20120424

Tableau vs. Qlikview

Some people pushing me to answer on recent Donald Farmer's comments on my previous post, but I need more time to think about it.


Meanwhile today Ted Cuzzillo published an interesting comparison of Qlikview vs. Tableau here:


http://datadoodle.com/2012/04/24/tableau-qlikview/


named "The future of BI in two words" which made me feel warm and fuzzy about both products and unclear about what Ted's judgement is?


Fortunately I had a more "digitized" comparison of these 2 Data Visualization Leaders, which I did a while ago for a different reason. So I modified it a little to bring it up-to-date and you can see it for yourself below. Funny thing is that even I used 30+ criterias to measure and compare those two brilliant products, final score is almost identical for both of them, so it is still warm and fuzzy.


Basically conclusion is simple: each product is better for certain customers and for certain projects, there is no universal answer (yet?):


[googleapps domain="docs" dir="spreadsheet/pub" query="hl=en_US&hl=en_US&key=0AuP4OpeAlZ3PdHJ1eXZyWFJjUVoxbUl0cTlCVDlXU2c&output=html&widget=true" width="510" height="1060" /]

20 comments:

  1. Dear Dr. Pandre,

    Thanks for your post on the DV visualization tools comparison (Spotfire vs Tableau vs QlikView) on

    http://www.practicaldb.com/blog/data-visualization/tableau-qlikview-spotfire/

    It's indeed a very valuable information before investing on DV tools.

    Our Organization is looking at DV tools to reduce manual efforts for reporting, improve operational efficiency, perform smarter analytics. Our current BI system is built on Microsoft BI suite with database on SQL Server and OLAP cubes on SQl Server Analysis Services and hence we are looking for DV to connect to SSAS OLAP cubes.

    Your assessment on these tools indicate that for Multi-dimensional, Tableau is very good and QlikView is limited and Spotfire is None.

    a. And what do you mean by Limited for QlikView?

    b. On Which version of Spotfire is the assessment done? (I can see that Tableau 7 and QlikView 11)

    c. Your assessment on Spotfire indicate none for Multi-dimensional. Going through some of the blogs in TIBCO community it is indicated that TIBCO can be connected to SSAS OLAP cubes through Application data services ( http://spotfire.tibco.com/community/forums/t/1203.aspx?PageIndex=1 ) or through Linked Server
    ( http://spotfire.tibco.com/community/forums/t/1203.aspx?PageIndex=2 ). What's your opinion on that?

    Kindly appreciate your valuable feedback.

    Thanks
    Shankar

    ReplyDelete
  2. Andrei Pandre30/4/12 03:14

    a. Qlikview will not inherit any cube infrastructure (dimensions, measures, KPI, etc. and that is LIMITED) but I saw very smart Qlikview developers were able to get data from SSAS cubes with sophisticated Qlikview Scripting. In addition internal Qlikview data structures (Columnar database and associative Queries) in Qlikview can be used to emulate the Multidimensional Cubes

    b. Yes I used the latest to compare: Tableau 7, Qlikview 11 and Tableau 4.0 (version of Tableau Server still 3.3.1)

    c. Links you provided are not completed and they only can help with multistage extraction of Data from Cubes; when Data will arrive into Spotfire, all Cube infrastructure will be gone/lost.

    I have a good news (rather rumor) for you but you will have to wait for a while: I guess/hope that the next version of Spotfire (4.x or 5.0 or 6.0?) will have the ability to interact with SSAS cubes. In any case, if you need to choose the DV tool now and connection to SSAS Cubes is the Requirement, you have only one choice - Tableau 7.0.

    ReplyDelete
  3. Hi Andrei,
    1 Thanks for sharing so many of your observations on the DV platforms. I have found you observations very useful.
    2 It is definitely a mystery why Microsoft with all its resources in terms of money and people is not investing in making a DV tool whether within Excel or out of it. It seems to be having no integrated strategy towards BI. Things seem to be done in bits and pieces. Whether it has an overall strategy which commoners like us don't understand is not known.
    3 We do most of our practice in MS-BI, and lack of a DV has been a very sore point. We have lost a lot of our customers / prospects to QV (Qlikview). We are now trying to partner with one of the big DV companies viz QV or Tableau or Panorama.
    4 Panorama has still not come into the Gartners magic quadrant, while Tableau and QV have come in.
    5 Would it make sense to rely only on QV (Qlikview), or to do the back-end work on MS-BI and then have Tableau/Panorama do the front end ?

    Your opinion would be helpful to me..

    Sanjay

    ReplyDelete
  4. Andrei Pandre2/5/12 01:12

    Hi Sanjay:

    I suggest to stay away from Panorama until they will figure out of how to compete with Tableau and Qlikview.

    My answer on your questions are not simple, unfortunately. I also will not rely solely on Qlikview, because of

    a) they are limited by RAM, so Qlikview is not very scalable.
    b) Tableau is very competitive with Qlikview, in some case even better then Qlikview
    c) also Tableau allows server-less distribution of DVs with free Tableau Reader
    d) Tableau can read SQL Server Analysis Services Cubes AND PowerPivot Cubes

    e) I will try to keep MS-BI as the back-end, especially if you can use SQL Server 2012 with its Columnstore Indexes, xVelocity in-memory engine, and new Tabular Mode in SSAS. Try to stay away from Power View as it is immature and has many problems (SharePoint is the biggest).

    f) if you wish the real Visual Analytics, you may need Spotfire with its S-Plus, API and integration with .NET, IronPython and Visual Studio.

    g) In some cases you need either Spotfire Silver or Tableau Digital

    Andrei

    ReplyDelete
  5. Charles Gu11/7/12 16:13

    Hi Andrei,

    A great comparison of Tableau and QlikView. May I get a copy of the Excel comparion? Thank you very much!

    ReplyDelete
  6. I have done similar comparisons for clients looking for an evaluation between different BI Products. From a Qlikview point of view I can offer the following:

    - Although Gartner has now recognised QV as a leader, it is important to realise that the difference between QV 9 (when it first entered Gartner Magic Quadrant ratings) and the current v11 is vast. QV is not most certainly mature and stable product. The QV user community is well established with many examples and helpful posts.

    - QV connection to SQL SSIS cubes.... In most cases a cube is a data staging database whilst as a QV Developer I would try to jump past this and hit the original data sources directly. I have seen cases where QV has bypassed the cubes and after a few months it is evident that QV is able to much more easily adapt to reporting changes and changes to business data sources that the Cube (which is sheepishly being maintained by administrators who are finding their staging database becoming obsolete as a data source in favor of QV.

    - QV scripting is a combination of SQL and VB. these are well known and with a little experience I cant think of a data source or a business problem that would be too difficult for QV.

    ReplyDelete
  7. A little late here but I'd like to submit my company for consideration.

    http://anametrix.com/product-showcase/

    http://anametrix.com/technology-overview/

    Thanks for the great comparison above!

    ReplyDelete
  8. As far as Qlikview not being very scalable - could you define this? We have a database of 400 million records in qlikview and it runs very nicely.

    ReplyDelete
  9. Qualifier - We do have 512G of RAM in the server however.

    ReplyDelete
  10. Thank you for the comparison chart. As a QlikView user, there are ways to connect to SSAS and MS PowerPivot. 1 would mean that it cannot be done...this is inaccurate. It can and has been done. Tableau and QlikView get compared to one another although I haven't really seen an article exposing their true core competencies...in short, Tableau is a terrific tool for helping individuals navigate data sources that are pre-baked with formatting. It is a terrific visualization tool. QlikView is a functionally deeper product in that it can provide ETL-like behavior, store data offline (if required), and provide visualization. So, if both provide visualizations, which one is better at identifying patterns, outliers, anomalies?...the difference...QlikView's associative analytics engine...Green/White/Gray color identification. Also, not covered in your article was the ability to perform massive loads of data, incremental loads of data, lack of chaining from one workbook, unlike doc chaining in QlikView. I appreciate your table. It is certainly a good start, but you may wish to also include things like: # of functions available (QV), Meta data management (QV), server abilities (QV), # of wizards (Tab), Reusability of objects (QV), Auto date formatting (Tab), collaboration (sharing screens and annotations (QV), Internet integration with numerous social media sites (Tab)...

    One thing I cannot agree with stating is ability to connect to certain data sources...E.g. QV has an API, Tableau has an API...these API's can be used to connect to pretty much anything. When people bring up Cube connections...I know immediately they do not "get" QlikView's core purpose. Provide QV the raw data and let it empower a dept, a team, a division, the enterprise to connect, model, build a UI and collaborate for findings.

    The article is skewed for Tableau...I understand. I am a QV user. The tools are good solutions for organizations...just do your homework and be sure to see the tool in action...don't simply rely on demonstrations or a sales pitch. If you were to purchase a car...you would first, most likely, drive the vehicle before dropping down cash...do your homework.

    Thank you again for the article! Nice job!

    ReplyDelete
  11. a) QlkView can scale. It doesn't need to go after everything and place it all into memory. There are design best practices (as with any tool) that should be discussed and fitted for particular situations.

    b) agree - but not when it comes to sheer data volumes, loading speed, reusability, # of functions and calculations available.

    c) not sure why a business would want to permit its users to distribute content like this

    d) QV can read SSAS and PowerPivot. Partner solutions exist, MDX can be used, or frankly - can't you just go after the original data sources? Break free of the handcuffs of OLAP...there are other ways of modeling and interacting with data

    e) QV integrates very well with MS Sharepoint (Not deep on knowledge about Tab with this); Agree - keep data in db, unless you need control of it or access to it because of excessive controls or restrictions

    f) "real Visual Analytics" - not sure what that means, but I would be hard-pressed to find technologies better than QlikView, Tableau and Spotfire. Seems like new tools emerge each day...When will Oracle, IBM, SAP, possibly Google gobble up one of these tools? Only time will tell.

    ReplyDelete
  12. Andrei, can you please tell me what you think about KNIME (.org) solution and how it compairs to Tableau or other in the space...? Thank you.

    ReplyDelete
  13. Andrei, can you please tell me what you think about KNIME (.org) solution and how it compares to Tableau or other in the space…? Thank you.

    ReplyDelete
  14. Andrei, thanks for creating such a comparison table it's very helpful. I see Tableau and Qlikview as two different solutions, with a different approach. Tableau in my opinion is an out of the box data visualization tool that connects with sources that are already have been prepared for analysis. You don't need any training or technical expertise as a business user. Qlikview needs training, coding, is more difficult to maintain because it is used as a ETL and datawarehousing tool. I think Qlikview wants to enter the enterprise market and will focus on developing a more mature integrated datawarehouse and BI tool suite. My guess is that they will try to become one of the leading BI vendors like IBM, SAP and Oracle in the next fve years and that they will aquire what they need to broaden their suite(predictive analysis and production reporting for example).

    ReplyDelete
  15. [...] not Spotfire, for example. I just want to choose one. Qlikview vs Tableau. Comparison articles like this and this are very helpful, but a man is a man with his circumstances. Each starting point is [...]

    ReplyDelete
  16. I like your business intelligence software criteria (and the logical breakdown between business and technical) very much, will use for internal purpose, thanks Andrei

    ReplyDelete
  17. To get an idea of where Tableau is going and what Version 8 looks like, have a look at their keynote address of their conference in November http://www.ustream.tv/recorded/27070401 . This will give you and idea of the way forward and the thinking that goes into the next generation of visualisation tools. There are also some excellent demonstrations of new functionality.

    ReplyDelete
  18. Qlik view can read SQL, and I think the difference of visualization is bigger, Qlikview is superior to tableau in making graphics, reports,and all relationed with the interface.

    ReplyDelete
  19. Great observations and information is very well laid out. I would like to learn what it means when Qlikview scalability is "RAM Limited".

    ReplyDelete
  20. Jonah Randal19/3/13 05:05

    I would definitly add SiSense Prism to the equation:

    ElastiCube (SiSense technology) is disk-stored, which means the limit on its size is only tied to the size of available disk-storage, not RAM. Similar to other in-memory technologies, ElastiCube utilizes in-memory query processing for speed of calculation. But in contrast to these technologies, ElastiCube's query processing engine (Elastic IQ) only loads and unloads data to and from RAM on-demand. This means RAM is only taken up by data you need access to right now, but the rest of the ElastiCube data is stored on disk, available instantly when requested.

    For high-performance storage, ElastiCube utilizes a column-store, where data is stored on the disk as separate columns, rather than consecutive rows as in relational databases like MySQL. This storage tactic eliminates the notorious table-scan bottleneck, and also opens up opportunities for extremely effective compression, which ElastiCube employs extensively.

    ReplyDelete