Comparing Pyramid Analytics and Power BI

With so many different BI tooling on the market today it can be daunting to choose the right one for your company. Many things influence your final decision. Your IT strategy, for instance, may enforce a cloud-only policy or may even state that all (BI) products should be from a certain platform (e.g. Oracle, Microsoft etc.). Let’s not forget your most important customer, the user. Their demands and wishes also effect which product is best for your company.

The goal of this blog post is helping you make that decision. This post is inspired by Jen’s article in which she compares Tableau, QlikView and Pivotstream with each other. I encourage you to read that article too.

First off, some background information on Pyramid Analytics and Power BI helps you better interpret the comparison made in this article. Information on Power BI can be found here and here. Information about Pyramid Analytics can be found here.

The comparison made in this article reflect my personal opinion. You may share a different opinion. In any case please share your thoughts (respectfully) with us through the comments section. I’ll start with a detailed comparison followed up by my conclusion.

Detailed comparison

Business criteria

Visualization criteria

Technical criteria

Conclusion

Both Pyramid Analytics and Power BI aim at simplifying information discovery through self-service BI. I think both tools succeed in providing a clean and user friendly way of accomplishing this. Of course, user friendly being dependent on what your typical user is comfortable working with. Pyramid Analytics requires a predefined OLAP cube, thus requiring users to understand the concept of dimensional models. Power BI requires Excel and uses Tabular which basically is the concept of relational modeling. Although Power BI can also work with dimensional models it is not mandatory.

Pyramid Analytics relies on an OLAP cube. SSAS OLAP cubes have been around for a while making the product mature. Power BI relies heavily on Excel 2013 making it ideal for Excel (power) users. Both Pyramid Analytics and Power BI work with a subscription based licensing model. Power BI is part of Microsoft’s Office 365 offering.

Below a graphical representation of where I think Power BI and Pyramid Analytics fit. I used the maturity model provided by TDWI. Note that this does not say anything about which tool is “better”. It is where I think both products ideally fit.

I hope that I have helped you better understand your options. One piece of advice, play with different tooling (most vendors offer trial versions), make sure you are informed by attending demo’s and / or asking peers / co-workers what their experiences are. Of course you are welcome to ask me for further advice on this matter. Good luck!

**** UPDATE ****

I have spoken with various people regarding this post. These conversations led to better insights which I would like to share with you.

Not so BIG data and the enterprise way of things

I think Power BI is excellent in providing ways to connect to “BIG data” sources like Facebook, Hadoop and many more. However, Power BI is not suitable for actually “housing” BIG data. Power BI relies on Tabular in Excel which cannot house trillion rows of data. Moreover, Power BI itself has a restriction on workbook size (currently 250 MB).

Pyramid Analytics works on top of Microsoft SQL Server suite (both TABULAR and OLAP). OLAP knows no technical restriction like maximum size. Moreover, OLAP has excellent scalability features. These features make Pyramid Analytics a good candidate for enterprise solutions.

The “grey area”; a feeding ground for endless discussion

In making my comparison I followed a simple guideline; “What do I get when buying this product”. This led to some big differences mainly caused by the fact that Pyramid Analytics works on top of Microsoft SQL Server suite (Tabular and / or OLAP).

Example: I think Power BI provides more ways of connecting to external data sources. You could be of the opinion that Pyramid Analytics provides the same because it can work on top of the same model. The difference I make is that with buying Power BI that feature is included. It isn’t included with Pyramid Analytics, you need to buy Excel through Office 365.

Simply one of those things I feel you should be aware of.

Final points; “in – depth functionality” versus “for the masses”

I feel I have left out an important part in this post, and that is a proper conclusion. So, without further ado, here it is.

Pyramid Analytics works on top of Microsoft SQL Server. Together they provide excellent extendibility and scalability features. Pyramid Analytics is more an on premise solution then it is cloud. There are ways of “cloud enabling” Pyramid Analytics but it will require third party platforms / applications like Microsoft Azure.

Power BI is a cloud only solution for which you will need an office 365 subscription. Power BI is an all-in solution meaning it does not require other tooling / platforms to unlock the full potential. Power BI is more restricted when it comes to extendibility and scalability.

Power BI is something you can easily start with but will start to feel restricted when more enterprise features are required. Pyramid Analytics offers more on the “enterprise” side. Being an enterprise solution Pyramid Analytics certainly feels that way, there a ton of features to use which at times can feel daunting. The required installation and configuration makes Pyramid Analytics feel less agile.