Azure Analysis Services Vs. Power BI Premium

Is that even a valid question?

A friend of mine from Microsoft wrote a blog questioning Azure Analysis Vs. Power BI Premium a few years ago, but at that point, I think it was a little easier to answer and my own answer would have been different at that time!

It is probably not a valid question at first glance and that is the initial response I get when I have brought this up in discussions in the past. But with higher dataset sizes and more refresh options than Power BI Pro, would you ever need Azure Analysis Services in most cases?

In some scenario’s this is very easy to answer (this is not an exhaustive list) –

  • High level of configurability, especially around scaling up and out, or start/stop the service – Azure Analysis Services
  • Greater than 500 Power BI Pro licenses – Power BI Premium
  • A Microsoft house with lots of money – both… maybe ????

In order to figure this out and cement my thoughts around my recommendation, I did some research into Microsoft’s reference architectures and compared feature sets.

And to ensure we are on the same page; this article is focusing on the data modeling capabilities of Power BI Premium.

Microsoft’s reference architecture for the modern data warehouse – Click image to enlarge

In this architecture, Azure Analysis Services is included and of course Power BI for front end reporting and visuals at the least. The description that goes along with the architecture does not go into detail about what version of Power BI is included here.

Azure Analysis Services Vs. Power BI Premium Feature Comparison

A couple of years ago, looking at the features made this an easier choice as there were some clear differences between the services. There were hard limits on dataset sizes in Power BI Premium, and you could not use 3rd party reporting tools to connect to the data models. The later made the choice very simple if the organization uses reporting tools other than Power BI, and the former made it very easy if the dataset sizes were bigger than 10GB. This would mean Azure Analysis Services would be recommended.

With monthly release cycles for Power BI important and requested new features are released all the time. The last 12 – 18 months being especially big when comparing these two services.

Figure 1 compares some features based on use cases

Power BI vs Azure Analysis Services | Simpson Associates

Figure 1

 

Figure 2 is taken from the Power BI blog and compares features of both services. As you can see both services have very similar features now available, with Power BI Premium also bringing a lot more functionality for other tasks, and if there is a feature missing, it is marked as not yet rather than no.

Figure 2

As you can now see the feature list is a lot more comparable than it was just a couple of years ago. This has made it even more difficult to pick between the services for most cases.

I think the direction of Power BI is clear and most people understood that all the functionality of Azure Analysis Services would be made available in Power BI over time. Could there even be a rebrand on the horizon for Azure Analysis Services – who knows? (outside of Microsoft).

Having said that, in the Power BI blog mentioned above the following quote from Richard Tkachuk, Principle Program Manager on the feature set confirms our thoughts.

Power BI Premium is the focus for enterprise BI and the primary target for future investments. In time, Power BI Premium will provide a superset of the capabilities when compared to Azure Analysis Services. The clear future direction is Power BI Premium.

With a few exceptions, Power BI Premium provides a superset of the capabilities available in Azure Analysis Services. In time, these exceptions will be eliminated making Power BI Premium a clearly superior choice when considering capabilities alone.

What does this mean for Data & Analytics Consultancies?

If I can offer any advice on this from my own experience it would be to consider the following points:

  • Get a good understanding of your client’s requirements and get a very good understanding of your client.
  • The pricing models for both services are very different.
  • Managing the expectations of your client is key.
  • Good consideration of the overall analytics strategy and future road map.
  • Assess feature lists, assess costs, assess client working practices.

Get in touch with Simpson Associates for any help with this!

And in my honest opinion Power BI Premium wins this comparison for a lot of projects!

Blog Author

Liam Wilkinson, Lead Consultant, Simpson Associates

 

 Don’t forget to join us on Linkedin

Back to blog