bring on the bar charts

This month's #SWDchallenge was a straightforward one: create a basic bar chart. We sometimes avoid bar charts because they seem common or boring. Our view is that's the wrong approach. We should use them and use them frequently because they are common. The beauty of a bar chart is that our audience already knows how to read them and they can instead focus on what the data actually says. So we challenged you to teach us something new with a basic bar chart with bonus points for thoughtful use of color and annotations to call action to the main takeaway.

85 of you delivered, with bar charts ranging from single series vertical & horizontal to stacked to diverging. In keeping with the theme, we visualized your chosen topics to see what's trending on #SWDchallenge readers' minds.


Within the business category, profitability of top companies was top of mind as three readers—Josh S., Sateesh and Tausif—chose the same dataset! (Possibly a #MakeoverMonday repurposing?)

A couple notes to those who submitted examples: THANK YOU for taking the time and sharing your work. The makeovers are posted below in alphabetical order by first name (+ last initial when needed; we omitted full last names in respect of those who would rather remain anonymous). If you tweeted or thought you submitted one but but don't see it here, email your submission (including your graph attached as .png) to and we'll work to include any late entries this week.

Without further ado, here is a recap of March 2018's #SWDchallenge submissions. As you scroll though these, notice how easy it is to actually see the data! We can't overemphasize the beauty of the basic bar chart.


Conventional financing represented two-thirds of the Northern Virginia real estate transactions in February 2018. Home buyers using government-backed financing will have challenges competing against individuals using conventional loans in the upcoming spring selling season. This chart was designed in Canva. Visit for the latest real estate sales trends in the Northern Virginia area.


Adam C.

I created a basic bar chart showing why my hockey team is in second to last place in the NHL. It turns out you need to actually score to win in hockey. I stuck to minimal monochrome colors used only to call out my team's performance in this area. I scraped with Google Sheets and visualized the data in Tableau. Link | Blog | Twitter | LinkedIn


Adam G.

March's challenge centered on the humble bar chart, the data subject, a homage to the recent passing of Sir Roger Bannister (3rd March 2018), the first man to go under 4 minutes to run a MILE. This viz focuses on the world record holders below this magic 4 minute barrier, set in 1954.
Blog | Twitter | LinkedIn

adam g.png


Just a simple bar chart comparing the Electricity Consumption by Household in 1990 vs. 2016, just 1 country is consuming less electricity compared to 1990, Germany. For the 2016 values I used the bar charts and for the 1990 values I used a dash, I cannot add the value labels on the bars due to lack of space and the graph would look crowded so I decided to put the values at the bottom of the chart, just below the name of the countries, the 2016 values are bigger and the color is the same as the bars; the 1990 values are smaller and they are in black. The issue with Excel is that sometimes, if you want to add details, you practically have to handmade your chart, in other BI tools just a few clicks and done however I still love the detailed process that involves vizzing in Excel! Twitter | LinkedIn



For this month's challenge I was rather shocked to hear that there were as many as 8 Fast and the Furious films. If I'm being honest I think they should have stopped at 2 maybe 3 but 8 (with plans for more) seems excessive. I've made a pretty simple bar chart showing Adjusted gross revenue in chronological order of film release date. I've used colour to highlight those films that have exceeded the revenue of the original film and have used a reference line to help with the encoding. As I said nothing very fancy but I did download a similar font to the one used on the film posters and chosen the colour of the original car in the film. I did consider sorting by descending revenue but thought the film order possibly helped with the story as people would ask what order they were released.

Amy C.

Bar charts don't have to be boring! They can be playful and maybe even spooky. These use Halloween data and bring in other elements to make them more visually interesting, while keeping data integrity. Twitter

Andy S.

In BYU’s management communication program we teach students to write, design, and speak. We emphasize the techniques of storytelling with data when it comes to visualizations. This visualization, built primarily in Microsoft PowerPoint, explains a pattern I discovered in some market research while preparing a presentation on disruptive innovations.



I have created a simple bar chart depicting the fastest growing private industries and the revenue made by them. Tool used by me is Tableau.



For this challenge I decided to visualise the fatal effect of the recent “Beast from the East” cold wave in different European countries, focusing on Poland and the UK where almost half of its victims were based. You can view an interactive version on Tableau Public and reach out to me on Twitter for comments or feedback.



This infographic chronicles the history of college football games between the University of Miami and Florida State University. Three interrelated bar charts tell the story of this rivalry. I built the visual in Tableau, and the interactive version is on Tableau Public.


Here’s a big bar chart I did several years ago for the Winston-Salem Journal about health-care plans. It’s a bit of a different presentation from a “typical” bar chart, since the bars are stacked and their totals are the same. I felt that was the best way to show the shift in plan usage over time—since there are so many bars, it would’ve been unwieldy and much less clear if all the plans for each year hadn’t been stacked. (I never like graphing when there are years of data missing, because it can create false impressions, but in this case we felt the data was valuable. I spaced those bars out more than the others and added the dotted lines between years to indicate the separation. Still not sure that was sufficient or the best thing to do.)


Charles S.

For this month's entry, I have produced the following bar chart from CDC data. This was made with Power BI, and underlines how physical activity and the proportion of students classified as overweight are going in opposite directions for grades 9 to 12 students over the period covered. Although we cannot speak of (negative) correlation, I find that the bars make it easy to underline the trends for the dataset used.


Charlie H.

I was watching some Premier League football in the UK here, saw a guy score and knew he was one of the top scorers in Europe and decided to research and chart it. The colour is a reference to Salah’s team (whose team wear red), with all other teams pushed into the background with grey.



Colin W.

Here's a pilot viz for part of a collaborative project with my wife Ann, who is a keen small garden blogger. Her blog ‘Green City Gardens’ focuses on gardening in an urban setting, and particularly the role domestic gardeners can play in collectively improving the urban environment both for themselves and the wildlife with which we share our planet. We recorded how many birds visited our garden over a 10-day period in February 2018 including which species. The data visualisation aims to add insight to the title, use colour effectively as well as practice in bar labeling for a less cluttered look. I also practised the storyboard approach, which will inform the final version of the viz for the blog post. The Interactive version is available on Tableau Public.


My analysis is the positive and negative effect data for three countries of personal interest (Belgium, Holland, and USA). All data from recently published world happiness report.


Daniel C.

I looked at the top 50 best-selling video games and how Nintendo console exclusives dominate the top 50. I added indicators to show where games have been bundled with consoles which can boost their sales. I also added a highlight action to show multi-platform games that were playable on Nintendo consoles. Data is sourced from Wikipedia. Viz on Tableau Public | Twitter


Daniel L.

I used Tableau Public to create the top ten FC Barcelona goal scorers.


Dave A.

With the Academy Awards and Times Up this past weekend, I was thinking about women in film, and wondering what data was out there. Luckily I found on this great set on Bechdel test, which really asks "Is a female character a character in her own right, or just an extension of a male character?" 


Dave N.

This bar chart was a quick and fun visual narrative of my kiddo's (unfortunately) one-and-done playoff hockey round. It shows how the playoff game very closely mirrored their overall seasonal results—essentially matching how they and their opponents played in all but the penalty categories (which ended up being the reason they lost the game). For creating the full dataviz, I gave myself a time limit, from pulling the data from our league's website to the final image creation, of no longer than the actual game time (which was 1:15 - an hour and 15 minutes). I did it in 1:14. : ) Twitter


David O.

Back in 2007, more folks lived in cities than in the countryside, so I tried to tell that story with some diverging bar charts. Made with Tableau, data from the World Bank. Tableau public link.


David S.

For my basic bars I decided to look at Olympic medal count, using colour to highlight gymnasts and swimmers, who dominate the list. I created this in Tableau and used the viz in tooltips functionality to add an additional level of detail. 



This is my second version of visualizing my annual tax summary showing how my tax and national insurance contribution payments were spent in the public sector. Here I see that 75% of my Tax % NICs were spent in the top 5 public sector areas: welfare; health, state pensions, education and UK's national debt interest. Read more about how I was inspired to create this visualization in Tableau in my recent blog post.



I thought I would share my latest viz that I have done for the UK health TUG around diversity as I really enjoyed creating a slightly different type of bar chart (diverging). This viz looks at some of our health care data comparing population figures to those seen by our local District Nursing teams in one of our boroughs in the UK. The district nurses provide an essential home based service to patients that aren’t able to leave their homes to attend clinics. The district nurses are amazing, and will go out in all weathers to ensure they see their patients!


Last year I read the fascinating book Stalin's Daughter: The Extraordinary and Tumultuous Life of Svetlana Alliluyeva by Rosemary Sullivan. After Stalin's death, his daughter defected to the US. It got me wondering about other Soviet defectors during the cold war. Being a huge fan of ballet, I felt like most of the defectors I had heard of were ballet dancers. I was curious about the breakdown by profession. This chart is more exploratory in nature and was made with Tableau.



This isn’t “simple” but I couldn’t resist.  I built this a few years ago in Excel 2010 for a project and the concept has proven useful. I updated the dates so the “today” vertical line shows up. Briefly, this displays the server environments and what happens in each of them during an ERP upgrade.  Some environments were used for multiple phases (e.g. TMV 1 and TMV 3). We didn’t have enough hardware to keep all environments available all the time so we had to take them up and down and schedule the teams’ work accordingly.  


I pulled in causes for mortality, from 1999 thru 2015, on I used Tableau to create this bar chart with a purpose to bring forth the comparison between the various causes of death and a hope that it educates people and helps align resources in fighting the most critical causes.
Twitter: @hemalsanghvi



Here's a historical look at federal surplus and deficit trends over time.


Jamie B.

Tried to make it as simple and self-explanatory as possible.

Jason P.

Here is my submission for the #SWDchallenge in March. The timing couldn’t have been better since I was already inspired by the Winter Olympics and I was analyzing historical medal count data. I thought I would focus on some outlier nations that are extremely specialized in only one sport. The stacked bar chart is based on volume (medals in specialized sport and medals in other sports). Percentage is also a key factor in the story, but I chose not to visualize that information because by definition, these are the only nations with at least 90% of their medals in one sport. Visualizing percentage could be useful information, but the trade-off might be that I have too much information. These 5 nations are anomalies in terms of their extreme allocation of medals in only one sport. Hopefully, I have done these unique and talented nations justice by visualizing just how specialized they are. You can find me on twitter at @jaydpauley and on Medium.


As the deadline for this challenge is midnight just before the beginning of International Women’s Day, I wanted to take a look at some gender-related data. I turned to the just-launched Open Think Tank Directory for a dataset on think tanks. The chart shows the share of think tanks in each region that are led by women, men and, in a few small cases, when they’re led by more than one person with a mix of genders. At one level, it makes sense to give the breakdown as a percentage of the total — but I also wanted to give a bit of context to these numbers, as our dataset is substantial but not complete. The chart on the right shows the number of think tanks included in each region and how many we have information about the gender of their leader. Who ever said you can’t visualize missing data?


Jeff P.

As a huge sports fan from Minnesota, I've been wanting to tackle this viz for quite some time to see just how bad our men's professional sports teams have been. My goal was to look at the four major professional sports in the U.S. and for each sport, sort the teams from most to least number of seasons played without a championship. I was certain the Minnesota teams would be near the top in all sports and had a hunch Washington D.C. teams would be up there too. However, I didn't realize just how close their droughts would be in total. I used color to call out just the teams belonging to these two fan bases and included maps that can be hovered on for more detail.
Blog | Twitter: @JtothaVizzo


Jeremy and Sarah

To pick a topical subject, we’ve looked at steel in the United States. US President Donald Trump is justifying the proposed 25% and 10% tariffs on US steel and aluminium imports on national security grounds, avoiding World Trade Organisation scrutiny as the multilateral trade body does not have the powers to rule against measures imposed on ‘national security’ grounds. However, our graphic shows that the steel and aluminium sectors are not as nationally important as Trump believes. We’ve tackled this as a bar (well, column) chart looking at the steel industry’s decreasing share of total US GDP and jobs since 1993. As this is a fairly simple time series, we’ve gone for eye-candy to engage the reader, creating the bars from a photo of a steelworks. We’ve shown the workforce share as a line on a second Y axis; we tend to tackle this by making sure the line, Y axis labels and legend use the same colour, with legends for left and right axes to the appropriate sides of the charts. The chart was first constructed in Excel, and then taken into Illustrator so we could do most of the finessing.


Jim S.

For this month's challenge, I looked at the ratio of rushing to passing yards in the NFL since 1970.  Even with the trend falling steadily, the last 5 seasons have been historically low, built using Tableau.  Workbook and data can be found here.

Jim VS.png

Josh S.

I love this data set because it makes big numbers easier to think about. As always, I started by discovering the story: figuring out who my audience is (the heroes), the big takeaway (the theme), how all the evidence supported the takeaway (the plot), and how the audience would consume the information (the setting). Then I worked on the narrative that would tell the story: how do I place my audience, as heroes, into the data story? How do I make the analysis relevant to them and invite them to learn? What visualization best illustrates the plot (spoiler alert: bar chart!)? And, of course, how do I flesh out the plot with the things that make the story worth experiencing: color, annotations, and other design elements. Once I had a general wireframe drawn, I built the visualization in Tableau. 

Author: Joshua Dean Smith (find me on LinkedIn or on Tableau Public!)


Justin D.

When I was a kid, I loved watching Wayne Gretzky at the tail end of his career. Recently, I’ve been getting back into the NHL, so I decided to visualize the Gretzky's dominance in his prime for the #SWDchallenge | Tableau Public Link



The bars may be basic but this chart went through a few iterations as my story kept changing! Data viz created using Adobe Illustrator.

Kate B.

I used Tableau public to show the 3 hardest holes at the 2017 Masters golf tournament. I defined the hardest as the holes that had the highest percent of scores over par. I sorted the data by hole number instead of by the highest percent because as a golfer it seemed more logical to me to see them in order of the holes.



I have made a bar chart to show the acceptance and the enrollment rates for universities in the US. The tool used to build the chart was Tableau. The data was only county wise. To get the state information, I used the zip code and mapped a code to the state. Also, the acceptance and enrollment rates were calculated as follows: Acceptance rate = total admits/total applicants and enrollment rate = total enrolled/total admits


Created with Tableau | Tableau Public | Twitter


For March's #SWDChallenge, I have explored my 16 months of daily walking data collected from Moves App. The bar graphs visualize average and median number of steps I have taken for each day of the week. It is clearly highlighted that my weekends are comparatively more active than weekdays. Also since I am not much into outdoor activities, there are not many outliers in the data, because the average and the median are fairly close to each other.

LinkedIn: Public:



I'm as big a fan of getting down to basics with bar charts, so I welcomed this challenge. But in data storytelling, I sometimes find that a single-measure bar chart is an incomplete story unless it is a precursor to additional context.  So, I used one of my favorite visualization techniques which I learned in Stephen Few's "Now You See It" which is called a table lens, or side-by-side bar chart. It allowed me to compare categories across two measures. In this example, I created a build that started with a possibly premature conclusion with only viewing one measure, and completed it with more context around where the needle may be better moved. I also used a standout orange hue to call out an area of attention. Twitter | LinkedIn



I used the Excel tool to create my graphic. I like use this tool because is very accessible to all. I love analysis the Brazilian economic and Pharmaceutical Market and create insights of easy understanding for all. LinkedIn | Facebook | Twitter | Blog

Liyang W.

I ended up using my personal data for the chart, very simple.



I went to an exhibition and wanted to show how many Lego bricks were in each model. By using the bar chart it allows the user to see the difference in number of bricks. You can then click onto the exhibit and see the description of it.  Interactive viz 




My submission is a look at Brazil's Population Estimate for 2017-2018. I only looked at most and less population cities, trying to emphasis the biggest and lowest ones. The data came from IBGE - Brazilian Statistical Office and I use Tableau for DataViz, because I'm trying to learn more about this tool. Interactive Viz




While Women’s Day is all about celebrating women’s achievements, it is also important to reflect on what holds them back. Gender violence and domestic abuse predominantly affects women. I look at some stats on it from the UK. More in the post here


Nathan H.

I scraped beer data from, filtered on U.S. breweries, aggregated by style, and calculated the percent of beers that fell into each style. To my delight IPAs and double IPAs came out in the top two spots. I used R to make the chart.



A simple bar chart visualizing the most commonly used words (excepting stop-words) in my blog: “Data" and “visualization" feature most heavily!  Twitter: @theneilrichards



My submission is about the ratio of men to women who committed suicide in 2015. This ratio has been the same in during the last decade and is about the same across Latin America. I did it using Python's Matplotlib and Seaborn. Twitter

Nick K.

I wanted to find some data that was interesting that could show how using some color would revel changes. I was surprised at Iran's data so when I saw that outlier along with Morocco's so I thought it would something that other people would pick up on easier like I did with the color. 


Nick Z.

My “readers” are often internal stakeholders so when I annotate charts to call attention to specific data I often take that opportunity to pose questions to them. I find this helps guide discussions and keeps them more interested in reading (and responding to) what I give them. I made this chart in Excel, sorted my data in descending order for more clarity, formatted the x axis to 100% (it defaulted to 70%), and added two text boxes, one for the title and one for the annotation.



I focused on the gender gap in tech companies and how it is time for a big change. It works well in conjunction with Women International Day tomorrow. Last year marked the first year since 2006 that the percentage of women working in the technology field declined, a troubling reversal for an industry that had been making progress in closing the gender gap and industry leaders should work urgently to revive progress toward parity. Link | Twitter | Interactive Viz 


Paul M.

In the 2016 Alabama Senate special election, Doug Jones flipped 12 countries won by Donald Trump in the 2016 Presidential election from red to blue. Bars were not my instinct when I first looked at this  data. However, this seemed the perfect opportunity to challenge myself to keep my chart choices simple. For me, the story is in the drastic change over a relatively short period of time. For that reason, I focused on the counties that flipped from Republican to Democrat.
@RelatableData | Paul's Tableau Public Profile

Paul W.

I reviewed the data on Super Bowl Coin Flips to see how often the winner of the coin flip goes on to win the game. While I created a simple bar chart using Tableau, I didn't use the bars to aggregate the data, as is the typical use case for a bar chart. I looked at doing it that way first, but it only resulted in two bars—one for heads and one for tails, which wasn't too exciting or interesting. Instead, I used the bars as a binary indicator, either heads or tails for each Super Bowl, and then color coded the bar to indicate whether the Coin Toss Winner won or lost the game. So I ended up putting a twist on the regular old bar chart but still kept it simple.
Interactive Viz  | Twitter: @PaulWachtlerPMP



I'm visualizing how policymakers estimated the share of female labour participation in my home country Colombia. Interactive Viz

Paula J.

KDnuggets conducts yearly polls on what tools data scientists use and this is how the tools rank. I am glad you were curious about the previous year results because when I looked for that data I decided to use 2016  poll numbers which seemed to be different from the summary columns in the 2017 table indicated. Lesson learned—always check your data!


For this month’s challenge I decided to take a look at our blog stats. We were founded in April 2016 and are coming up on our 2 year anniversary so thought it might be awesome to analyze our stats via your challenge. Our new project #DuoDare has received great attention and Adam (my co-blogger) blew up our November stats by creating a stunning visualization on UFO sightings. My entry goes to show that color, annotation and bars when used effectively can help deliver your message at a single glance! Hope you enjoyed my entry. Blog | Twitter | LinkedIn


You don't have to be a professional developer to create charts with D3. This is the first in a series of post walking you through the process. Here you'll learn how to build a simple bar chart. Blog | Code | Chart




Following your guidelines, I went with a Bar Chart on the Top 20 Highest Grossing Domestic Box Office Movies (adjusted for inflation), and highlighted the Star Wars Franchise dominance. Interactive viz


Sarah B.

Given that the brief was to use a simple bar chart, I opted for a simple dataset. I recently read an article by Forbes which listed the top 5 wealthiest hip-hop artists and their overall net worth. I thought this would be ideal for the challenge as it’s not a subject-matter that’s commonly visualized. I included the net worth of each artist listed for both 2018 and 2017, opting to use bar-in-bar charts to show the growth in net worth over the last 12 months. Interestingly, this is the first year in the history of the annual list that Diddy hasn’t placed first. I included some text to the right of my visualization to help explain how each artist had made their millions too. Ironically none of the artists listed have become super rich from their music alone. They have all branched out into other areas such as alcoholic beverages, music streaming services or clothing. I wanted to draw attention to Jay-Z since he topped the list so I included his bar and the headers of any associated text relating to him in red to make it stand out, keeping everything else in black and grey. You can interact with my visualization on Tableau Public.  



For March Basic bar chart challenge, I looked at what the top 25 profitable companies made in 2016. From the chart, we can see Apple leads the game with $45.7B net income making $1,444 per second in 2016. That’s 1,0 17,219 times the *median earnings of an American worker per second.
*Median weekly earnings of an American worker $859


Sean M.

I’m a semi-professional appreciator of music and when you pair that with my knack for data analysis and visualization, you end up scraping every single Billboard Hot 100 chart going back to 1958 to look for trends. Well, in my initial exploration, I was curious what the distribution was for amount of weeks spent on the chart. So I built a histogram depicting just that. And the result intrigued me! The MAJORITY of songs to ever chart on the Hot 100 have spent exactly 20 weeks. As a data analyst, I knew this was the point of entry into further analysis. By breaking it down further by number songs each year to spend exactly 20 weeks on the chart, I again saw a massive jump in 1991. This led me to do some research and come to find out there is a little known rule in the Hot 100 criteria. “Any song to spend 20 weeks on the chart AND not rise above the 50th spot will be designated as “recurrent” and removed from the list.” Billboard implemented this rule as a way to keep having fresh new songs added to the list.



My viz attempts to test the assumption ‘drive for show,putt for dough’ which suggests the more successful US PGA golfers are also the ones with better putting statistics. I approached this by creating 2 diverging bars,one depicting the top 50 driving performance golfers in 2017 and the other the top 50 putters. I then visually coloured the bars relating to players in the top 50 prize money list of 2017 using the same colour as the word ‘dough’ in the title (to achieve an association). In addition I only provided the names relating to the bars of a top 50 earner as these were the golfers pertinent to the analysis. This allowed me to visually demonstrate more of the top 50 earners were in the top 50 putters,this crudely confirming the assumption. BlogInteractive Viz


Simon R.

Inspired by my recent attendance at the storytelling with data workshop in London, I decided to set up a Tableau Public profile and give the next challenge a go.  My visualization shows day by day breakdown of cycle hires in London but I wanted to focus on the reasons for some of the larger entries using colour to do so. 


According to the Singapore’s Prime Minister speech on National Rally Day** in August 2017, Lee Hsien Loong highlighted 3 main focus areas for the nation long-term development. They are (1) pre-schooling, (2) diabetes fighting and (3) smart living. Obviously, education comes the first in priority. Unsurprisingly, with the vision in bringing Singapore towards smart nation, science is the foundation driving the aforementioned long-term vision. Despite the already strong human capability of Singaporean, science and technology are being focused at all levels of education. In particular, Physics is apparently the one given the highest emphasis, in term of number of programs generated comparing to other topics in science. Although the data shown represents science learning labs and classrooms beyond school classes, this guides how science is being focused in Singapore. Singapore is recognized about its success in national development. I thought that every country can learn from Singapore from any particular aspect. Although I am interested in education as it is a foundation for sustainable development at all levels. So I would like to present a piece of aspect that might inspire my own (and maybe others al well) life-long learning. Data from by the Government of Singapore and created with Tableau Public. LinkedIn



Here is my very simple attempt to use bar graphs for showing the tallest 25 buildings in the NYC with their purpose and building material used. I have kept it as simple as possible focusing on bars as a medium to show the height. I have purposely not sorted the bars so as to give it a look of a skyline. 


Shane W.

At our beginning of the year professional development day for staff, our superintendent wanted to communicate the gains we have made over the last decade in UC/CSU “A-G” completion rate (it’s a measure in California of how many high school graduates complete a set of classes from different curricular areas). We used to have a table of values for each year, with yellow highlights for our schools. I wanted to create something more visually striking, illustrating that our schools have moved from down here, to up there.

I originally used the school colors of our schools. But there were two issues with this. First, the point wasn’t individual schools but the district as a whole. Different colors separated them rather than unified them as our schools. Second, the different colors for each school were distracting, too many colors going on. So we simplified it to the district’s particular shade of blue. Also, why the black background? Change this viz to a white background and the brightest part of the viz is now the gaps between the bars. The gaps become more prominent than the bars. With fewer bars this isn’t as much of an issue, but in this case, with this number of bars, the black background is a better choice. (Why so many bars? That’s what my boss wanted, so that’s what we do and it’s my job to figure out how to do it.)

Data came from California’s school data system, DataQuest (this is a good website for a lot of school data in California). The viz was created in Excel. @shanewaggoner#RJUHSD



I used Google Data Studio to create this bar chart. The bar chart visualizes website traffic and conversions by marketing channel. Blog

Stephen R.

The chart depicts the detected and undetected crimes in South West Dublin as reported by the Gardaí (Irish police). The graphic is used for emphasis.
Source: John Lahart, TD (Member of the Irish Parliament), Created in: Tableau + Paint.NET




I've picked the topic of Super Bowl 51 and using a column chart have depicted how the New England Patriots overcame a huge deficit to stun the Atlanta Falcons in overtime. This is a column chart with zero gap width between each data point, hence at first glance it provides an illusion of an area chart. The chart tells the story of how Atlanta held a lead for the major portion of the first 3 quarters only to slowly lose out in the 4th quarter, eventually losing in overtime 34-28.



For this month challenge on Basic bars, I used the MakeoverMonday archives data. Twitter: @takazi88


Teresa B.

The data consists of a sparse time series of number of incidents/delays, which makes it a great candidate for a bar graph. In the visualization I wanted to highlight the fact that over the course of four months there has been many incidents and delays that have affected many bay area commuters. LinkedIn | Tableau Public



Multiple charts on climate change and you can see the remainder of them on Twitter


My goal was to show that in a first past the post election (like the one in the United Kingdom), the percentage of votes a party receives does not reflect on the number the seats a party wins. The bars were colored according to the traditional color scheme for the political parties.


Thomas O.

On January 30th, we founded the Linz.AI data science group. The bar charts shows the growth of group members until the first meetup 35 days later. I used R (ggplot2) for the visualization. The data consists of all attending guests and does not show the 25 members who registered for the group but didn't visit the first meetup. Twitter | Link




In light of International Women's Day on March 8th, I decided to pick a topic close to me - the proportion of female engineering students in Canada. As an industrial engineering graduate myself, with several female relatives graduating in the 80's and 90's, we always spoke about the low representation of women in this field, and how ratios haven't seemed to change over time. I wanted to explore the proportions across the various disciplines, and if/where there has been any improvement. 
Interactive Viz | LinkedIn | Twitter: @valeriemais027



I have taken the statistics of the top 20 happy countries. I wanted to check what all contribute to the overall happiness of a country. One important observation - not all countries on the list are generous.


This is my very first submission for the SWDchallenge, created with Tableau. It shows the cost assessment of antibiotics used to treat bacterial infection, and the rate of resistance against the selected antibiotics, sampled globally between 2012 to 2017.


Thanks again to everyone who took the time to create and share their work! Stay tuned for the next #SWDchallenge, which will run the first week in April.


SEARCH STORYTELLING WITH DATA: © 2010-2018 Cole Nussbaumer Knaflic. All rights reserved. STORYTELLING WITH DATA and the STORYTELLING WITH DATA logo are trademarks of Cole Nussbaumer Knaflic.

“how do I incorporate visual design into our monthly deck?”

After reading storytelling with data or participating in a workshop, people often ask how they can incorporate the lessons into a recurring (i.e. monthly, quarterly) report. These reports often materialize as a PowerPoint deck, which started sparsely, but over time has taken on a life of its own and now resembles the “slideument”: part presentation, part document but not exactly either at its best.

Consider the slide below, which is based on an actual slide from a recent client workshop. (I’ve anonymized the client’s data to preserve confidentiality.) Today’s post demonstrates how to apply data storytelling lessons to a visual from a monthly deck, illustrating the thought process to improve it.


This slide shows a monthly trend of customer service complaints: in total (top chart) and broken down by category (bottom chart). The commentary section tells us (the audience) what the important points of reference are: what happened this month compared to last month (complaints are up 14%), where it changed (Employees) and their proposed next steps. However, notice how much work takes to read through all this text and then find evidence of this in the graphs.

Imagine if you were given this slide to determine an action plan. If you were in a live meeting, would you be able to read all of this text and listen to the presenter at the same time? If you weren’t in the meeting and were reading through the deck, how much time would you realistically spend trying to digest the information presented? We can improve on this visual in both scenarios with a few design changes.

In both cases, I used the commentary as a guidepost for the important takeaways and re-designed the visuals accordingly.

First, let’s a closer look at the top chart. The commentary tells us that complaints were up 14% vs the prior month.


Where did your eyes go first in this graph? Mine went to the red Average line, which I visually estimated to be about 410 per month.  In looking for evidence of the 14% increase in December, I had to do a lot of mental math (add the Solicited + Unsolicited for November and compare it to Solicited + Unsolicited for December) which took more time than someone would likely spend doing this.

If that 14% increase is what the audience should know, check out the difference between the original visual and this:


When applying the “where are your eyes drawn?” test, my eye went straight to the data markers & labels at the end of the total line, where I could see both the absolute numbers and annotations telling me it’s a 14% increase. Since we’re visualizing time, I changed the graph type from a bar chart to a line chart, unstacked the data series, and added a series for the total. This was intentional based on the commentary, which only referenced the total trend. I chose to de-emphasize the subcomponent pieces (Unsolicited and Solicited) by using grey.

Side note: what about the Average line? If the monthly deviation from average was really important, one option would be to keep it in the graph for reference with the tradeoff that adding a fourth data series could create clutter. Another option is an entirely different choice of visual, depicting the monthly change (from average), with a visual cue to indicate that December’s data is acceptable. Both are choices the information designer would make knowing the audience and what context is relevant. In this case, I didn’t feel that this additional point added anything to the overarching story, so I chose to eliminate it altogether.

Let’s take another look at the second visual now. The commentary tells us that complaints were up in a specific category: Employees. Not only did they increase, but they increased from 87 to 117. Apply the “where are your eyes drawn?” test again with the original visual.


If I took an informal poll of readers here, some might have gone to the black line, others might have noticed the blue list first and others (like me) went to the red line. Regardless of which line you focused on first, I’d likely bet that you didn’t focus first on the November to December increase in the Operations line (red).  In fact, it’s difficult to discern the absolute numbers (87 and 117) here because of the general clutter: overlapping data series, gridlines, color, heavy chart border and legend at the bottom requiring some visual work to figure out which line goes with which complaint category.

When setting out to improve a visual, there’s not necessarily a right or wrong answer in choosing a visual type: it often takes looking at the same data several different ways to find which view is going to create that magical “lightbulb” moment. Let's look at a few different variations of this visual.  

First let’s keep the existing line chart, remove some of the clutter and focus attention on the November to December change in Employees.


This view gives the audience the full context of the 12 month trend, while focusing attention strategically on a specific point. However, if the emphasis is really about the November to December change, we could also visualize only those two data points. Let’s look at a few different ways of displaying this.

First, this horizontal bar chart compares this month (December) to last month (November). Horizontal bar charts are useful when your category names are long and therefore can be displayed horizontally from left to right on the y-axis without having to rotate or shorten them.


Another option is a vertical bar chart, if you’re more inclined to preserve the left-to-right construct of displaying time.


As a third option, we could use a slopegraph. Slopegraphs can work well in making change visually apparent across categories. Check out how clear it is that some of these categories changed more drastically than others. In fact, looking at the data this way, we see that there was also a marked increase in service-related complaints, something that didn't stand out as much in the other views of the data. You can read more about slopegraphs, including design considerations, in this previous post.


Any of these three visuals could work for depicting this data, I chose the slopegraph for the final version to keep the emphasis on the change in the two data points.

Here's what it could look like if all of this needed to be on a single slide:


In the remade version, I’ve moved the text to be closer to the data it describes and used color strategically to create a visual link between the text and where to look in the graphs for evidence. I’ve also made the call to action more visible—remember when communicating with data for explanatory purposes, we should always want our audience to do something with the data we’re showing them!

Check out the difference between the original and the remade version:


This single view works well as a remake of the original, but not as well in a live presentation. There’s still too much text to read and process, while listening to a presenter at the same time. For a live setting we can still use the same visuals, but build piece by piece (using animation), which forces the audience to listen to the presenter describing the data. For example, consider the Complaints over time visual again:


Now imagine if each of these images were its own slide. Sparse slides lead to better presentations because a person is there to narrate what’s happening.


One final note on the choice of red as the emphasis color. Some readers may be surprised to see something different from our usual blue & orange as emphasis colors (and readers who are Michigan fans are probably having heart palpitations!). In this case, red was the client’s brand color so we chose to stay consistent with the rest of their visuals. If that weren’t the case, we might avoid red because it could a negative connotation, even though this is a somewhat positive story (complaints declining over time).  

In conclusion, we can indeed incorporate visual cues such as strategic use of color and words into a monthly recurring presentation so that our audience clearly knows 1) what’s important and 2) what action to take.  You can download the Excel file with accompanying visuals here

Elizabeth Ricks is a Data Visualization Designer on the Storytelling with Data team. She has a passion for helping her audience understand the "so-what?" Connect with Elizabeth on LinkedIn or Twitter .  


SEARCH STORYTELLING WITH DATA: © 2010-2018 Cole Nussbaumer Knaflic. All rights reserved. STORYTELLING WITH DATA and the STORYTELLING WITH DATA logo are trademarks of Cole Nussbaumer Knaflic.

#SWDchallenge: basic bars

If I could only use a single graph forever forward, it would be a bar chart. It's sort of like that exercise: if you were stranded on a deserted island and could only bring a single book to read or only have one type of food to eat from then on out—what would it be? (Easy: The Great Gatsby and peanut butter toast). Is it a realistic scenario? No (I'd have my Kindle with me and likely not an endless supply of PB and bread...or electricity...or a toaster). Would I get sick of my choices? Probably. Would it work all of the time? No. But I really do have an affinity for stories about the Jazz Age and peanut butter toast.

I also really like bar charts.

Bars are my go-to graph for a number of reasons. They are common. While this might be cause for some to avoid, this is one of my top reasons for embracing: your audience already knows how to read a bar chart, so you don't face a learning curve for getting your information across. They are not intimidating—you aren't likely to scare anyone with a bar chart. Bar charts are also easy for us to read. When we look at a bar chart, our eyes compare the ends of the bars relative to each other and relative to the axis. Because of the alignment to a consistent baseline, it's easy to see which category is the largest, which is the smallest, and also the incremental difference between categories. Note that for the visual comparison to be accurate, bar charts must have a zero baseline (read more). Bar charts can be vertical (also known as a column chart) or horizontal (great use case: if category names are long—allows you to orient text in a legible fashion, avoiding slow-to-read diagonal text that would be needed if you stick with vertical orientation). Below is an example of each from storytelling with data:

Basic Bars - Vertical.png
Basic Bars - Horizontal.png

The #SWDchallenge this month is to create a basic bar chart. Nothing fancy. No need to stack it or do anything else crazy. Have you made a bar chart before? Probably. But here is an opportunity to focus on making your best bar chart yet. Find some data and teach us all something new. While not a strict requirement, I do encourage you to look to the prior challenges when it comes to annotating and thoughtful use of color and words (including the takeaway title!)—these bar charts are meant to inform, so those important lessons will apply here as well. DEADLINE: Wednesday, 3/7 by midnight PST. Specific submission details follow.


  • Make it. Identify your data and create your visual with the tool of your choice. If you need help finding data, check out this list of publicly available data sources. You're also welcome to use a real work example if you'd like, just please don't share anything confidential.
  • Share it. Email your entry to by the deadline. Attach your image as a .PNG. Put any commentary you’d like included in my follow up post in the body of the email (e.g. what tool you used, any notes on your methods or thought process you’d like to share); if there’s a social media profile or blog/site you’d like mentioned, please embed the links directly in your commentary (e.g. Blog | Twitter). If you’re going to write more than a paragraph or so, I encourage you to post it externally and provide a link or summary for inclusion here. Feel free to also share on social media at any point using #SWDchallenge.
  • The fine print. I reserve the right to post and potentially reuse examples shared.

I look forward to seeing your beautiful bars! Stay tuned for the recap post in the second half of March.


SEARCH STORYTELLING WITH DATA: © 2010-2018 Cole Nussbaumer Knaflic. All rights reserved. STORYTELLING WITH DATA and the STORYTELLING WITH DATA logo are trademarks of Cole Nussbaumer Knaflic.

connecting... the dots

Today's post is a guest post written by Daniel Zvinca. When Dan first reached out to me via email about a blog post I'd written, I thought to myself, That name sounds familiar... Why do I know that name? Some time later, it struck me—it was because I'd recently read one of Stephen Few's quarterly updates that introduced the Zvinca plot named after—you guessed it—Daniel Zvinca.

Dan has a mechanical engineering background, spent much of his career developing business related data applications in the ERP area and beyond, and today enjoys, among other things, considering and practicing data sense-making and data visualization. He lives in Romania (Romanian is his first language), but most of his projects were implemented in Belgium, where he runs a small IT company together with his business partner, Wilfried Van den Bosch.

Dan and I have had some good conversations in blog comments as well as behind the scenes (I'll use this as a opportunity to mention how impressed I am at the technical conversations we have with him speaking in a non-native language). At one point, we were discussing lines and all the ways they can be used. I've run into the scenario several times recently in workshops where it's become clear that many people are under the false impression that lines can only be used for continuous data. That's actually not true. The guideline is that with graphs that use lines, you have to make sure that those lines make sense. In some cases, that can be true for non-continuous data as well. At any rate, Dan and I were discussing this at one point and I invited him to pen a guest post outlining the many uses of lines in data visualization. The following is his post. I hope you'll find it eye opening how many different ways we can use lines in data visualization.

A good communication requires reasonable knowledge of a common language in order to succeed

I had the experience of learning English by starting with a minimum vocabulary and only a few grammar rules. At one point my job required me to travel in another country where English was commonly accepted for communication, even though none of my coworkers were born in a natively English-speaking country. After a while I thought I had a reasonable command of English, but all my confidence collapsed during a 2-hour meeting in London. I could not understand half of what the participants were saying and I probably misunderstood the other half. I am still not sure if they used the cockney dialect or another language but, I am sure it felt awkward to me. Obviously, we can’t properly communicate if we don’t speak the same language, in some cases even the same dialect. The same rule applies so well to data visualization as an extension of our communication language.

The line, a powerful encoding graphical element

One of the basic entities we see in different forms of display is the line. Actually, we use the line term to describe what in geometry is called a line segment. Sometimes we use it to describe curved graphs as well. Just to make sure we are on the same page, in data visualization a line has a finite length, being delimited by two points in a two-dimensional space (paper, screen). Besides its ends coordinates, a line also features a geometric property called slope. This is defined by the rise over the run, the change of “Y” over the change in “X.” The most common data visualization form using lines to encode data is, obviously, the line chart, but there are many other graphs that use lines to encode information. It might be useful to see a few of them and the roles that lines have to encode information. Basically, in data visualization we use lines to:

  1. Encode end points position.
  2. Show connection between two points.
  3. Show orientation, direction or sense.
  4. Encode variation. A slope shows the change in vertical direction over the change in horizontal direction.
  5. Show pattern of change. A group of connected lines show pattern of change, possibly indicating trends.
  6. Define separation. A line can indicate the separation between two regions.

Although important, I deliberately skipped the use of lines for reference constructions like axis, ticks or gridlines. Before I dive into the subject, I would like to remind a couple of things. First, in data visualization we encode two types of variables: quantitative and categorical. Examples of quantitative variables: Cost, Price, Volume of Sales. Examples of categories: Country, Customer, Months of the year, Days of the week. I should also mention that some of the quantitative variables can be considered categories in certain contexts where they can be used for grouping or aggregation purposes. Another thing I want to remind is the nature of information assigned to variables useful to describe the axis of a graph. Psychologist Stanley Smith Stevens developed Scale of Measurement, a classification that is widely accepted in the Data Visualization world. According with this, there are four levels or scales of measurement:

  1. Nominal (items have no particular order and no quantitative meaning),
  2. Ordinal (items have an intrinsic order, but not necessarily a quantitative meaning),
  3. Interval (items have an intrinsic order and the same difference between consecutive values), and
  4. Ratio (items have an intrinsic order, same difference between consecutive values and have a zero as reference).

With these two pieces of information we may consider that most of the graphs are forms of display that encode one or more variables (quantitative and/or categorical) with the help of one or two scales of measurement (nominal, ordinal, interval, ratio). Let’s have a look at the different roles the lines can play across several forms of display.

Line graphs


The most common form of display that uses lines to encode values, is the line chart. While in most of the graphical tools a line chart is considered just an alternative to a bar chart, they are important differences between these two graphs beyond the variable types they share. A bar chart encodes the values of a quantitative variable across a categorical variable for comparison purposes. A line graph displays the variation of a quantitative variable across the items of a categorical variable. Its main purpose is to show the pattern of change across all the items of a categorical variable. Each end of a line encodes the value of the quantitative variable (Y) associated with an item of the categorical variable (X). 

In line graphs, the slopes encode the variation of the quantitative variable between two successive categorical items. For this to work the change in X direction has to make sense. In Data Visualization it is widely accepted that a line chart works fine with interval (and implicit ratio) scales, for which the difference between consecutive items of categorical scale has a quantitative sense.

For instance, a time series (fits into the definition of interval scale) works well with line charts. A chart showing the sales over the months of a year, the sequence March ($50M), April ($60M), May ($40M) can be interpreted as: Sales increased (in one month) from March to April by $10M, but then significantly decreased in May. Trying a similar exercise and use a line chart to encode the sales across the product categories ordered alphabetically, we might have Computers ($40M), Mobile Phones($70M), TV’s ($20M). The interpretation of the slope would be something like: Sales increased from Computers to Mobile Phones, but then significantly decreased for TV’s. It doesn’t make sense. We can repeat the same exercise and order the product categories in descending order of sales adding as prefix their rank: 1. Mobile Phones($70M), 2. Computers ($40M), 3. Televisions ($20M). This time we may read the graph as: If we look at figures from sales rank perspective, the variation from 1st category (Mobiles) to 2nd category (Computers) is as large as $30M, almost as much as the value of 2nd category... This time it works because each category is displayed in its rank position, so actually our categorical variable is the rank (1, 2, 3, …), an integer variable for which we can obviously have clear metrics defined.

In case we don’t remember the classification of S.S. Stevens mentioned above, we can consider that line graphs can be used only with those categorical variables that:

  1. Have an intrinsic order,
  2. The change (difference) between consecutive items makes sense, and
  3. All the changes between consecutive items have a similar meaning.

Please notice that I avoided the strict definition of interval scale that requires the same difference between consecutive items, in the favor of more general, similar meaning to make possible the inclusion of logarithmic and fractional scales. 

I made a short list of examples of categories that can or cannot be used with a line chart.

  • January, February, March (it works, intrinsic order, difference between any two consecutive items is 1 month)
  • January, February, September (it doesn’t work, intrinsic order is there, yet the difference between February and January is 1 month, but between September and February the difference is 7 months)
  • Monday, Tuesday, Wednesday (it works, intrinsic order, difference between any two consecutive items is 1 day)
  • 1, 2, 3 (it works)
  • 2, 1, 3 (it doesn’t work, not ordered)
  • 1, 2, 4 (it does not work, order exists, but the difference between consecutive elements is different)
  • 3, 2, 1 (it still works, descending order, the difference between consecutive elements is -1)
  • Apple, Oranges, Pineapples (doesn’t work, no intrinsic order)
  • 1st Oranges, 2nd Apples, 3rd Pineapples (it works, the rank is actually the categorical variable)
  • 1, 10, 100, 1000 (it works, logarithmic scale, but does not fit into S.S. Stevens' classification)
  • 1, 1/2, 1/3, 1/4 (it also works, fractional scale, but does not fit into S.S. Stevens' classification)



A slopegraph is a form of display that shows the variation of one quantitative variable over two categorical variables. The quantitative variable value is encoded by Y, first categorical variable is identified by the lines (S-Category, S from Slope), and second categorical variable is encoded by X (X-Category). Each end of a line encodes the quantitative variable value (Y) and X-Category while the line itself identifies the S-Category.

On his site, Edward Tufte writes “Slopegraphs compare changes usually over time for a list of nouns located on an ordinal or interval scale.” I agree, but I think that slopegraphs usage can be extended just fine to show the comparison between two groups of any categorical type, therefore they can belong to a nominal scale as well. Cole wrote this post about a slopegraph showing the comparison between groups.

When there are more than two elements of the X-Category, slopes comparison has to make sense across the entire graph, therefore the above mentioned line graph rules apply (intrinsic order, differences between any two consecutive items have sense and similar meaning). Or, if you prefer, follow Edward Tufte guidance, but you might consider also the cases that do not fit S.S. Stevens’ classification (for example, logarithmic, fractional, and cyclic scale).

Frequency Polygon


A frequency polygon is similar to a histogram. It displays the distribution of a quantitative variable over bins defined for the same quantitative variable. Instead of using bars to encode values it uses lines to connect the encoded values. A frequency polygon is the preferred form of display when we need to look for the distribution shape. I haven't figured out why it is called a polygon (this is the name used in geometry for closed two-dimensional figures), but I assume that polyline was not good enough. Each end of a line uses Y to encode the frequency value (the counter of the values that belongs to certain bin) and X to encode the bin position. The slope can be interpreted as the frequency variation between two consecutive bins.

Parallel Coordinates


Parallel Coordinates is a form of display that shows relations between multiple variables. Used in multivariate analyses, Parallel Coordinates usually works better with quantitative variables. Categorical variables also work, but in their case the lexicographic order is used to define the scale. Each line end uses Y to encode the value of one quantitative variable and X to identify the variable. Parallel Coordinates are used to discover relationships between variables. For this form of display the lines have just two roles: to encode the values and to connect correspondent values of two adjacent variables. Unlike Line graphs and Slopegraphs, the line angle has no meaning for this form of display. To remind the definition of a slope as the change of Y over the change in X, we cannot give any quantitative sense to X, other than a conventional position for variables axis.

Pareto Chart


Pareto charts are one of the few cases where it’s acceptable to use a secondary axis. This chart shows the values of a quantitative variable (encoded by bars) and their cumulative values (encoded by lines) calculated in the descending order of values. A correct design should have the two scales synchronized to make sense of dual data encoding/decoding in variable unit of measure and in percentages. The line uses Y to encode the quantitative value and X to encode the category. The slope of one line can be interpreted as the change of the cumulative value between two consecutive ranking positions.   

Connected Scatterplots


Connected scatterplots are scatter plots that have connection lines between the encoded X and Y positions given by a third ordered variable (very often time). The only role of the line is to show the order of the pairs. Sometimes the lines can be decorated with arrows to indicate the parsing order. There were many discussions over the years about the utility of connected plots. I participated in one of them on Stephen Few’s forum. I need to admit that since then I found a very useful particular type of connected scatterplot, that is often assimilated with a line graph, but is not. This is a design I made a few years ago in a forum as a respond to one participant question. Is the below line graph correct or not (the elections are not equally spaced in time)? The graph was designed to show the decline of the interest in politics by measuring % of participants from total possible electors for all elections organized between 1949 and 2009.  


My makeover is a correct and useful connected scatterplot, but it is not a line chart. The slopes of the lines connecting consecutive events indicate the participation change from one event to another and the distance in time between events can vary. This particular connected scatterplot has the third variable (connection order) the same with X variable (time).

Contour Plots


Contour plots are forms of display that encode 3 quantitative variables with a continuous variation, two of them encoded accurately by X and Y position and the third one (commonly Z, elevation) encoded by the variation of a color intensity. A sequential palette usually works best. The lines which shape the contours have the role of delimiting the bins of the third variable (Z-levels).

Tukey Bagplot


Tukey Bagplot is the two-dimensional generalization of a boxplot. Investigating the distributions of both variables with independent boxplots does not reveal anything about the simultaneous behavior of paired values. The dark gray area is called “the bag” (containing 50% of the points), the light gray area is called “the loop” (the other 50% of the points minus the outliers) and the outer polygon of the light gray area is called “the fence.” Without going into details, a Tukey Bagplot reveals similar metrics as a boxplot: location (the depth median, white cross), spread (bag size), correlation (bag orientation), skewness (the shape of the bag and the loop), and tails (the points near the boundary of the loop and the outliers in red). Even if the drawn polygons go through different points, they are just conventionally computed convex hulls, used to enclose sets of values. In this case, the lines have just a separator role.

This was by no means an exhaustive list, but it gives a good indication of the many roles lines can play in different forms of display. There are many other graphical forms that feature lines: regression lines, dendrograms, hierarchical trees, and the list goes on. Do you know of any other roles for lines in visual displays? What are your thoughts on this subject? Leave a comment!

Note from Cole: Dan, HUGE thanks for writing this post and teaching us all about the many different roles of lines in data visualization!


SEARCH STORYTELLING WITH DATA: © 2010-2018 Cole Nussbaumer Knaflic. All rights reserved. STORYTELLING WITH DATA and the STORYTELLING WITH DATA logo are trademarks of Cole Nussbaumer Knaflic.

#SWDchallenge: education, color, and words

Throughout my life, I’ve known February to be Black History month. Interestingly, though, that’s simply on account of my age, as this year marks only the 43rd year we celebrate and recognize African American achievements in the US and Canada—achievements that took place against a backdrop of inequalities and often injustices politically, economically, and socially. For me, I believe that one of the most important pillars to ensuring access and opportunity for all (as well as ending ignorance and racism) is education.

To raise awareness and celebrate Black History Month, storytelling with data is collaborating with, Tableau Public, #MakeoverMonday, Viz for Social Good, and Data for Democracy to ignite the imaginations and talents of our respective community members around the datasets and data stories connected to Black History. Each week’s focus is on a different sub-topic. I’ve decided to make this month’s #SWDchallenge to be centered on education, specifically the access, benefits, opportunities, and ignorance-curbing power. Create a visual with this in mind and let’s use data to recognize the importance—today perhaps more than ever before—of education in our society.

Your work doesn’t stop there. Last month, the challenge was to create an annotated line graph (nearly 90 people shared their creations!). I felt that singling out a graph type here would be too limiting, however (we’ll come back to that in future challenges). Rather than dictate a type of visual, this month we will put into practice a specific tip I find myself giving often when it comes to creating effective visual stories: be thoughtful in your use of color and words.

This may sound like simple advice. It is, I suppose, but there are nuances and the impact of these two straightforward elements executed well can be huge—and can even help overcome other design issues. Let’s talk a bit more about each of these.

Color, used sparingly, is one of your most strategic tools when it comes to the visual design of you data stories. Consider not using color to make a graph colorful, but rather as a visual cue to help direct your audience’s attention, signaling what is most important and indicating where to look. Note that for this to be effective, the use of color must be sparing. If we use too many colors, we lose the ability to create sufficient contrast to direct attention.

Words used well will both ensure your visual is accessible as well as indicate to your audience what you want them to understand in the data. There are some words that must be there: every graph needs a title and every axis needs a title (exceptions will be rare!). Don’t make your audience work or make assumptions to try to decipher what they are looking at. Beyond that, think about how you can use words to make the “so what?” of your visual clear. I advocate use of a “takeaway title”—meaning, if there is something important that you want your audience to know (there should be), put it in the title so they don’t miss it. Also, when your audience reads the takeaway in the title, they are primed to know what to look for in the data. When I’m putting a graph on a slide, I’ll use the slide title for the takeaway (and put a descriptive title on the graph). When the graph is on its own, I’ll often title with both—typically “descriptive title: takeaway.”

As illustration, below is an example. Here, I’ve shown the progression (no need to do this for your challenge, you can simply share the final product) from base graph, then added color, and finally words. Notice how we immediately know what to look for and where to look in the final graph.

Education color words.png

To recap the #SWDchallenge: find some data of interest related to education (you have free range within this: academia, higher education, black scholars, access, how education has helped ensure progress and opportunity, etc.). has curated a short list of datasets, or you can find even more in this list of publicly available data. Analyze the data to determine the specific story you’d like to tell. Harness the power of color and words to create your visual story. DEADLINE: Wednesday, 2/14 by noon PST. Specific submission details follow.


  • Make it. Identify your data and create your visual with the tool of your choice. If you need help finding data, check out this list of publicly available data sources.
  • Share it. Email your entry to by the deadline. Attach your image as a .PNG. Put any commentary you’d like included in my follow up post in the body of the email (e.g. what tool you used, any notes on your methods or thought process you’d like to share); if there’s a social media profile or blog/site you’d like mentioned, please embed the links directly in your commentary (e.g. Blog | Twitter). If you’re going to write more than a paragraph or so, I encourage you to post it externally and provide a link or summary for inclusion here. Feel free to also share on social media using #SWDchallenge and #VisualizeDiversity and/or upload to the page.
  • The fine print. I reserve the right to post and potentially reuse examples shared.

I look forward to seeing what you come up with. Thank you for helping to celebrate Black History Month and the importance of education in our society. Stay tuned for the recap post!


SEARCH STORYTELLING WITH DATA: © 2010-2018 Cole Nussbaumer Knaflic. All rights reserved. STORYTELLING WITH DATA and the STORYTELLING WITH DATA logo are trademarks of Cole Nussbaumer Knaflic.