With Microsoft Fabric, it will no longer be necessary to acquire different components individually while everything will be centralized in one place, which is a big advantage. With this advantage comes another one, the fact that the various data professionals (eg data engineer, data architect, data scientist, BI analyst, etc.) will now be able to collaborate in a single tool. An integrated tool for all professionals, at ventriloc, we love it 😊.
Again, that’s all well and good, but what is game changer?
Here I am giving you my opinion based on my past experiences and current knowledge of Microsoft Fabric. I already see some big advantages to using Microsoft Fabric. Here are the most significant ones:
1. The introduction of the OneLake concept.
The OneLake concept try to solve the biggest challenge in BI, implementing a single version of the truth. Each organization that will use Microsoft Fabric can only own a single OneLake. This represents a further step towards centralizing all the data of an organization. Also, OneLake will not be exclusively reserved for technical people since the data is structured in the same way as a OneDrive. This will help democratize data access for non-technical business users.
Not sure how to write your complex measure in DAX? Ask CoPilot, it will do it for you! However, make sure to confirm that the result is what you are looking for since CoPilot is not always right.
Want to create a new Power BI report but have blank page syndrome? Ask CoPilot, it will do it for you!
Do you find that the report you created is not visually appealing and you would like it to look like a report that one of your colleagues developed? Ask CoPilot to adjust the design of your report to match your colleague’s report!
I don’t think I need to convince you any further, you have no doubt understood that CoPilot has the potential to become your best friend in your data journey.
On the other hand, in order for CoPilot to guide you effectively, you must have an irreproachable semantic layer. By irreproachable, I mean that you must give explicit names to your tables, columns and measures so that CoPilot can couple your questions with your semantic layer. For example, if you ask CoPilot to generate a sales report on a dataset but the measure that represents your sales is called “Value”, CoPilot won’t know where to retrieve the sales.
3. Data Activator
Another new toy from Microsoft! With Data Activator, you can have peace of mind and avoid having to scan through all your organization’s reports looking for data anomalies because Data Activator will do it for you. Indeed, Data Activator allows you to set up a monitoring and alert system to notify you when an anomaly is encountered in your data. Isn’t that every manager’s dream?
4. The new Direct Lake connection mode in Power BI.
For those who have worked with Power BI, when you want to import data into Power BI, you have to choose between 2 main connection modes, Import or Direct Query. As shown in Figure 2 below, the Direct Query mode provides real-time data but the performance is not there, while the import mode provides high performance but requires duplicating data from your systems in Power BI in addition to not allowing real-time.