Business Intelligence Blog

Latest Stories

Featured Stories

Filter By Categories
Matt C Atkins
By
March 08, 2019

Tableau Dashboards - The Do's and Don'ts (Part 4)

Due to being offline last week, I skipped a post, apologies, so let's catch up!  In the third post of this blog series, I talk briefly about a couple more things you should think about when designing and building a dashboard.  Using inspiration for your own build, and properly utilising your dashboard real estate.

 

working_matrix_4

So, you have Tableau Desktop, you have a dataset, and you can already see the insights and stories you can extract from said data.  A dashboard will be an excellent medium for communicating those insights.  For your dashboard to be genuinely awesome and empowering for those that make decisions, you should really take the time to think about the design and content.  Plan your build and consider some very important elements.

 

 

green tick_smallUse BANs (Big-Ass Numbers)

 

One sure-fire way of quickly and clearly communicating those key metrics - creating that Jedi dashboard that reaches out to its audience and says; 

 

numbers_looking_for

 

Okay, so I'm not sure our friend Obi-Wan has used Tableau yet, but you get my point.  You can really guide your audience to the most pertinent points of your dashboard. 

 

In Visual Analytics, there is indeed evidence that the human eye is naturally drawn to bigger numbers.  Imagine you put these BANs toward the top-left too...you could well be on to a winner there!  Now let's be absolutely clear here...whilst you've got the attention of your audience and put their gaze where you want it to begin with...this does not necessarily mean the information is being processed.  It is where the user starts...be sure to add descriptors where needed, and the flow of information can begin.

 

Let's just take a quick look at the two examples below...which style of design would you feel you could more comfortably get to grips with?

 

BANs_1

I prefer this one below personally...it gives me a clear idea of the top level values and then I can go from there.

 

BANs_2

With these BANs, you can even think about using them as colour legends or comparatives.  You might want to consider using say, Green and Red to compare your values to Target values or Prior Year values.  You might want to use your numbers or backgrounds as legends for other elements in your dashboard.  I mean the options are numerous...it's up to you...and your audience. 

Visual Analysis Best Practices: A Guidebook

red cross_smallDon't use the same old charts

 

Part of visual analytics is choosing the appropriate chart for your data.  Tableau has a wealth of different chart options available via the 'Show Me' function and I urge you to explore these.  However, don't just stick with these...there are some really cool charts you can build, some with mathematics - sunburst radials, sankey diagrams, word clouds, donuts, slopes...you really can be very creative.  

 

Let's get back on point though. What I'm saying is, don't just settle for the first viz you come across.  Yes...with certain data types, there are charts that lend themselves naturally - like time based data suits a line chart.  But don't assume this chart will tell your story as clear as you'd like.  There could be other options.  

 

Below, I give two examples which I have taken directly from the Tableau site.  They demonstrate perfectly the point I am trying to make here.  

 

same_1

same_2-1

Hopefully, you can see that using an alternative to the line chart here, a heat map gives a much clearer picture and we can see the larger concentration in September.  So, again, consider your options, look at your viz and 'be the customer' for a moment.  Does it tell you what you want quickly and efficiently?  Can you visualise this data another way to really cut down the time it takes to reach a conclusion?

 

When you're looking at things like changes in rank, or part changes for a whole element, the line chart could well be the worst choice.  Tinker with your viz - you will have numerous iterations perhaps but ultimately, if the end result presents the story effectively...everybody wins. 

 

That's it for another week.  I hope you found this blog post useful.  Next week will be my last post of this series and you'll have another couple of things to think about.

 

If you'd like to find out more about how Promatix can help you with Tableau, please do send us a message! 

Contact Us

 

Subscribe Email