July 21, 2014

Makeover Monday: Slicing Up the La Liga & Premier League Revenue Pies

Over the weekend, Reader Ben Jones sent me a message on Twitter pointing me to this post by @kmgfootball, which he thought would be great for a Makeover Monday example.


The point the writer is trying make is clear: Barcelona & Real Madrid combined get almost as much revenue as the rest of the clubs in La Liga, whereas in the Premier League, there's much more parity and revenue sharing. The problem is that the charts are basically unreadable.
  1. There are 20 slices in each pie.
  2. Each slices contains way too much content: team logo, team name and revenue, yet it lacks the percentage each team takes in, which is more meaningful than the revenue in this case.
  3. The image is blurry.
  4. The fonts are tiny.
I recreated the data and used Tableau to build a lollipop chart instead of a pie chart.


I chose a lollipop chart because I wanted to show the data in a bar chart view, but accentuate the end points. I then color-coded the dot on the end of the lollipop by the revenue and kept the range consisted across the leagues. In addition:
  1. I kept the scales for the bars the same on both charts so that you could see how the leagues compare to each other.
  2. I converted the Pounds to Euros (1 british pound sterling = 1.26 Euro) to make the data more comparable.
  3. I included the share of revenue as a label on the lollipop.
This view makes two points very obvious:
  1. There are only two teams that matter to TV networks in Spain.
  2. The Premier League is very, very rich. The team will the lowest revenue allocation is higher than the third highest team in La Liga.
How would you visualize this data differently? The lollipops are merely one approach. Download the Tableau workbook here and leave a comment with a link to your version.

June 30, 2014

Makeover Monday: How Americans Spend Their Online Time

Time spent online - we all do a LOT of it. Whether it be Facebook, Instagram, YouTube, ESPN, or any other site, the data shows that the average American spends a LOT of time online. Business Insider took a look at this recently with this circular chart:


The content of this slide is quite interesting, but their approach for visualizing it could be better.  I feel like they went after cute instead of simple. 

In his article "Our Irresistible Fascination with All Things Circular", Stephen Few talks about how hard it is to make comparisons with circles. Many designers use the diameter of circles to distinguish size. It's hard to tell precisely if Business Insider made this mistake, which in itself is a problem.  If I can't easily tell that the 19 for Online Games is just over half of Social Networks, then I'm having to do too much work.

In addition, they have plotted the circles on two rows. Comparing any two of these circles that are not directly adjacent to each other is too hard.

I would have created a simple bar chart instead.


I've made a two notable changes:
  1. I converted the circles to bars, making comparisons much easier.
  2. I added the % of total time spent online to the end of the bars to give additional context.
Do you agree that this is easier to read? If not, why not? How would you do visualize this data differently? Download the data here or the Tableau workbook here.

June 16, 2014

Makeover Monday: Where do World Cup players play professionally?

Unless you live under a rock, you know the World Cup started last week. In this spirit, Chart of the Day published this chart showing where the players play professionally.
This chart seems simple enough, yet they make it too hard on the reader.
  • A horizontal bar chart would be easier to read
  • The sorting is backwards
  • They're not doing enough to show the geographic distribution
In about 10 minutes this morning, I built this viz with Tableau for Mac to address these concerns.  Go USA! I believe that we can win!

June 9, 2014

Makeover Monday: Label bar charts for easier comprehension

I'm in the market for Chromebooks for my twins and was reading quite an excellent overview by The Wirecutter. In the middle of the article is this chart comparing the performance of various Chromebooks:


This chart seems innocent enough, yet I found myself having to constantly reference the legend because they didn't bother including the labels directly on the chart. A more understandable alternative might look like this:


In this chart I have:
  1. Added labels for the bars
  2. Removed the legend and the different colors for each Chromebook
  3. Made the bar horizontal bars so that the labels are easier to read. I also find it easier to compare the length of the bars on horizontal bar charts, but that's a personal preference.
  4. Added a metric to show how much slower the other Chromebooks are compared to Wirecutter's recommendation (Dell Chromebook) and colored the bars by the % difference. This helps provide more context to the speed comparisons and I don't have to do the math in my head.

June 5, 2014

Hey QlikView - If you were the best tool, I would have chosen you!

Earlier this week several people sent me a slide from a sales presentation from QlikView. In the slide, QlikView used a quote from me, my image and my name without my permission. (NOTE: Please see update at the bottom of this post.) The quote was in reference to a poll I ran (which I've since taken down because I wanted to stop the voting) in which I was critical of Tableau's performance. Apparently QlikView is using this quote as a competitive weapon against Tableau. QlikView has sent this to multiple customers in an effort to impact sales.

I can think of only one reason why QlikView would do this: Tableau must be killing them!

QlikView is obviously grasping for straws. While I'm sure Tableau is having to do some unplanned firefighting, they might see this as a compliment. Only competitors that are getting their asses kicked would resort to these tactics. All you have to do is look at the latest Gartner Magic Quadrant to see that Tableau is the leading platform.
I like to tell Tableau, and I know many other people tell them this too, that we chose them because they are the best end-to-end solution on the market. But they need to be aware of complacence. If another tool comes around that better meets my needs, you can be sure that I will switch. In the end, I want a tool that makes me look good.

Performance remains my #1 concern with Tableau. Tableau is not an exclusively in-memory solution like QlikView, so Qlik comparing their performance to Tableau's is not an apples-to-apples comparison. I like to tell the Tableau execs that if they can make Tableau fast for data at Facebook's scale, imagine what they can do for the more typical customer. I am extremely confident that Tableau is listening and are actively working to address my performance concerns.

Here are some random thoughts about why I love Tableau:
  • One of the reasons I respect Tableau, its management and its employees so much is that they would never use anything I say without asking my permission. Tableau has quoted me quite often: on their blog posts, on their website, as part of ads, in sales presentations. But Tableau has asked permission EVERY SINGLE TIME.
  • I have never seen an example where Tableau has acted in the manner in which QlikView has.
  • I chose Tableau because it's the best overall data visualization tool on the market.QlikView, I didn't choose you because you simply can't compete. If you want to win, make a better tool.
  • Tableau listens unlike any other vendor I've ever known. They care about their customers. Customers drive feature development. Two great examples of Tableau listening to their customers are: (1) the upcoming Mac client and (2) the Apply button on multi-select drop-downs. I personally worked with Tableau to help create the Apply button, which they initially called the “Andy” button internally. They took the time to personally reach out to me and gather my feedback and showed me prototypes along the way.
  • Tableau has a community of users unlike any other product I've used. If you've ever been to one of their conferences, you know what I mean. If you haven't been to one, you must go. Tableau's forums are completely user driven. If you ever have a question, ask on the forum and you're all but guaranteed to get a high-quality response...for free! There are people like Joe Mako and Jonathan Drummey that hold working sessions with users to help them learn. I write this blog because it's a way that I can share with the community. The number of wonderful Tableau bloggers is astounding. Do you ever hear of other vendors having their users promote their product like Tableau's users do? I think not!
  • They have a mission and vision that matches my own.
  • I've spoken at four Tableau Customer Conferences with two more planned this year. I've spoken at their annual company meeting. Why would I speak on their behalf if I wasn't their biggest advocate?
  • Without Tableau, there's no way I would be working at Facebook. I have some incredible success stories from my days at Coke (like using Tableau with buyers and making $30M in 10 minutes). And the success stories have continued here at Facebook.
I'm disappointed and saddened by QlikView's behavior. If anyone knows who I can contact to get them to stop using my name, my image and the quote, I would greatly appreciate it. They didn't ask permission and I don't like how they're using me to market their product.

If there are any companies or customers that are seeing this slide from QlikView and it's making you question which tool you should use, then consider why they would be resorting to such tactics. Only companies that are playing catchup behave like this. Think about why they would do this. Question their sales staff.

I've always been more than happy to talk about the pros and cons of BI tools as I know them. If you are a prospective customer comparing QlikView and Tableau and have any questions, feel free to email me. I'd be more than happy to answer any questions you have.


UPDATE: 
6/6/2014 - I need to recognize Maria Scurry from QlikTech for being so responsive to my concerns. Maria left a comment shortly after this blog post went live. In addition, she has been very responsive to my emails and she has confirmed that she found the source of the slide in question and it will not be used publicly or internally from this point on.

June 2, 2014

Tableau Tip: Use the "refresh" URL parameter to avoid hitting the Server cache

I've been experimenting with Tableau for Mac quite a bit. This past week I've been focusing a lot on live connections to MySQL to replace some existing internal dashboards with more insightful ones. The great thing about our MySQL environment is that it's super fast and I can turn around a dashboard built on a template in less than an hour. While this has been awesome for proving the value of Tableau for Mac, I learned a lesson over the weekend about the Tableau Server cache.

We're at a luxury here at Facebook that we can have a separate environment for the 8.2 beta. In production, we all share a single server, so it's unlikely that you hit the cache. However, on the beta server, there are only a handful of us using the Server, which led to email over the weekend from my users that their dashboards with live connections are not showing the latest data. Why? Tableau is using the cache (as it should).

To solve this, all I have to do is force the dashboard to refresh every time the user hits the server.  I accomplish this by adding the ":refresh" parameter to the URL. My URL now looks like:

https://[TableauServerURL]/views/[WorkbookName]/[ViewName]?:embed=y&:toolbar=no&:refresh

The refresh option on the end is what tells Tableau Server to re-render the view.  Of course, you don't have to have the embed and toolbar parameters in the URL, but I like the way my view looks on Tableau Server when I include those options. For a full list of embed parameter, go to this link.

Makeover Monday: The Face Pie - Taking an Analogy Too Far

Edward Tufte likes to say "the only worse design than a pie chart is several of them." Today's makeover takes this even one step farther. Someone at Pew Research decided that since the topic of their chart was "The Changing Face of America" that they should uses faces instead of pies.

Humans are poor at judging angles in a pie. I can't even imagine how bad we are at judging angles in obscure shapes. Also, the purpose of this graphic is to show change. It’s very difficult to understand trends in a series of face pies. I would present the data as a line chart like this:

image

Now it’s much, much easier to see the changing demographics of the United States. Keep it simple people!

May 19, 2014

Makeover Monday: India's BSE Sensex as an Area Chart

I'm not a huge fan of area charts, especially stacked area charts. Much like bar charts, the axis for an area chart needs to start at zero, otherwise you're not showing the total area, thus defeating the purpose of using an area chart in the first place.

As an example, here's a recent chart from Chart of the Day about India's stock market.

Notice how they've started the axis at 2,600. This distorts the slope of the graph and also the area of the graph is not complete.

Contrast this to Yahoo!'s chart, which is executed perfectly.

2014-05-19_09-18-12

Yahoo! is showing the entire scale and nice proportions. Without noticing these subtle differences, you might interpret a very different story. Moral of the story: always start the axis for area charts at zero.

May 16, 2014

Tableau Tip: How the familiar Windows Right-click+Drag, CTRL+Drag and Shift+Drag work on Tableau for Mac

Most people that have been using Tableau for years have become very familiar with right-click+dragging pills to bring up context menu (particularly for dates), plus lots of other keyboard shortcuts. The Mac OS works differently, so if you're considering switching to Tableau for Mac, here are some handy tips to keep in mind (via this response by Vijay Doshi on the Tableau Forums):

The Option+Left Click is necessary because Mac and windows handle the right-click event slightly differently.

On windows there are effectively two events:

    1. On right-click press down, no menu is shown and you can then drag the pill around.
    2. On right-click release, Tableau shows the "right-click menu".

On Mac, the convention is on right-click press down you see the context menu (try it in finder). Tableau cannot show both the context menu and drag. Thus, Tableau has to use a keyboard modifier.

Command+Drag will duplicate a field (i.e., copying the same field from/to a shelf from/to LOD). On Windows this is Control+Drag (Tableau could not keep it the same because on Mac Control+Left-click = context menu by convention).

Option+Drag will show you the aggregation options for the field.

One bug that I've reported is that when you Command+Drag to duplicate a field, you won't see the little + that you do on Windows that is a visual clue that you're duplicating the field.

May 12, 2014

Makeover Monday: Will Johnny Manziel stop the run of terrible QBs for the Cleveland Browns?

The NFL Draft is somewhat of a national holiday here in the US. It’s the day when all fans can dream of their team using their picks to turn the fortunes of their franchise around. QBs are particularly in the spotlight. In this spirit, Chart of the Day published a chart on Friday after the first round of the NFL Draft showing the number of starting QBs for each NFL team since 1999.

Accompanying the chart was this statement:

“Since 1999, 20 different quarterbacks have started for the Browns, the most in the NFL. Meanwhile, the New England Patriots have had just three starting quarterbacks over the same span.”

This statement implies that there is a relationship between number of starting QBs and success (because they’re only talking about the outliers), yet they provide no additional context. I downloaded the winning percentages for every NFL team since 1999 from SportingCharts.com and joined it to the Chart of the Day data. 

I like how they’ve sorted the bars in ascending order by number of QBs, yet I don’t like how they always have the labels rotated. A horizontal bar chart would be much easier to read.

Given that we can easily compare number of QBs and win %, I turned to Tableau and build this simple view.

Looking at the data this way, it becomes much more clear that there is no direct correlation between the number of starting QBs and win % (as implied by COTD).

  • Detroit is an absolutely horrible franchise, yet they’re right in the middle of the pack with starting QBs. 
  • Chicago has a winning record, yet they’ve used the third most QBs.
  • Cincinnati and Houston have had pretty stable QB situations, yet they don’t win even half of their games.

One particular insight that sticks out to me is the amazing amount of parity that exists in the NFL. 25 or 32 teams have between 40-60% win percentage. In any given season, you can pretty much count on around 80% of the teams winning between 9.6 and 6.4 games per season. This is exactly what the NFL wants and is a large reason that they run a socialist type model of revenue sharing.

What else do you see? You can click on a team to highlight them. Download the data here and the workbook here.