12 Replies Latest reply on Dec 8, 2015 5:16 PM by Patrick A Van Der Hyde

    Grand Total at the Top

    Viswanath Sankar



      Is there a way to show the Grand Total at the top as opposed to the bottom?




        • 1. Re: Grand Total at the Top

          Hi Vishy,

          No, it is not built-in yet.

          Workaround: Make a separate worksheet with grand total which you put above the table in the dashboard.


          You can vote for getting this feature in Tableau here:


          Forum discussion

          • 2. Re: Grand Total at the Top
            Joe Mako

            Making a separate sheet and using a dashboard is less than ideal.


            Thank you for linking to the other threads, as you will find one comment the route that I recommend as a more ideal option, duplicate the data with custom SQL, granted, this may not be a good route in all situations. You can see an example in the attached workbook. This route also enables other capabilities not available in Tableau's built in Grand Total.


            see also Jonathan Drummey's 3-part post at http://drawingwithnumbers.artisart.org/customizing-grand-totals-part-3/

            • 3. Re: Grand Total at the Top
              Chris Gerrard

              This discussion is cousins with innumerable others having to do with the organization - what goes where, and decoration of the various elements within a viz.


              There's a crying, even screaming, need for a better model for structuring and populating the viz space.

              Tableau has an internal model that works very well in its narrow domain, but it doesn't accommodate the flexibility that many of us see as desirable, natural, even essential. It's simple and crude, with extremely limited configurability, and even that is much clumsier and awkward than it should be.


              There's a real, and vast, disconnect between Tableau's fast-and-easy approach to getting analytical results with very little effort and the real struggles involved in achieving anything other than Tableau's cookie cutter outputs. It's as if Tableau thought they solved the whole human-side problem once they figured out the first bit - getting the data connection and organization, and standard rendering mechanisms, in place, and stopped there.


              The product keeps advancing, but this hole isn't getting filled in, leaving huge gaps that take very specialized skills and deep insight to bridge.


              The whole idea that one would need to use custom SQL, under specific conditions, with an awareness of the limitations, subtleties, complexities, and consequences just to put a total at the top of a viz, is depressing.


              We might as well go back to writing code to render our analyses, where we'd at least have the control we need.

              Or start using the new upstart product that has Tableau's initial ease of use, embrace of visualization best practices, AND enables the flexibility and control over the visuals we need.

              • 4. Re: Grand Total at the Top
                Joe Mako



                I believe we have different expectations. It also sounds like you are in pain with desire for a tool that assists you in your approach to data and visual analysis, that Tableau is so close, yet so far away, that it hurts.


                > "It's as if Tableau thought they solved the whole human-side problem"

                I believe this to be a false statement, and it is not constructive.


                I would think that if Tableau's model does not address your needs, why not build an application that does address your needs? Scratch your own itch, 37signals style, or is there something out there now or in development that does address your needs, the "new upstart product" you mentioned?


                > "leaving huge gaps that take very specialized skills and deep insight to bridge."

                Tableau is not a complete BI solution, if you think it is or should be, I would be interested in discussing why you think so.


                To me, Tableau is great, and my tool of choice, precisely because of its model, and its narrow domain. It does one thing very well, the cycle of visual analysis, and I fear that if Tableau tried to be everything to everyone, solve every problem, built-in just one click away, and not requiring any understanding, then it would be a very poor product, and I would look for another tool.

                • 5. Re: Grand Total at the Top
                  Shawn Wallwork

                  To me, Tableau is great, and my tool of choice .... I fear that if Tableau tried to be everything to everyone, ... not requiring any understanding, then it would be a very poor product, and I would look for another tool.




                  • 6. Re: Grand Total at the Top
                    Jonathan Drummey

                    Hi Chris,


                    I know you've had years of experience with Tableau and I imagine from the title of your blog that you've been rubbed raw by all the little (and bigger) points of friction in the interface. I'd like to offer a different perspective:


                    I just spent several days in the last few weeks in demos and reviews of Tableau, QlikView, SAP, and SAS. I don't think you're going to find the mix of ease of use, data visualization excellence, and control over layout that you want coming from any of them, nor any of the other major vendors. They all have their market niches and philosophies carved out. If you want (mostly) total control over your dashboards, go with QlikView. If data governance is your thing, go with SAP. Hardcore statistical modeling, go with SAS. They all have "data visualization" tools (SAP and SAS each have at least 4 products apiece), but the latest and greatest from the other vendors are Tableau clones. Qlik.next has been announced and the descriptions all read like something from Tableau. I got to see SAP Visual Intelligence & SAS Visual Analytics, and both products are lacking in a lot of features because they are so new, and I highly doubt they are all going to get the bells and whistles of their predecessor products (though they certainly have gradients and 3D exploding pie charts). And Tableau, based on what they've been putting out over the last 16 months, is focused on getting more embedded into organizations by building new database connectors, supporting tighter integration with the new Javascript and extract APIs, getting more people interacting with Tableau with the web editing and dashboard improvements.


                    A couple of months ago I thought Tableau would be open to disruption from a new entrant in the combination of ease of use/visualization "best practice"/control over layout, but now I doubt that. With the other major vendors all building their Tableau clones to be "good enough" in this space, there might be some company (small or large) that builds a Tableau-like-we'd-want-it-to-be, but I don't think that would be awesome enough to really change the market. I think the next disruption will come from another direction. It might be in the cloud, maybe it's real-time analytics on "Big Data" with APIs to hook into operational systems (like an analytical app in the Emergency Department that can compare your medical records and current symptoms to identify your risk of complications & readmission in order to direct your care), I imagine with the new rendering pipeline and web editing of worksheets Tableau is trying to generate their own disruption with self-service BI, but really I don't know what that disruption will be. However, I don't think the disruptive threat is going to come from a product that enables people to put the Grand Totals anywhere they want. I'd argue that the really important issues are around enabling the right insights to the people who need them at the right time.



                    • 7. Re: Grand Total at the Top

                      Hi Chris, Joe, Shawn, and Jonathan
                      As happy I am that Chris is a strong advocate of a less frictious Tableau, just as happy am I that Joe, Shawn, and Jonathan are ready to protect the heart of Tableau - the visual analysis process - and emphasizing that Tableau is doing a good job. What I don't understand is the disagreement. I don't see any conflict, and believe Joe, Shawn, and Jonathan also want less frictions, such as grand totals at the top and other minor issues. If the topic is if Tableau is doing a good job or not, than I do understand the disagreement.


                      I agree with Chris - and hope we all do
                      I agree with Chris that there are (too) many minor frictions in Tableau today. These distract "the visual analysis cyclus" when we spend time in (unnecessary) workarounds and time-consuming routes to do our job.

                      I intended not to share my opinion this evening, but rather work on a visualization, but the first thing I met was a friction! namely changing the data source of an extract:

                      1. Tableau Desktop
                      2. Right-click datasource > Edit Connection > Ups, cannot edit connection
                      3. Create new workbook > Connect to data > Tableau Server > Login > Choose site >  Download a copy
                      4. Wait (sometimes quite long time)
                      5. Right-click datasource > Edit Connection > Edit Custom SQL
                      6. Publish to server

                      It would be much more intuitive, faster and less distracting to do this already in step 2 

                      So I chose to share my thoughts and feelings, because Chris is touching something I agree in.


                      Fact : Tableau is doing a good job
                      I do understand Tableau cannot solve all things in a short time, and even think they shouldn't, because new features will likely be better if they get necessary incubation time before implemented. That said, I am impressed how fast Tableau is moving forward ... including toward a complete BI solution (data server, extract API, etc). I look forward to version 8 and future versions, and have no plan to switch to "the new upstart product" ... yet.


                      Prioritization : Improve existing features
                      If I understand Chris correct (including reading his website), he wants to cut off frictions from existing features and - if necessary - slowing down with new features until existing are less frictious (better). I don't think he asking for a complete BI solution or a system that is everything to everyone. That said, I think Tableau is moving toward a complete BI solution with data server, extract API, etc.... and pleased they do


                      Conclusion : More developers on "The Friction Fighter Team"
                      I hope the team that are working on improving existing features will be up-prioritized with more developers than current 2-3 developers I think they have today, if I remember correctly a comment in another discussion. Maybe Chris with his passion to lessen frictions in Tableau will be great addition to that team!


                      Wish : Friction Overview
                      I wish there was a well organized and categorized friction list of open ideas. It is easier to discuss when we have something explicit to talk about, especially when I think we actually do agree about this.

                      • 8. Re: Grand Total at the Top
                        Joe Mako



                        I agree there are many little frictions in Tableau, and I put up with them and workaround them every time I open the application. This not what I was addressing in my reply, and I do not believe there is any disagreement on the points you raise from any user, we all have our pain points.


                        What I disagree with is the claim that there is a "need for a better model for structuring and populating the viz space" and that Tableau should change their model.


                        I am not disagreeing with the issues presented, I would absolutely love to have more control over grand totals, I am disagreeing with the solution suggested.


                        Johan, you have a lot of great feedback, and I'll note Dustin Smith so he can help, or route your concerns to those who can best help. I hope you continue to share your pain points, as you are not alone, and I appreciate how you framed your points. 

                        • 9. Re: Grand Total at the Top
                          Chris Gerrard

                          Holy moly! Do I really sound like that much of a grump? (I know, I know)


                          In case it's not clear: I love Tableau. Loved it when I first found it in the way back. Love it today. It's the first, best tool for almost every data analytical need I have. I've used it constantly for a long time, and make a very tidy living using it and helping my clients dramatically improve how they work by adopting it.


                          I'm really, really happy with the way Tableau Software is moving the product forward - introducing new functionalities and expanding its reach across multiple areas that matter to individual data analysts and organizations.


                          I bought and paid for Tableau out of my own pocket long before there was any realistic hope of getting paid for using it. I've followed along picking up crumbs from the experts, Joe first among you, constantly amazed at the depth of insight and understanding, and creativity that goes into solving some of the problems that have come up.


                          So, how can I be critical when I'm such a huge fan?

                          I want Tableau to be even better than it is. I want it to be perfect.

                          Because when it's perfect it will essentially (or effectively) vanish, leaving only us and our data, intimately coupled and in harmonious understanding.


                          Does it really matter that it's difficult to get just the table formatting you want?

                          Does it really take away from Tableau's intrinsic value that putting totals at the top requires the equivalent of tying our ankles behind our head?

                          Does it really matter that Tableau's bar spacing on multiple-bar graphs doesn't facilitate easily discriminating one dimension member's bars from it's neighbors, or that there's no straightforward way to make it do so?

                          Does it really matter that there's no way to make the pane invisible, or otherwise disappear, when you're really just trying to organize your data, to see what the inter-dimensional relationships are? (I use the ' '-valued calc field method and it takes only 5-12 seconds, but it still interrupts the flow)


                          Only in that every little thing that causes us to stumble, every little hiccup, every little wart and niggle, is magnified because of Tableau's core elegance. And the accumulation of all the bumps, knots, pits, and slippery bits creates a drag on the overall experience, which at some point becomes an onerous burden.


                          Tableau is far and away the best tool I've ever seen for rapidly and effectively analyzing data. In its sweet spot it's almost impossible to make it much better without developing the amazing Kreskin mode.


                          I might not be relevant or interesting, but... 25+ years ago I went to work for IBI as a FOCUS consultant because FOCUS provided unparalleled abilities to access and analyze business data. I was a product manager for PC and Unix versions of FOCUS during the transition from mainframe and character-mode computing to GUIs and network awareness. I watched what happened to an absolutely terrific product with a fantastic core ability - data access and ease of analysis, when it grew too many warts and knobs, kept expanding its operational base - adding additional data connectors and enterprise features, and didn't keep a bright light on the need to continually improve the user experience by adopting the best possible design concepts and ruthlessly removing even the smallest grains of sand in the gears. WebFOCUS is still out there, and IBI is undergoing something of a resurgence based upon its enterprise abilities, but in 1985 it was the dominant company in the business data analysis world.


                          Hmmm, what started out as a simple comment became a confessional. Odd, that.

                          • 10. Re: Grand Total at the Top
                            Dustin Smith

                            Hi All,


                            Sorry for not jumping in earlier.  Wanted to let everyone know that we (me + devs + prod. mgmt) have been watching this thread with a lot of interest.  These types of discussions are incredibly insightful for us since they mirror those we have internally.  How broad should the Tableau product be?  New features vs. Refining current capabilities - which should we focus on. Etc., Etc.  Obviously there are no definitive answers to many of these, but well stated and exampled threads like this never cease to help us consider things we may not have thought of before.


                            In other words, this is awesome on so many levels.




                            P.S.  Joe, thanks for the ping. 

                            • 11. Re: Grand Total at the Top

                              @Joe, thank you for kind words. This encourages me to continue to share more pain points.

                              • 12. Re: Grand Total at the Top
                                Patrick A Van Der Hyde

                                Hello Viswanath Sankar,


                                Yes it is years later but it is finally released.  As of version 9.2, Totals can now be placed on Top as desired in this post.  You'll see the idea is marked as 'released'.