here’s Claire’s explanation about the difference between Analytics Engineer and Data Engineer and Data Analyst roles (link):

Well may be for now that’s the definition of analytics engineer, i.e. nothing about AI!

Please help yourself with the resources below if you want to dig deeper into Analytics Engineering.

References:

1. What is Analytics Engineering? by Claire Carroll: link.
2. Analytics Engineering Guide, by dbt: link.
3. Data Engineer vs Analytics Engineer: How to choose the career that’s right for you, by Madison Schott: link.
4. What is an Analytics Engineering? Everything you need to know, by Adel Nehme: link.
5. dbt and the Analytics Engineering – What’s the hype about, by Oliver Molander: link.
6. Introduction to Analytics Engineering, by Ust Oldfield: link.