Tag Archives: data analysis

Map your maps.

During the holidays season, I’m having more time to catch up watching movies. On that long list a film “Another round” can be found. In a nutshell, the plot is about four friends and their unexpected alcohol experiment. Everything is done in the spirit of science, of course. In truth, this dark comedy-drama touches on a very sensitive social problem that affects many people around the world.

I’m wondering how Poland looks compared to other European countries and if Poles on average drink more or less in comparison to Danes? According to WHO (World Health Organization) data from 2018 average Pole drinks 11.71 pure alcohol and Dane 10.26 (15+ years). The difference is 1.45. Is Poland near or far from Denmark? Depending on the colour palette and applied scale we can perceive it differently, and consequently, convey different stories or draw misleading conclusions.

5 stepped colour

I used Tableau Public to visualize data. This visualization is automatically chosen by Tableau. According to the visualization, Poles are not in the lead for European countries and Danes are somewhere in the middle of the scale.

3 stepped colour

But wait a minute. What a shame! Poles are heavy drinkers. Now I can see it clearly.

7 stepped colour

OMG… how much beer average Czech had to drink to win this competition? When it comes to Poland, it is not so bad. Poland is near the middle of the range.

Reversed 3 stepped colour

Hm… I’m a little bit confused. I have the impression that Poles don’t avoid occasions to celebrate the fragility of life, but now I can see is opposite. (Who would check legend description? Waist of time, data visualizations are intuitive!)

Attention: Remember in our culture stronger colour saturation means increased occurrence of the phenomenon.

As we can see, each of the four above examples depicts the same information differently, and that difference can be significant.

Maps are commonly used in public media and people like them. The same is in the business world. However, knowing it from experience, it is very easy to manipulate information presented on maps. Before you publish or share your map ask yourself:

  • Does scale represent the statistic bins,
  • Are colours adjusted to the topic,
  • Is reverse scale justified?

Data source: https://data.worldbank.org/indicator/SH.ALC.PCAP.MA.LI?view=map

Data storytelling for busy people – strategies which always work

Do you need to know how to tell stories with data?

Ask yourself how often do you use data in your daily job? Or maybe how many times do you use data to convince others to your ideas? If your answers range from rarely to often, then this post is for you.

One scene from the movie “Silver linings playbook” stuck in my memory. The main character after having an explosion caused by hearing his wedding song, is sitting in the therapist’s office, and complaining that it would not have happened if that song had not been played in the therapist’s office. The response of the therapist was clear and brief “You need to build your own strategy how not to be afraid of that song”.

Building strategies helps us to be more productive and perform better, whether it is in our work environment or our private life. Our brain just loves mental shortcuts, and strategies are those shortcuts.  Especially when we are in a hurry and need simple solutions which always work.

 Let’s see what strategies we can prepare to make data communication more effective and efficient.

Comparisons

Comparisons are always a good choice when we want to present the progress of initiatives, outcomes of introducing new processes, or showcase sales performance in different markets. People compare things in their brains all the time, so any story based on comparison will be easy to understand. But it needs to be well-crafted.

Before and After

This strategy works well when delivering outcomes of recently introduced new initiatives or processes. Old state data is the best background to emphasize big change or the success of a new approach. You can present benefits or results in several dimensions: process, employees’ satisfaction, increase in a number of clients etc. Anything you deem valuable for your business.

As an example, we can put together two dimensions: employee’s satisfaction and a number of human errors. In Picture 1, it is easy to see that changes have improved the employees’ satisfaction and resulted in a decrease in human errors. Simple column charts displayed side by side will suffice to represent this data. Adding lines connecting columns makes visualizations more suggestive.

Picture 1

Us vs. All

Every good manager should brag about her or his team and highlight what a great job they do for the organization. If your team, the product, sales, or converted leads are the best, show how they stand out from the rest of the company.

To draw attention to your data, you can change its colour. This simple trick will distinguish your data from the others and push it to the foreground. See Picture 2.

Picture 2

Where are my stars?

When analysing revenue growth, we consider what is pushing it forward and what is holding it back. A very popular concept is to present leaders and laggers. The popularity of this concept stems from human nature. We admire and envy the best, but love the worst because they are worse off than we are!

C-Levels managers like to see contributors of the growth on the waterfall chart because this visualization shows at a glance which contributors have made money for the company and which have lost. For our revenue growth example, we can use two different colours to indicate leaders and laggers.

Picture 3

Changes over time

Changes over time are the next group of strategies which use the familiar comparative idea with a whole story set in time.  We can present how something develops over time and what is more appealing for our audience how it might be in the future. For such stories, we use line charts.

Show me the bright future!

Who would not want to know the future? Well, I do not… But, when it comes to the business environment the answer is always: everyone. When I work with clients, the trend of any phenomenon is a must. Many decisions within an organization are based on current trends and an estimation of future outcomes.

However, every data scientist will warn against relying too much on historic data. There is a strong tendency to predict future business performance behaviour based on past results. To temper expectations, we can provide several scenarios based on the same dataset. This approach will add value to our analysis if we introduce factor parameters to each scenario. Typically, three scenarios are provided: optimistic, realistic, and pessimistic.

To illustrate the technique, I will use an example with revenue growth (every CEO cares about revenue growth). The main factor in the example could be the launch of product A in a new market. As we all know, launching a product on a new market can be a huge success, but on the other hand, it can also be a spectacular failure.  Using sales of product A as a parameter, we can create three separate revenue scenarios for the upcoming fiscal year.

Picture 4

Factors of success or failure.

Another story which is attractive for the audience is about factors which influenced the results of the phenomenon. This narration is based on our natural tendency to look for cause and effect relationships. Maybe if we knew what had triggered results in the past, we could use it in the future to prevent bad impact or use identified factors to achieve better outcomes?  This strategy is great when you want to convince senior managers to spend money on the next marketing campaign. Simply show them the periods with and without marketing campaigns on the line chart, where they can easily observe the ups and downs of the line representing sales. Do not forget to add some call outs to strengthen a message. See picture 5.

Picture 5

Connecting dots

The last strategy which I want to bring closer to you is about presenting the most crucial business metrics on the one-pager. This strategy is a master level, because whoever prepares it must be aware of connections between separate metrics and the overall influence which they have on the business health. This is very practical when trying to understand which processes drive others. The one-pager can show usual suspects, threats, and opportunities. For instance, if your core business as a company partner is selling services to the specific hardware, you can expect a drop in sales if hardware sales fall down.

Picture 6

Circle Charts – when design meets data

  • Circle charts are better to use for entertainment or information purpose. They are not the best choice for a business environment.
  • Circle charts are attractive for receivers and can pull them into your story.
  • Using multilayers demands providing a well-defined legend.

Humans always have had a special attitude to the sun. In prehistoric and ancient ages, in some cultures sun had the status of God. Without any scientific theories, they just knew that the sun is unique and has a crucial role for our planet and any living creature. Even in cultures where ancient humans did not worship the sun, the sun motif was commonly used to decorating buildings, everyday items, or apparel.

Nowadays, we still willingly use the image of the sun, especially in art and architecture. Something is appealing in this figure. Centric circle shape with rays around them somehow reminds me of the wheel of life with rays as special moments.

Maybe that is why the pie chart and all variations of pie charts are so popular and like among people. The father of the most known data visualizations is William Playfair. He invented a pie chart in 1801, and it is still commonly used to depict data.

My personal relationship with a pie chart is …. complicated. I do not use them often in a business environment. It is hard to present accurate data on a pie chart, especially with a good number of categories. When it comes to present information for making decisions it is better to go for more readable visualizations like bar charts (check my post: “PIES ARE FOR EATING NOT FOR DISPLAYING DATA”).

However, a different story is with data journalism, when the purpose is to entertain, or inform the audience. In such case, I would give green light to anybody, who would like to present any complicated data on any variation of a circle chart like a sunburst, radial chart, or spiral chart.

Those charts give you an opportunity to present complex hierarchical information on one chart, so even though there are maybe not idealistically readable, they are still concentrated within one visualization, which is an advantage for the audience. Do not forget that data journalism has a different purpose. The main goal is to pull readers into the story. Surprisingly, more complex visualizations with a huge number of details, colors and shapes can be a better agent than simple one to achieve that mission. It is because readers must spend more time decoding that visualization and retrieve information from it. Another aspect that increases the involvement of readers is chart interactivity. Of course, that case can be applied only on website media.

EXAMPLES

Below infographics are good examples of the complexity vs. the reader engagement. It is hard to understand them at glance. You need to hang your eyes for a longer time and go deeper to acknowledge these images.

The huge advantage is adding other layers or rings to the image. Thanks to that technique additional data are introduced into a chart and we can interpret or read information from different angles or levels. Looking on the same image with several layers of information helps us to find interesting patterns and observation. Would be much harder to achieve that effect when having separate charts.

Global statistics

Our Mother Earth is round at it has a connotation with a round object like a circle. Why not use it to strengthen the message. The chart is combined with several charts placed on circle x-axis: life expectancy and average hours of sunshine is a bar chart. Life satisfaction is a heatmap.

https://www.designboom.com/design/sunshine-and-happiness-infographic/
https://www.visualcapitalist.com/visualizing-all-of-earths-satellites/

Time

The time in western culture is perceived as linear from years perspective. When we present years the line chart or bar chart would be our first choice. However, when it comes to the elements of one year, we perceive them as a cycle. What I definitely admire in circle charts is the possibility to present any periodical phenomenon connected with time:

  • Seasons: Summer, Autumn, Winter, Spring
  • Months
  • Weeks
  • Hours
  • Minutes
https://www.digitalartsonline.co.uk/features/graphic-design/award-winning-infographic-designer-nadieh-bremer-on-how-create-powerful-data-visualisations/

Hierarchical information

Presenting hierarchical data is challenging. However, sunburst charts can handle that. Sunburst charts consist of rings that represent a separate level of hierarchy. This visualization gives us an opportunity to present very complex information in one view.

Note that hierarchical information can be presented as qualitative or quantitative.

The below example presents types of cheese categorize by type of milk and their hardness. This information is qualitative. Another type of visualization that we could use would be a treemap. However, a treemap does not look such good as a circle chart.

https://stackoverflow.com/questions/17069436/hierarchical-multilevel-pie-chart

DOS & DONTS

  • Use colours to catch the attention but remember to choose them in accordance with best practices for colour blindness disabilities. Studies show that around 10% of people population have some disabilities in colours distinguish.
  • Always provide the legend. The legend should explain the meaning of colours, shape, sizes and even positions of objects on your visualization.
  • Add short text on visualisation. If there are points that should be emphasis place additional text with an explanation nearby them. The well-balanced text provides context for a particular point.
  • Plan the objects’ size with available space in mind and readability aspect.
  • Do not use too small fonts.
  • Do not use decorative fonts as they are not readable.
  • Remember about the title and short description of the data visualization.
  • Leave whitespace around the visualization to not clutter the page.

How to speed up information decoding by simple data visualization tricks – the story of one chart.

How many times have you struggled to quickly understand what a chart is presenting? It is something that I often experience in media when reading articles or watch some statistics on TV. Sometimes is extremely hard for me to make sense of what I see, just because I am not the subject matter expert and those data at a glance do not seem familiar. And let face it, I am a data person. What must feel ordinary people, who do not work with data on a daily basis and are not highly data literate?

This post is inspired by data visualisations in the article that I have read recently about the employment situation in the UK. You can find the link to the paper at the bottom of the post.

We are going to focus on three easy to introduce improvements to make any chart more readable, impactful, and thoughtful:

  • Additional Axis labelling
  • Annotation
  • Preattentive Attributes                

As an example, we will improve the below chart that presents changes over years of staff availability index.

Additional Axis Labelling

I am not familiar with the staff availability index. From the title and footer of the chart, I understand that the higher, the better. However, that information could be served on the plate. Based on my experience, I can see an easy fix for such a case that speed up the cognitive work of my brain. Most of the time, when some charts are presented, they present some changes over time or comparisons between two or more phenomena. 

In this case, adding small arrows to the Y-axis and additional words describing axis directions give much more sense to the chart and improve the audience experience. Now the chart presents not only changes over time but informs the expected direction of change.

Annotation

There is a common myth that “Data speaks for itself”. No data can speak because it does not have a tongue. The responsibility of proper understanding of the message lies on the messenger side.  Another quick win is adding more text to the chart itself. Additional description or insight help people to process information more effectively and, thanks to visual presentation, make it easier to remember. 

I have added a sentence from the article next to the point that I have wanted to emphasise. The rich text pays attention to the audience eyes, and the soft grey line directs to the specific point on the chart.

Preattentive attributes

Each object on Earth has properties like shape, colour, size, position. This is what we notice without using conscious effort, and because we do not involve too much conscious effort, we must take advantage of it to decoding information faster. Thanks to them, we can guide the audience eyes through our data visualisation and point them exactly where we want.

Introducing a small red dot is a true game-changer for presenting information on the below chart. We can get this effect by taking off the line chart colour and add to the chart another object with a different shape (circle), size (the circle is significantly thicker than the line), and by adding contrasting colour (the red one). At the final stage, let us analyse our eyes movement. First of all, our eyes start looking at the chart with the title (that is why do not forget about titles! Never!). Then they go straight to the red dot. Just next to the red dot is an insight that explains that point.  Next, they track the line chart and finally look at additional Y-axis labelling. Now, our brain, after collecting all this information, can process them and make sense of those data.

I would recommend those three easy to remember and use tricks to uplift any data visualisation that will improve your audience experience.

The link to the article:

https://www.theguardian.com/business/2021/jul/08/uk-employers-struggle-with-worst-labour-shortage-since-1997?CMP=Share_AndroidApp_Other

Use the force of tables, but choose wisely.

“You must unlearn what you have learned”, said Master Yoda. Tables are not visuals! Truth? Have you ever heard that?

Nothing more wrong. Tables are a very powerful tool for visualizing data if you use them wisely. The main advantage of tables is the ability to present several measures for the same category in one row. This allows your audience to make quicker decisions because all important information is “on the table”.

However, the human brain READ the table. There are plenty back and forward iterations which it does to understand table content. So to make understanding easier, some additional elements should be introduced into tables. In the end, we don’t want to overload the lazy brains of our audience. Let’s see how we can improve tables to make them more accessible for people.

What makes the bottom table better than this at the top? There are several bullet points, which I’m going to address. You should have already noticed titles. Titles, itself, are introducing a huge difference.

Flat table

This table is simply flat. All information is at the same level, which means that they equally attract your attention. Nothing is highlighted, except for the second rows… which is unnecessary. Well, it’s hard to read, right? There are more sins: small fonts, cluttering elements such as lines, grey backgrounds, no formats of values.

Meaningful table

In the table, I’ve introduced information hierarchy by using different font colour. Rows and columns headers are in the background. Values have the darker, bold font. What is more, visual elements are added. Bars differentiate revenue volume, RAG icons simply convey the message about target realization, arrows indicate the direction of the year over year change. Columns headers well describe a column content and columns order leads through information importance.

Spaghetti Monster. Visualising multicategories.

When I say “multicategories”, I mean more than 4 categories. Sometimes, a challenge of visualizing multicategories is like an old polish proverb “eat a cookie and have a cookie”, which is hard to put into practice. I often observe how data analysts try to approach this challenge. Common scenarios are for products, countries, businesses, departments, teams, agents or cost centres. For all these data, they try to find out meaningful insights by depict patterns and highlight interesting points… mostly on one chart. That visual decision creates a beautiful piece of abstract art riched in colours, shapes, different sizes of objects, patterns or crossing lines.

Could you imagine how someone must be determined and persistent to look at and try to understand the column bar chart with 15 categories presented on 5 years horizon? Which category has upward and downward trend? Which is a leading one? And foremost which one is which?

When I think about “multicategories”, my first association popping into my head is “clutter”. The clutter is one of the greatest factors of cognitive overload. To understand clutter impact imagine that, you try to talk to your friend in a crowded space like a bar. You are all ears to follow her or his, and even then, you are not successful. The same effort your brain does, when it is exposed to the flashy visualization.

So how to overcome this challenge?

Both visualisation present the same information:

  • trends over time,
  • product comparison,
  • the best and the worst-selling products.
Spaghetti Monster
Clear view

Doesn’t it look like Spaghetti Monster? You rummage with a fork to find a juicy bit of meat. Similar is with decoding some information from this visualisation (line chart), it costs a lot of effort and time. Our brain decoding one information eg. line colours, then stores it in memory, then compares lines position on the chart, then look for trends for each of lines goes back and forward through the chart to make a sense of it.

On the second approach, I’m showing the different strategy to present the same data set. Splitting information into two visualisations gives clarity and ability to draw a conclusion. The left chart enables the receiver to compare sales amount between products and memorise them easily. The right chart provides information about particular product trend in comparison to average sales. In this approach, we guide the audience through data. We pay their attention to important points. We don’t leave them alone having hope that they draw a conclusion on their own.

We are the data storytellers.

Let’s start from 0

I came up with the idea for this article on the last webinar, which I had the pleasure to conduct with my coworkers. One of the participants paid attention to the starting point of the line chart, which I presented. He noticed that the starting point of the axis wasn’t in “0”. He addressed it with the famous book by Alberto Cairo “How Charts Lie” and commented that the line chart should have started at 0.

There is no doubt when it comes to the bar chart that it should ALWAYS start at 0. Bar charts encode data by length. People have developed the ability to compare objects in terms of length for thousands of years of evolution. Thanks to that they were able to estimate how high the food hangs on a tree branch or compare themselves with the enemy to fight or escape. Placing starting point in non-zero skews data and misleads our audience, because in the first place, unconsciously, they will start comparing bars length.

Of course, we can label bars and axes properly. The crime would be to switch off the Y-axis (in such case), what I observe from time to time. But even then in our brain, there is cognitive dissonance. Numbers don’t reflect lengths and proportions. Lengths and proportions are what our brain will remember because numbers are quite fresh phenomenon for our brain.

Let’s compare below examples for the bar and line chart with zero and the non-zero starting point and check what consequences it might have in the interpretation of data.

Skewed Y-axis & Bar Chart

To have no heart attack in the near future and be still in fit, WHO (World Health Organization) recommends taking a 10 000 steps per day. There are plenty of apps which can track your daily physical activities. Above charts presenting my recent results from the same range of dates. On the left side chart, a proper baseline is applied in 0. All daily results fluctuating nearby the daily goal. In one second, the level of dopamine in my blood pomps up looking at bars achieving the daily target.

The right chart doesn’t give me a reason to be proud of my self at first glance. Firstly, my brain notices gaps between bars and target line. And OMG, twice I almost took no steps! If you don’t notice Y-axis label, you can interpret this chart so dramatically. Worse, if you just had a chance to see it for a few second, you would probably make such a conclusion. Your brain wouldn’t have time to notice Y-axis labelling. But two times I exceeded the target more than twice. Awesome! Everything WRONG.

Skewed Y-axis & Line Chart

A different situation is with line charts. There is no length to compare. There are only slops and positions. In this case, context and narration play first fiddles.

On these line charts, the same data set is presented. From the chart on the left side, we can take out a similar story. The performance is almost aligned with the target. However, looking on the right chart, our brain doesn’t make automated assumptions on lengths because there are no lengths. We see connected dots.

And now is a question. Does the non-zero axis skews data at line chart or not?

There is a discussion around it. Still, non-zero baseline, even though there are no bars to compare, can mislead the audience, presenting steep slope of tiny mountains. However, in some cases, having a particular purpose in mind, it can be the best option to choose. Non-zero axis at line charts is good for presenting minor fluctuations or changes of phenomenons like stock exchange rates, products quality tracking (production series) etc. Especially, tracking performance within companies. Even small changes can have a huge impact.

In our scenarios. Well, to pat on the back myself, I would choose the bar chart with “0″ baseline, but to be able to control my daily results in details, I would definitely choose the line chart with non-zero baseline.

Key Ingredient of Compelling Stories. The Power of Context.

Seneca said, “We are more often frightened than hurt, and we suffer more from imagination than from reality.” Imagination is a powerful weapon. Designing compelling data visualizations to sell stories might get a human imagination down to work.

To make it happen, the context is a key player. Without the context, it is hard to understand presented numbers or outcomes. The human brain always seeks for comparisons to create a meaningful picture of the world. In this article, I would like to talk about how we can add context to presenting the behaviour of the phenomenon over time.

From my experience, I often see a single line of eg. revenue, sales, costs or number of claims presented on a line chart. However, without the proper highlighted background, it’s hard to say if what we see is positive or negative. Is this change is for better or worse. Using additional information, the message is strengthened and helps tell a thoughtfully crafted story.

This approach is especially important when the report supports the decision-making process. Quick business insights can be easier revealed when decision-maker can benchmark presented data to thresholds.

Let’s check how different stories can be told. On this chart, we can see a single line represents revenue of company X. Analytical eyes will see the downward trend over time. However, maybe this observation is not so clear for people who have other skills then analytical.

The first story can be about a decline in revenue over two last years. The declines in revenue can be depicted with an added trend line. In real scenario would be good to highlight specific points in time which caused this change.

The second one can focus on now and then. Comparing the two times period, current and last year helps see the magnitude of change. However, it’s good to remember that on such visualization trend over the longer period is lost.

The last one doesn’t emphasize changes over a longer time at all. It just presents performance vs. budget and directs the audience attention to “here and now”.

In conclusion, there are three different contexts for the same dataset, which changing the data perspective. Frankly speaking, combining these three perspectives gives an insightful story of revenue condition.

Numbers with Human Face

Recently, I’ve taken part in a discussion about how to present numbers to convey a message about true people stories.

We often forget that these are not only numbers. Each number represents a human being, his/her tragedy and tragedy of her/his relatives.

Statistics often show numbers, % of populations, rises, falls and trends. There is a huge challenge and effort to depict context and tell the story behind datasets. Especially, when we try to depict in numbers the phenomenon such as #COVID-19. We have to remember that “confirmed cases” are real people, who are diagnosed with coronavirus. A number of deaths is a number of people who lost their lives because of this disease.

Daily, we are exposed to numerous statistics in media, workplaces, schools. They describe current situations, accidents, local and global events like car accidents, infants mortality or unemployment. Most of them are expressed as a ratio or percentage. These formats are not intuitive and for most people are hard to interpret. However, there are some methods, which connect numbers with people. Maybe not with individuals, but with countable human beings, with whom we can empathize.

KPI approach

A good example is the unemployment rate, which is one of the most important economic indicators. In the governmental statistics, unemployment is presented as a ratio of employees to all people who can work.

An unemployment rate expressed as a percentage does not cause any emotions among most of us. Most of us understand what see, but … it is nothing personal. Percentages are abstract objects. It is about closer indefinite part of the population throughout the country.

As studies show, we can transform this message in a way to evoke people feelings and make them start to take a more human perspective. Instead of abstract 20%, we can present that 1 out of 5 people is unemployed. Each of us can count to five. Each of us can easily list five people. Behind this number, people’s faces may stand. In such a small group of people, our neighbour or our family member may be out of work. This is no longer an abstraction but a very real threat.

Human approach

The Nature of the Phenomenon. Linear vs. logarithmic scale

The one dataset, two charts, two opposite stories.

The introduced scale has a huge impact on how we digest and interpret the presented data. The linear scale represents natural numbers, which we can easily compare. The logarithmic scale is not intuitive for us. It’s a mathematical concept, which we can use when we want to describe multiplicative factors or when is a huge skewness towards large numbers. We need to use brainpower to understand it. What is more, we are so used to linear one that we can easily overlook that visual is depicted on a logarithmic scale. We should inform our audience that logarithmic is used… and make sure that they understand how to read it.

Because of COVID-19 huge amount of statistic are generated and published across the internet. Those statistics try to tell a story about COVID-19 phenomenon. Most of them focus on a number of confirmed cases and deaths. I notice two data visualisation’s trends regarding presenting data about this virus. The first one concentrates on the growth of a total number of confirmed cases and the second one on the pace of disease spreading.

Let’s feel the difference.

“PANIC chart” — I saw somewhere a good name of such a linear chart. I couldn’t more agree. Tell me, what feelings this chart evokes in you?

This is an exponential chart (another mathematical concept), which depict the growth of the phenomenon. Very rapid growth to be specific.

Below we can see the same data. However, embedded on a different scale. Please, look carefully. Each grid represents 10 to n power. Don’t you think that the below chart isn’t so scary?

What stories these two charts tell us?

Let’s base them on 18th of Mar and 4th of Apr. The Linear chart tells us that till 18th of Mar nothing spectacular happened. Totally opposite to the Logarithmic one, where we can see the fastest growth of confirmed cases. Between 18th and 4th on the Linear, we can see the huge growth. On the second one, the pace of growth decelerates. After 4th of April, the Linear continues to present the same pace of growth (steep hill), but on the Logarithmic, it’s plain to see that the curve flattens.