Purchasing decisions on self-service BI software often come down to a three-way competition: Tableau vs. Power BI vs. Qlik. "Typically, they're the ones winning out in self-service BI. They're the leaders," said Rick Sherman, founder and managing partner at consulting firm Athena IT Solutions.
Sherman deploys the three self-service BI tools for corporate clients, and he teaches students how to use them in BI classes at Northeastern University's Graduate School of Engineering. Tableau, Microsoft's Power BI and Qlik Sense -- which became Qlik's primary BI platform in 2014, although the company also still offers its older QlikView software -- are about equal in the overall functionality that they provide, Sherman said. But he noted that each technology has its pluses and minuses, which he discusses in more detail in this Q&A.
Before we talk about Tableau vs. Power BI vs. Qlik, what are some of the key features and capabilities that users should look for in self-service BI and analytics tools in general?
Rick Sherman: Certainly the first thing users want is to be able to do analytics without, or with little, IT support. The second thing is to be able to access a variety of data sources and blend the data from different sources -- for example, if you pull in data from a database or application and want to blend it in flight with some data from Excel. There's a range of what self-service BI vendors can do on that.
The third thing is creating data visualizations and dashboards; some tools do dashboards and others don't, and some only let you do certain types of visualizations. There's also being able to drill down into data and apply filters, slicers and hierarchies. Different users, even within the same group, have always done that differently. One of the reasons why a lot of people use Excel for BI is that it was the only way they could filter and slice data in the past. Now, most of the self-service BI tools can do that, too.
Are Tableau, Power BI and Qlik all in the same ballpark on functionality?
Sherman: From a high-level point of view, they're very similar. When it comes to getting to different data sources, they might do that in different ways -- Microsoft understands metadata better than the others do -- but all three can get to pretty much any data source. When it comes to data visualizations, Tableau is definitely still a cut above, but Qlik and Power BI are close. They all do filters and hierarchies.
The reality is that all three are big enough that they can see what works best and copy each other. Part of choosing between them has to do with what you already have and your inclination: Do you like Coke or Pepsi?
Now let's look at Tableau vs. Power BI vs. Qlik Sense individually. What are Tableau's biggest strengths and the primary weaknesses or challenges that prospective users should be aware of?
Sherman: Certainly it has the best name recognition and still the best data visualizations. Also, excellent ease of use -- it's easy for business users to get started with Tableau on their own.
Cost has been an issue; it's the priciest of the three on a per-user, per-month basis. Another weak point for Tableau has been making the shift from desktop users saying, "We can do whatever we want" to people sharing data and publishing dashboards and reports via a server. All three vendors initially were getting their market share with the idea that users didn't need IT or a semantic data warehouse or data governance. It took a little time for all three to get their act together on server technologies, but with Tableau having a wider user base, it hit the wall on the need for that sooner.
What about Power BI's strengths and weaknesses?
Sherman: Power BI is very cost-effective -- or free for users who don't need Power BI Pro. Microsoft is also pushing out updates every month, and it has a tendency to be first up with support for different data sources; it isn't trying to be in a Microsoft-only world. And you don't have to be familiar with Excel to use Power BI. It's a plus if you are, but Power BI doesn't assume that you're an Excel jockey.
However, Power BI is still used primarily by people who know Excel and SQL Server. The first wave of users is in companies that are inclined to be Microsoft shops. Power BI also doesn't have either the number or quality of visualizations that Tableau and Qlik do. Microsoft is increasing them, and I'd say it has all that most business users need. But it still lags behind on advanced data visualization.
And what about Qlik Sense?
Sherman: Qlik has really good visualizations and some powerful data sourcing. I consider it fairly in between Tableau and Power BI on those things. And Qlik's Associative Engine is a big plus; it makes it easier for users to look at what tables and columns are applicable to each other in data sets. That's a capability the other two vendors haven't copied yet.
But when Qlik first introduced Qlik Sense, it kind of froze the market on itself because people didn't understand the positioning of Qlik Sense and QlikView. Qlik Sense also had to mature and grow a little bit, and that slowed Qlik's momentum.
Another issue has been data blending and more extensive data preparation work that does require changes to data. Historically, to make a lot of changes, you had to understand Qlik's scripting language, but Qlik has been addressing that. You still have some work to do, but it's better than it was.
Do companies evaluate Tableau vs. Power BI vs. Qlik Sense and try to standardize on one? Or is that not feasible when business units often can buy self-service BI tools themselves?
Sherman: I think there's a desire to go to one self-service BI vendor. But in many enterprises, you'll find Tableau or Qlik and Power BI. If it's Tableau vs. Qlik, one of them usually wins out. But Power BI sort of stands on its own. It goes with Excel, and you're never going to get rid of Excel.