QlikView or Tableau - Which is better?

Dear Sir,
Qlikview, Tableau, Spotfire,... are all front end tools which replace the Excel user interface. All have strengh and weaknesses but it's still visualisation tools and all are good. The major point is that, no it's not magic tools and you have a lot of work to do before (data preparation, cleaning, MDM,...) and it takes time (minimum 30 to 50 days for a "normal" project based on one DB source). If you work on multiple sources, you still need a DWH event if some says that the have their own DWH inside the software. Those tools are not yet the perfect "self service" BI tools, business still needs IT people. And please for all those tools, don't forget the technical and functional documentation if not, in 1 years time nobody will be able to manage the solution.

Then yes they are good tools, yes it's more or less the same as visual tools, NO it's not a magic world where in 2 click all it's done moreover if you work on multiple source.

If you need more help, contact us, no problem.

Regards

Hi,
We did this exercise of evaluation of tools qlikview vs Tableau.
One thing that I want to make sure that you should be thinking about is that these tools are going to complement your existing enterprise BI tools or if you have none then you will have to think about having one. Why do I say so, like probably most people above would have said (I am sorry I have not read every person's comments), these tools are primarily for data visualization and data discovery. It is not that everybody in the company will be doing that or needing that. There would be scenarios (and based on my experience a lot of them) where in the users are happy to consume a standard report with pre-defined navigation. Both these tools are not your tools for what we call canned reports and/or pixel perfect reports.
Talking some specifics about them - Tableau (as you must have read) works on the principle of "best practices in data visualization" so if you are looking for 3-d charts for example or people in the organizatoin loves that then Tableau is not your tool. If the data is too scattered then again Tableau will have challenges as it does not even have scripting (like qlikview) for basic ETL. Applying data security/scheduling etc. is not that straight forward. But if you are looking for end-users to be able to create their visualizations and do some data discovery (it is an often misused term in my opinion..:)) very quickly then Tableau is your tool.
With respect to QlikView, it is more of an application approach where users are given a prebuilt application (like HR, Sales, or even by role, CEO/COO etc.) and they can play with the data in those boundaries. This application is generally built by IT people. Also it is not easy to merge the two applications so like I mentioned above if we create an HR, Sales, operations etc as different applications and then tomorrow we need to build a dashboard for COO or CEO which include data from some or all of the above already created applications then we will have to create another application combining that data, thus creating redundancy and maintenance overhead.
In a nut shell, Tableau is for you if your primary usage and requirement is best practices visualization and data discovery. Qlikeview is your tool if you know the boundaries for people for each application and are looking for more jazzy (read 3d) visualization. Also mobile could be another criteria Tableau does not have any native mobile interface (although it does have something for tablets), qlikview has a very nice and native app for android and ios which works on mobile and tablet.

Hope this helps.

Have no experience with either but you may find the link provided by Jackie Lieber useful and fairly detailed and also the gartner industry research in the link www.informationweek.com Cheers.

Both are compliant for Trend/Visual Analytics. Tableau works with local data processing as well,
For deeper insights need you should look for a complex architecture.

I've not worked with it but to reduce your frustration, do a needs assessment. Make a list of the most important tasks the tool must do for the users and the rationale for using it. Maybe you do not need to use either tool unless it is a mandate from the company. If not, based on the need, you might be able to use an alternative tool that can do the job without a heavy price tag. Here is a link to compare the tools:
www.experfy.com

Sorry I have not experience with either

This is a handy side by side.
businessintelligenceassistance.wordpress.com

My take on QlikView vs. Tableau-

QlikView performance depends on your data model, front end design and
server configuration.

If your front end has lot of calculations and tabular reports there is chance of a slow down in the performance.
Data Model - Try to use less tables and if required consolidate /
integrate the data sets to decrease the number of tables
Front End Design - Try to keep all complicate calculations at backend,
front end should be used just to create report objects
Server configuration - follow the QlikTech guidelines to calculate
server configuration

Tableau - One of the major advantage in Tableau is its connectors to
various databases,
user need not to procure any new connectors to connect SAP / Big
Data databases.
Tableau performance depends on the database performance.
Tableau is cost effective compared to QlikView.

See my comment above, but in order to be truly helpful, we need to inquire:

1. What is the size of your business?

2. How big is your IT staff (or will you be managing this tool by yourself)?

3. Do you have a budget, and if so, generally what is the size of your budget/any financial limitations for potentially purchasing this BI solution?

4. What sector of business are you in (banking, government, education, non-profit, manufacturing, etc. – as there are a variety of BI tools that have specifically geared themselves to a given business sector … public sectors especially)?

5. Are you just getting started on your BI journey, and if not, where are you at on that journey?

6. Do you have an executive champion designated for your BI initiative?

7. Are you the BI Steward for your business, and if so, what is your experience with implementing BI solutions thus far during your career?

It’s important to know who we are trying to help with our professional input, and how sophisticated their business may be at this point in time in truly adopting BI as a functional leadership tool to assist in the business’s strategic decision making.

Have to disagree on the performance issue. I have a demo I show that is over 1 BILLION rows of data, on my notebook, in Tableau. When it comes to performance, I think any solution can be good (with good DBMS and dashboard design) and any solution can have issues (with bad DBMS /dashboard design)...

Hi,

I am not familiar with Tableau. I had very limited exposure to QlikView. My real area of experience is with Microsoft’s set of BI Tools especially Analysis Services. I do work, amongst others, for a big Mining Company with millions of records of data and so far even the Standard Edition of SQL Server has done the job. The companies that I worked for use SQL for their transactional data so adding the BI tools does not cost them anything whereas any other toolset would cost them extra.

SQL 2012 Enterprise and BI Versions also includes a Tabular Model for Analysis Services which is very similar to the QlikView approach. It also runs in memory (but can use virtual memory) and is capable of doing the same type of tabular calculations with DAX.

The biggest problem I came across with QlikView is that it appears to be very hardware (memory/processing) intensive.

Regards,
Dawie

Use Jaspersoft, it doesn't have any scalability issues.

Try office365 (free for 1 month). It's way cheaper and simpler than both.
Some functionality is missing e.g. combined line and bar graphs, but O365 has
just arrived and I don't think it will be long before these missing
functionalities are added.

Regards,
Mandar

Choose Spotfire if scalability is your concern. I've been using Tableau for a number of years but kept running into performance issues. Spotfire is enterprise ready so no such issues there.

The answer to which is better isn't that quite easy to give as you already
found out. It depends on a lot of factors.

- What is the size of the information you are going to work with?
- What do you want to do with Business Intelligence / Customer
Intelligence (Dashboarding, Reporting)?
- What is the size of your company?
- Who are going to work with it (licenses)?
- What are the IT skills level of the people?
- What is the budget you have?
- How are the training facilities and/or support?

As to performance, I have tested QlikView with 10 million sales records
over 10.000 customers over 10.000 stores in 100 countries with 1 million
articles connected to the AWS cloud. For getting all the data over internet
to your local workstation you will have to write a kind of looping,
preventing a timeout, but that will be the same when you use Tableau on
your local workstation.

So it doesn't have to be a problem, but when you are talking about
performance there are better ways to improve that. One of them is to think
about how you store your data. Should it be optimized for specific goals,
before you hand it over to any BI tool?

What is to say about Tableau? I like it more for analysts. It looks more
appropriate for people who are more into business then IT.

What is to say about QlikView? I like it more for IT focused people
supporting the business. Although QlikView has a couple advantages above
Tableau, like showing what you didn't select and a personal free edition on
which you can test all you want as long as you want, getting the right
result is more technical than Tableau.

Kind regards,
Patrick de Witt

There is a fourth level, where no tool helps anymore. It is custom sql development. To cover that, you should have a (IT-) team who does the development. If you want to pass on that responsibility to the business too, you may consider a sandbox solution in addition to a data discovery tool.

Excellent answer!
Thanks a lot!

Regards
Elisabete Miranda

From my point of view people new to Business Analytics and Business Intelligence are like newbies buying smartphones: shinning colors are attractive but not the professional base to decide about a professional use of an IT tool. 

Having more than a decade of background on DW/OLAP/BI I can say that there are three levels of reporting:

1. Regular static reports which are normally generated at night to be available thru intranet the morning after. This kind of reports let you change a type of chart displayed and perhaps some level of drill down. These reports are generated by batch processes that are ready on time because IT staff take care of that.
2. On Line on demand reports needed at any moment which must be based on Olap Cubes reporting tools like those provided by the big vendors like IBM Cognos, Hyperion Essbase or Oracle Business Objects or SAP DW. Solid, stable and with a professional foundation and solid structure.
3. On demand and customized reporting like typical BI new generation tools offer  (Microstrategy, Tableau, Qlikview, etc.) that generally are managed directly by end users and no IT people. This means that the source must be reduced and oriented to the problem to be queried like a proper Data Mart is. 

If the DW/BI infrastructure of your client is well designed each corporate area should have their own Data Marts oriented to their needs, to be queried at any time by powerful tools like Qlikview. If it is the case then a few moments more or less during a labor day to obtain complex on demand reports has no importance at all. Qlikview and Tableau are equivalent in many aspects. In my opinion I prefer Qlikview but based on logic power, usability and support not in a few milliseconds or minutes of difference to get a report.

Corporate central decisions making data must reside on proper datawarehouses and OLAP databases. I would never use Qlikview or Tableau which are limited data mining tools, to support the big decision data of a company.

Regards,

Marcelo

Sorry

Researched both, but never actually used either one (?).

Sort of depends our friend’s use case doesn’t it?

Different business sectors, business sizes, and the sophistication of the BI champion involved (hopefully they have one in this case) have a lot to do with the deployment and adoption of any BI solution and its ultimate success in a business.

Hmmm …

Might this online resource help our BI puzzled friend?

community.qlik.com

CAUTION: This table may be somewhat biased towards Tableau! … but what on the net isn’t?

Some balancing factors in support of QlikView (from that same site):

- QlikView is certainly Enterprise ready with multi-server configurations.

- QlikView in the cloud offered by partners (Rosslyn being a prime example).

- ETL and data integration is very strong in QlikView - though very script driven rather than UI.

- I have seen articles re: integration with R in QlikView.

- Storytelling quite natural in QlikView - with tabs and navigation buttons.

- To my mind implementation time is very rapid in QlikView.

The ability to make data publicly available without the need to license each client and at a sensible price point (without the use of NPrinting or similar to push HTML or images to a web server) is somewhere I would like to see QlikView catch up though.

No mention of affordability for businesses or end user experience factored in, but from those perspectives, we’re learning about Information Builders’ BI solution …

Hope that helps!

Personally I haven’t used either tool. We are in the process of choosing a BI solution for our business and pretty much have decided in going ahead with Yellowfin.

Tableau was also considered, however, it has some weaknesses in regards to Collaboration.

Reading through the answers on this thread, it would seem that the aspects of ease of use, visual representation and user benefits have ben addressed. Your question, however, addresses issues of scalability and therefore you should read the whitepapers on this topic:

www.qlik.com

www.tableausoftware.com

Good luck

I am not sure about Tableau, b

I agree with jhornber. Additionaly, as of today, QV is a in memory tool. You first load your summary tables into the memory than start analyzing. Tableau can direct query the DB or load the data with its own patented columnar database onto disc and start querying it. QV has a complex data and metadata management, but you can create very smart pixel perfect dashboards. to sum up, QV is a perfect dashboarding tool that needs IT-involvement, Tableau is a cutting edge data discovery tool, which can be directly used by business (power) users.

Hi,

I have quickly gathered below my views about these two tools. Note that I have worked more with Tableau than QlikView. Since I wrote the below, I have attended a QlikView conference where they showed a little of the next version and it – as expected – had more drag/drop functionality similar to Tableau’s, so the two tools will keep fighting for the number 1 position. On the QlikView performance question, I have not experienced what is described. But I am convinced some people have issues because they put too much (more things than you actually need) into each model they make – which makes updates, etc. quite slow.

Original view:
In general I find both are powerful visualisation tools and think that their functionality overlap in a lot of ways, so if a customer already has one or the other, I would not try to push them to shift to the other.
I would recommend Tableau if business users are going to build dashboards etc. themselves or if exploring data is the key usage of the tool.
I would recommend QlikView if you want to build something that you will later model properly in a datamart/data warehouse – meaning using the QlikView data model as an interim solution (can be months as it is quite impressive what you can do) – as it can do proper star schema modelling etc.
Even though Tableau is easier to learn and more intuitive I am quite convinced there is nothing you can build in Tableau that you cannot also do in QlikView – it just takes longer and you kind of need to know what you want to build. On the other hand I am almost sure there are things you can build in QlikView that you cannot build in Tableau.

QlikView pros:

- Association model quite powerful
- You can configure almost anything
- Strong in building a proper star schema for your data
- “ETL” can do a lot

QlikView cons:

- Developer/Designer curve much higher than e.g. Tableau
- No help in what you can do with the data – no suggestions etc
- No distinguishing of what is measures and what is dimensions when designing
- Millions of properties to set
- No easy drag/drop etc. – all is property boxes

Tableau pros:

- Developer/Designer learning curve very low – highly intuitive
- Doing something you have not done before is almost always easy to find – is where you intuitively expect it
- Distinguish between Measures and Dimensions
- Suggest what visualisation is possible with the data you choose
- High use of drag and drop – and rightclick options etc.

Tableau cons:

- No data manipulation features when bringing data in
- Works best on a single view of data (one table etc.) though you can do link etc. in Tableau
- Some things can be tricky to do in Tableau if they are not “out-of-the-box”

Regards

Peter

Automatic Drill down provided by Tableau for date is just an awesome feature and yes, just by drag and drop we can create hierarchy.

I love to use heat maps in Tableau and the dimensions tab (like would you like the dashboard dimensions to be letter landscape or A4, so many option..Just WoW!!

QlikView, I love the expression window as I am more inclined to programming but when I published my dashboard and saw the final outcome,, I felt that the layout was not the same as what I had finalized.

What I do like about QlikView is that with every section they show you by color (Green, grey and white) whether data/fields are related or not with the current selection.

I hope QlikView will also provide something like Tableau Public with 1 GB free space. Its really very enjoyable experience to have Tableau public.

Showme option in Tableau is Ultimate. My thought is that, in terms of working on a dashboard, Tableau is way more professional. I don't buy QlikView's screen estate efficiency factor.

I have to create so many charts each day and then when I know that I have too many types of charts option in Tableau, i'ts really an awesome feeling to have dual axis on same charts with synchronization option. And you can at the same time create YOY changes or MOM changes within few clicks.

Last and not least, I try to connect with QlikView employees in India and Tableau employees. I got a very fast response from Tableau employees. QlikView employees believe that I should join their training first, but Tableau employees listened to my problem statement that I was facing and then replied with an email with the appropriate answer.

My final opinion, I want to see both these giants offering more options for the analytics domain. I salute both QlikView and Tableau for their service to the analytics domain.

Regards
Sid

I just know QlikView, Tableau only by name. But I know that QlikView is more a data vizualization tool than a BI one, there are no cubes and data modelisation is very weak/difficult.

If you're having trouble with the decision, it usually means either will do the job. In that case, go with Tableau. Much easier/faster to get started and to be productive. Easier user buy-in, great company, fantastic user community.

I haven't tried Qlik with really big data, but, have had no issue with a 1.1 BILLION (with a B) record data extract in Tableau, on my notebook.

In the majority of situations, performance is a product of good design (DBMS and views)...

They are both good and they actually both slow down with larger volumes. (Though I too have read that Tableau scales better)

Qlik’s interface was getting somewhat dated, but I heard something of a refresh was in the making.
Qlik is excellent for true “one filter affects all” paradigms, which is a paradigm I personally really like.

Tableau can be made to look more visually appealing.

I am personally a big fan of both, currently I think Tableau may have the edge, though a few years back I would have said the reverse.
This is of course completely subjective and as new releases come out things will no doubt keep shifting.

I hope this helps.
Ian

I think it is important to understand that it is not a situation of bias. It is a situation what fits for your organization, what would fit with your infrastructure that your company has designed.

You will need to do a comparison that works very wells against:

1. Architecture
2. Keeps data secure
3. Performance
4. Support
5. Ease of use
6. Expertise available in organization
7. More importantly does it work on what you already developed ?

I would also agree that Tableau is more data discovery than classic reporting. I also didn't find it to be a somewhat flexible tool for creating dashboards, which is where QlikView was stronger. And earlier someone touched on a good distinction as well:

If you want to drop a tool in front of the business users that they can jump into, explore, drag and drop, and create good visualizations on the fly, Tableau is likely the better bet.

If you plan to have some IT/BI involvement in the development and implementation, potentially some level of data modeling (externally, and within the tool itself), and design/integration control, then Qlikview maybe be the way to go, as I feel Qlikview is better for creating a packaged, polished BI product.

As for limits/performance, obviously hardware is the biggest impediment. In QlikView, I have run a load of multiple tables, a couple of 5 million + rows, and one with 60 Million + rows on a relatively old desktop with 8GB ram, and manipulating the data was still quite quick. E.g. Refresh after applying a filter was still only a second or two, and I can’t say I’ve heard many people raving about Tableau’s performance relative its competitors. If you’re working with larger data sets than that, which is quite possible, you’re going to want a beefier machine, and particularly more RAM. The same could be said for both tools. And for a larger implementation of either one, you’d likely need to go with a Server license and have that server sitting in-house, or utilize a cloud based solution.

I totally agree with Paul's comment, there is no magic here, just some more
help in discovering data and taking off some report creation off of IT once
data is prepared.

Hi,
I am beginning to think that any tool that is still using OLAP cube instead of a Columnar Database is not the tool of the BI future and seem to have limitations on how much data can be processed. They also seem to require a lot more IT involvement…I am currently looking into SiSense’s Prism tool.

Steve

I would recommend looking at Gartner's 2014 report which shows the pros and cons of both. www.gartner.com

I have to disagree with Francesco on Tableau being more of a classic reporting tool. I use Tableau for data discovery very effectively and using the existing features, ability to perform quick multi-regression correlations and now the direct link to the R statistical site the ability to conduct predictive analysis and forecasting has been greatly enhanced. A lot of times it just comes down to personal preference though.

We have just purchased Tableau, so soon I will have a first hand experience
in developing a project with this tool.

From my original research Tableau is much easier to use in terms of
functions provided where in Qlikview you have to write your own script for
time series, measures, aggregations.

I had no problem installing and using Tableu against our data, I did have
issues connecting from Qlikview.

On the other hand if the company has a spare developer that can learn
Qlikview and develop reports and applications they may want to choose it
for it's flexibility and intuitive search.

Not sure how Qlikview will behave with the amounts.

Sorry, I could not make up my mind as to which tool to select and the
financial side took over :-)

Renata

Well, it really depends on what you what do to with the tool.
Both are suitable for enterprise, both have a server installation with a fully featured web interface: vertical scalability is just a matter of server resources for both. Horizontal scalability is an issue for both due to the poor metadata layer.

As for QlikView, the real issue with "multiple tables" are:
- synthetic keys
- path length (number of hops - different tables to traverse in order to reach a field from a starting one)
generally speaking, synthetic key MUST be avoided in the model and the length of the path that connect the two fields should be as short as possible: data are stored compressed by column.

That being said, QlikView is a DATA DISCOVERY tool (i.e. discover new relations / patterns in your data) and FREE ANALYSIS tool. Tablaeu is more a classical reporting tool.
Both are really end-user oriented and have a above average visualization capabilities.

It is because both are desktop tools. It is back to the future. We did that with CR and Deski 10 years ago. You need a server component of any of these tools for performance. It all has to do with the size of the laptop of desktop. Both rely on the same resources for processing.

Disclosure: I work for SAP

Hi,

I have the same question!!!!!

I’m comparing both tools right now… but I haven’t reach a conclusion yet.

Regards,
Elisabete Maria Leitão de Miranda