Software Review: Power BI

Forays into Data Visualization as an integral part of Data Analytics, have brought me in contact with several Data Visualization tools. One of these is Microsoft's offering, PowerBI. The "BI" part of the name, of course, stands for Business Intelligence, which is one of the use cases for Data Analytics.

If you have a need for ETL that will ultimately factor into your Business Intelligence, this is what PowerBI is for. Of course, PowerBI is not just about Business Intelligence. Any use case you may have involving transformation and visualization of data, will be handled by PowerBI.

The Premise

PowerBI is a desktop (or cloud-based, depending on your package) application that takes a data source or several, and combines them into a slick narrative, cumulating in charts, dashboards and storyboards.

The Aesthetics

Power BI follows a similar aesthetic to most Microsoft products. This means that if you're familiar with Access, SQL Server or Excel, you will feel pretty much at home right off the bat, whether it's managing dataset relationships or cleaning data.

Outside of data visualization colors, white and grey are predominant with button mostly in yellow. There's the occasional flash of green.

The Experience

Honestly, I found Power BI intimidating. Just for the simple task of importing datasets, there was a whole host of options to choose from. Further into the actual cleaning and visualization of data, each task had a blinding array of configuration options.

Thankfully, Power BI manages to stick to a well-established workflow - import data, clean and merge, create analytics, visualize.

The Interface

While the interface can get a little crowded, much of it is safely ignored with little consequence. For the most part, controls are straightforward and easy to find especially if you have had prior experience with Microsoft products.

Of special mention is the relationship relation interface. For such a busy looking display, the process of merging various datasets was relatively direct. And it was not clumsy.

What I liked

The meat of the software, as far as I'm concerned, is in the visualization. On that score, I wasn't disappointed. There are tons of chart types and dashboard options, with each chart highly customizable.

Simple things are easy enough to accomplish if you don't get distracted by all the other interface elements.

Statistical information about the data is available right off the bat when you view it, with more options available if you want it.

I've mentioned this before but it's great that the UI is mostly consistent with other Microsoft products.

What I didn't

Power BI Desktop is free. The costs of Pro or Premium seem a little high, though perhaps this can be justified with scaled-up use cases.

I found the going slow at times. This may be improved with the Pro or Premium versions, and may just be a result of me running Power BI on my ridiculously old Windows machine. Use on a Macbook was a no-go last I checked.

The screen can sometimes get a little crowded with the sheer amount of controls provided. Half of which I will probably never use.

Conclusion

PowerBI isn't what I would call a slick little package. It has a whole host of functions and sometimes all that can be overwhelming. Still, it's good enough for most cases even if the functions I really need are sometimes annoyingly elusive, hidden amongst a ton of related functionality.

I would absolutely recommend PowerBI for your Business Intelligence use cases. And if you can afford it, PowerBI Premium or PowerBI Pro.

My Rating

7 / 10

Powerful stuff!
T___T

So you want to be a Web Developer? (Part 3/4)

So we've covered the front-end and back-end aspects of web development. And, as mentioned in the last segment, we're going to delve into databases.

Unless you're only planning to create static websites, which in this day and age isn't really feasible as a career choice, at some point or other you're going to have to deal with databases, or at least a data source. Some databases are more suitable than others - it all really depends on the scale of what you're trying to accomplish. An enterprise-level application might require SQLServer, whereas MS Access is probably better for a simple disposable solution.

Again, as with back-end scripting languages, learn one database platform, preferably two.

- MySQL (http://www.mysql.com/)
- SQLServer (http://www.microsoft.com/en-us/server-cloud/products/sql-server/)
- MS Access (https://products.office.com/en-SG/access)
- DB2 (http://www-01.ibm.com/software/data/db2/)
- Oracle (http://www.oracle.com/index.html)

 

What you must know

- SQL (well, duh)
- Views
- Indexing
- Stored Procedures
- Database security
- Normalization

Good-to-haves

The following aren't absolutely necessary, but if databases are going to be your specialty, you could do worse!

Do have a look at CSV, JSON and XML as alternative modes of data storage and transfer.

MS Excel is a spreadsheet application, but through the years it's picked up a heft amount of added functionality. Worth a gander. (https://products.office.com/en-us/excel)

LinQ is an alternative to SQL. (https://msdn.microsoft.com/en-sg/library/bb397926.aspx)

NoSQL databases such as MongoDB and Redis. They seem to be the in-thing now. For a more comprehensive listing of NoSQL databases, see this link (http://en.wikipedia.org/wiki/NoSQL).

What are my career options?

If you're going to do only databases, why go into web development at all? There's demand for database administrators all around.

Next

Coming up, the final ingredient needed to be a web developer. Stay tuned!