Data Viz Done Right

August 19, 2018

Makeover Monday: Africa's Deadliest Armed Conflicts

No comments
Week 34 is a collaboration with ACLED to help them visualize armed conflict in Africa. Here's the original visualization:

What works well?

  • The map is zoomed into the areas that are the focus of the data.
  • The color palette has enough variation to distinguish between the type of violence.
  • Putting every incident on the map helps show the volume of conflicts.

What could be improved?

  • Because the dots are overlapping and there's no transparency, you can't see what is behind each dot. Are there battles behind the violence against citizens? I have no idea from this chart.
  • Some of the dots look like they are sized differently. Why?
  • There's no title.
  • The data source isn't referenced.
  • While the color palette give a good range, I'm not sold on the color choices. These are all bad topics, but the blue could come across as not bad.
  • Have fatalities been accounted for?

What I did

  • I like the idea of a map for this data set, but I think a density map will work much better. As this is now only in beta, I can only post an image. I'll post the interactive version once Tableau 10.3 comes out.
  • I wanted to understand where the deadliest conflicts occur, so I create a metric for fatalities per incident.
  • The density is colored by this metric.
  • I separated out the types of violence to address the issue of overlapping.

With that, here's my Makeover Monday week 34 about Africa's deadliest armed conflicts.

August 16, 2018

Workout Wednesday: How many customers make up XX% of sales?

No comments
On yesterday's Makeover Monday Viz Review, guest host Ann Jackson gave me a bit of grief for not participating in too many Workout Wednesday's this year. I've taken on her challenge and this week I've complete week 33. This week was a community challenge from Donna Coles, who has completed every WW since it started in January 2017.

Find all of the details of the challenge here. The only requirement I chose to ignore was the dashboard size. I created it 700x700 instead of 800x800 because it fit on my screen nicely.

Some hints for this challenge:

  • You'll need to know your table calcs.
  • This video will be helpful in creating the Pareto chart.
  • You'll need extra calculations for the tooltip and the lines and the dot.

That's all I'm going to give you; the rest you need to figure out on your own. Good luck!

August 14, 2018

The Greatest Tableau Tip EVER: Exporting Made Simple!


UPDATE: 14 August 2018

Version 2018.2 of Tableau introduced Dashboard Extensions. To make exporting data from dashboards as easy as possible, The Information Lab CTO and Zen Master Hall of Famer Craig Bloodworth created the Export All extension. What does this do?
No matter how much you try to convince them, there will always be some users who want to reduce your beautiful Tableau charts to a table of numbers in Excel. So if they're going to do it anyway, you may as well give them a simple, controlled way, generating one Excel workbook and not a bunch of CSVs. With the Export All extension for Tableau Server you can place a simple button onto your dashboard, choose which sheets & columns are exported, and with one click your users can download a clean & tidy Excel workbook.
If you have at least version Tableau 2018.2, then I'd highly recommend the use of this extensions versus the technique outlined below.


We’ve all heard this question before: How can I export a CSV in Tableau? To be honest, it’s quite the pain and way more difficult than it should be. There have always been a few options.
  1. Users can click on a specific sheet on a dashboard and then export that via the tiny button on the toolbar, but that has a few of its own problems: (1) You may not want to show the toolbar therefore making the export impossible, (2) People have to be trained to know exactly where to click to get it just right, and (3) You have no control over the output of the CSV.
  2. You can export a CSV using Tabcmd, but that’s not useful for the average dashboard consumer.
  3. You can add .csv to the end of the URL like http://[Tableau Server Location]/views/[Workbook Name]/[View Name].csv.  But again, you never know what that output is going to look like.

Yesterday I learned an incredibly valuable trick that would make option 3 (adding .csv to the URL) export exactly the CSV you want. Let’s look at an example.

On this dashboard, notice how I added an Image object (I used an Excel icon) on the dashboard. I’ve floated it to the upper right and made it small. All the user has to do is click on that icon and they get a nice CSV. Go ahead, try it!

Did you try it? If you did, and you opened the CSV, you may have noticed this looks remarkably like the data that you would hope to get if you exported the data for the line chart. But I didn’t export the line chart at all. Here’s the trick.
  1. Add the image icon on the dashboard and place it wherever you like.
  2. Add a URL link to the image.  In the dashboard above, the URL is

    This is where the magic sauce happens. When you add .csv to the end of a Tableau URL, Tableau will export the first sheet on the dashboard alphabetically.  Yes, it’s that simple!  And it’s totally undocumented.  Special thanks to the one and only Tableau Jedi Mark Rueter for this tip!  But note that Tableau orders upper case before lower case. 
  3. What I did was float a sheet named AExport onto the dashboard.  I changed the height to 1 and made everything white and transparent and chose Fit Entire View so that it would be inconspicuous.  I had to name it with a capital A so that it would be the first sheet alphabetically on the dashboard.


    The AExport worksheet started off like this:


    Basically, you can put anything you want on this sheet.  I then changed the transparency to 0% on the Color shelf, changed the default worksheet color to white, removed the row banding and removed the row and column dividers.  The worksheet now looks like this:

That’s all there is to it! To recap:
  1. Create a worksheet that you want to export.
  2. Remove all of the formatting to make it look invisible.
  3. Be sure to give it a name that makes it first alphabetically on the dashboard.
  4. Place the worksheet on the dashboard, float it, make it fit the entire view, make it really small, move it somewhere inconspicuous.
  5. Add an image onto the dashboard, float it and add a URL to it that is the URL for the dashboard with .csv on the end.

This is a game changer!! Download the sample workbook here.

August 13, 2018

Makeover Monday: How many episodes aired for each of Anthony Bourdain's shows?

No comments
Anthony Bourdain, the great chef, travel documentarian, and host of the some the best show about international cooking, travel, and communities, died on 8 June 2018 and with his passing, we lost one of the most remarkable people on the planet.

To celebrate his life, for Makeover Monday week 33, we're looking at data about all of his TV programs. The visualization to makeover comes to us from Christine Zhang:

What works well?

  • The map makes the geographical breadth easy to understand.
  • The colors are easy to distinguish.
  • Using dots gives each location equal representation.

What could be improved?

  • Filtering would be good.
  • Locations with multiple episodes are not apparent.
  • The title doesn't really explain what the viz is about.

What I did

  • Focused on the frequencies of the shows as I'm not totally familiar with all of them
  • Made the map information supplementary through the tooltip
  • Use a title that explains what the viz is about

Special thank you to Christine Zhang for this week's data set.

August 9, 2018

The Petr Cech of Chelsea was outstanding...then he moved to Arsenal

No comments
Arsenal signed goalkeeper Bern Leno in the offseason. Rumors have been swirling about when, not if, he'll takeover as their top choice keeper. Three summers ago, Arsenal signed Petr Cech away from Chelsea, and with that signing, Arsenal had hoped to sure up their defense.

In an Arsenalesque sense of optimism, Arsene Wenger thought bringing in a great goalkeeper would solve their defensive woes. Many fans, however, knew that the real problems were in front of the goalkeeper. Without a solid defense, a goalkeeper cannot be effective.

This led me to thinking about how Cech's first three seasons compared to his first three seasons with Chelsea, when he was widely considered one of the best goalkeepers in the World. The data shows that Arsenal more or less ruined him. Or did Chelsea's stellar defense make him better than he really is?

History of the Premier League Table

No comments
The 2018-19 Premier League season kicks off tomorrow night with an enticing match between Manchester United and Leicester City. This reminded me about a viz I had created at the end of the last season as a way of practicing stepped lines in Tableau.

When I originally created this, I had to use table calcs to get the stepped lines to work, which can get complicated and is very time consuming. Now with stepped lines, it merely a matter of changing the lines type.

I decided to add in a couple of user options:

  1. Which team do you want to highlight?
  2. How do you want to compare the teams? By total points for the season of the final position in the table?
  3. Not all teams have been in the EPL for all 17 years, so I provided an option to filter down to just the teams that have been in the EPL for the user specified number of years.

And here's the viz for you to explore. Enjoy!

August 7, 2018

Makeover Monday: And The Ice Melts Away (as a Radial Chart)

No comments
The second session we had at the DS today was taught by Laine Caruzca and she showed us how to build a radial chart in Tableau based on this viz she created for Makeover Monday week 20:

I'd never built a radial chart from scratch before, so I was excited to learn to build a second new chart type today. In Laine's tutorial, she used the Custom SQL option that's available in the Legacy data connector in Tableau for Windows. However, there's no custom sql option on a Mac, so I decided to create the data structure using Tableau Prep.

Download the Flow here

This is a pretty straightforward workflow. You split the data into parts to create the start and end points, then union them back together, along with some cleanup along the way.

Laine provided all of the details of the table calcs and the bin needed to create the curves, so I follow her steps using the Arctic Sea Ice Extent data from Makeover Monday week 15. That worked rather perfectly!

From there it was on to the polish. I love Pablo Gomez's style, so I used this radial chart viz of his as inspiration for the overall design, and then I used the wording from Arpit Arora's Makeover Monday viz about the same topic to help frame the messaging.

Makeover Monday: Jumpy Curvy European Irish Whiskey Sales

No comments
A big part of learning at The Data School is the students teaching what they learn. Last week, Nil Macher created this jump plot of The Big Mac Index based on a technique he learned by downloading this viz from Mark Bradbourne and reverse engineering it.

Nils' viz inspired by Mark Bradbourne

Today, Nils taught us how he shaped the data and built the viz, then we each took a Makeover Monday data set and applied what we learned. I chose to use the Irish Whiskey sales data from week 11.

I started by shaping the data in Alteryx via these steps in my workflow:

I then created a jump plot similar to Nils and also found a curvy plot interesting too, so I decided to include both via a parameter. Another fun day of learning! Never stop!

August 6, 2018

Makeover Monday: America's Contribution to Worldwide Research & Development Steadily Declined From 2011 to 2015

No comments
Back to London and back to work this week after a wonderful holiday in Scotland. This week, Eva is asking us to makeover this visualization from

What works well?

  • The title is clear, making the topic easily known.
  • The big numbers work well for the larger circles.
  • It's an innovative design, but that doesn't mean it effective.
  • Citing the sources

What could be improved?

  • Pretty much everything
  • The sequence is too hard to understand.
  • The color legend is sorted descending.
  • The colors on the ends are too similar.
  • There's way too much text, making it all feel very squished.
  • The color ranges aren't of equivalent size.
  • The maps of each country inside the circles don't add any value.
  • It would take forever to find out where a country ranked.

My Goals

  • Throw away everything they did and start over.
  • I really liked the simplicity of Eva's viz, so I went with her idea of focusing on just the US.
  • I wanted to understand how the contribution the US has made has changed over time.
  • Use the official Makeover Monday color palette

July 30, 2018

Makeover Monday: How has The Big Mac Index changed since January 2012?

No comments
I'm on holiday in Scotland this week so this has to be a quick Makeover Monday for me. Pardon me for being brief. Here's the original visualization from The Economist:

What works well?

  • Using the footnotes to help describe the caveats in the data
  • Including a reference line at zero so that it's easy to see if the country is over- or under-indexed
  • Including the latest price to the right for context
  • Sorting by the latest price
  • Using two colors that are easy to distinguish from each other
  • Subtitle explains the metrics in the viz

What could be improved?

  • The timeframe is so short that it's hard to see much change at all between the data points.
  • I have no idea what happened between these two points.
  • There's no explanation as to why this is the selected list of countries.
  • Sizing the dots for the most recent price doesn't add much value.
  • The white gridlines are too strong for my liking; I find them distracting.
  • I would exclude the Euro Zone since not all Euro countries are included in the viz and it's also the only aggregate included. If you look at the chart alone without reading the article, it doesn't make sense to include it.

What I did

  • I created some simple sparklines. Right after I thought about it, Rodrigo Calloni posted this viz which was nearly identical to what I wanted to create. The difference for me was that I wanted to look at the change in the price of a Big Mac over time, whereas Rodrigo looked at the change vs. the US price over time.
  • I used the colors from the original viz. I liked how they worked together.
  • I included some numbers for context.
  • I limited the data set to only countries that had been in all of the 14 most recent surveys.
  • I create a simplified mobile version that removes some of the BANs in order to fit onto the width of a mobile device.

July 26, 2018

Makeover Monday: Maternity Leave Pay Rate in OECD Countries

No comments
During this week's viz review, Eva and I were providing some feedback on this viz by Anik Sircar:

While we were giving the feedback, I had an idea for how I could use Anik's work as inspiration for a slightly different version of the chart. Before I get to that, here's his next version after iterating.

I love the simplification of the view; it's now much easier to understand. My take on this turns his view into a more minimalist dot plot that breaks it down into two columns that conveniently are above the OECD average on the left and below on the right. Originally I had it as three columns, but Eva suggested two and it worked out perfectly.

Thanks for the inspiration Anik!

July 24, 2018

Tableau Tip Tuesday: How to group items into dynamic halves, tertiles, quartiles, and quintiles

No comments
This week's tip came about from a client request at The Data School last week. The customer was looking to understand how quickly different doctors adopted different medications. I worked with Alexander Fridriksson to solve this problem using table calculation.

In Alexander's example, he needed to break the customers down into thirds: early adopters, the next 33% of adopters and late adopters. The beauty of this solution is that it dynamically recategorizes the doctors based on the marks in the view.

As we can't share specific client examples, this video shows you how to use the RANK_PERCENTILE table calculation to "bin" states based on their adoption date. An adoption rate in this case is the first time a product was sold in a category.


July 22, 2018

Makeover Monday: Employment-Protected Leave of Absence for Mothers

No comments
Eva has picked a nice simple data set this week about leave of absence protection in OECD (and various other) countries. Here's the original viz:

What works well?

  • The countries are clearly ordered by the weeks of paid maternity leave.
  • Including the three aggregations for context and giving them their own color so they stand out.
  • The subtitle explains what the chart represents.

What could be improved?

  • Remove all of the dark borders
  • Remove "Panel A" and "Panel B" from the chart headers
  • Change the title to summarize the findings in the data
  • Make understanding the relationship between the two data points easier

What I did

  1. Removed any non-OECD countries and aggregates
  2. Update the OECD averages in the source data since it wasn't calculating correctly
  3. Converted the original diverging bar chart to a scatterplot
  4. Included OECD average lines for both metrics
  5. Color-coded each quadrant
  6. Highlighted the United States' horrific performance
  7. Include the OECD average in the tooltip for context
  8. Enabled the highlighter option to allow the user to pick a country of their choice