Back in 2019, Microsoft introduced a new model for their certifications, called “role-based certifications”. The old certifications like MCSE (Microsoft Certified Solution Expert) and MCSA (Microsoft Certified Solution Associate) are gone. Also the Microsoft Professional Programs (they were not so well known) are gone. Have a look at this blog entry if you want to read about data certifications and the professional program in the past: https://www.mmertens.eu/2018/08/microsoft-data-certifications/
The new certifications are very much “cloud first”, thus Azure based. Role based means, that Microsoft defines role in a way they see the world (like Data Engineer, Data Analyst). It is not necessarily how I see the world, but this is not a problem. I’m a mixture of a Data Scientist, Data Engineer and Data Analyst in Microsoft terms, but I don’t care …
The case that everything is about Azure, is much more important. That means (again: for data specialists), that there is no certification for the former “flagship” software “Microsoft SQL Server” (on premises) any more.
Now here is how I see the transition from the old MCSE / Microsoft Certified Solution Expert for Data Management and Analytics to the new role based ones. Be aware that the MCSE had a concept of how several exams are linked together (start with an MCSA, then add another exam), while in the new world the certifications are independent from each other. For each certification, one exam is needed. So you can pick the ones that you think are useful for you.
There is the Azure Data Fundamentals certification DP-900 that Microsoft recommends for the data. It has certainly valuable content, however if you go for specialized DP vertifications, I would say that the DP-900 is not needed. Maybe you can have look at the parts that you don’t want to specialize on. But I find the certification unnecessary. I wouldnt even add it to my cv if I did it.
I personally did the AZ-900, the Azure Fundamentals certification. It provides the basis of how Azure works and is useful for everything you do in Azure.
Analyst means “Power BI”
“Analyst” means working with Power BI in Microsofts view. There are two certifications: the PL-300 “Power BI Data Analyst” (formerly DP-100) and the DP-500 “Azure Enterprise Data Analyst”. The DP-500 is brand new in 2022, and it is – like the name says – the Enterprise version. It combines Power BI with underlying technology, especially Azure Synapse. I see (in the meantime) the PL-300 as “Power BI fundamentals”. The DP-500 fits much better into the DP certification series. It was a little different when DP-500 was not existing yet.
By the way, the “PL” comes from the Power Platform certification system.
Data Scientist means “Azure Machine Learning”
This certification is about the Azure Machine Learning service and Azure Databricks. I don’t like the fact that there are two technologies combined into one certification because it forces you to learn more than you want about the one of them that you don’t plan to use. It would be better if they Azure Machine Learning and Azure Databricks were separated.
Regarding preparations, I wrote a blog post about this certification here: https://www.mmertens.eu/2022/07/udacity-or-coursera-for-dp-100-azure-machine-learning/
Administrator / Engineer
Finally there is the engineer side. One certification is called “engineer”, the other one “administrator”. I find “administrator” a little bit misleading. It also is about engineering the data – more or less specialized on the Azure SQL Server flavours. The “engineer” is more universal, e.g. it is about Synapse, Data Lake Storage, Apache Spark, Stream Analytics and so on.
As I wrote in the beginning of this post, there are no paths like in the old certification world. For me the two pairs of certification make sense together. So my personal path is:
(AZ-900 ->) DP-100 -> DP-500
The more engineerish path would be:
(AZ-900 ->) DP-203 -> DP-300