good chart source: wallstreet journal, chartr, and consulting firms, the economist. 0:15 took detailed notes and compared them with some of the best chart makers in the business. 0:25 4 main reasons. 1:40 reason 1: simplicity. basic title, basic subtitle, and basic bar chart. 2:00 number 1 goal with charts: readability, not trying to impress the audience. if you scroll along their website and instagram page. compared to some consulting firm. 2:40 reason 2: they match the chart to the message. 3:15 expressed with a year over year percentage. 3:45 you can think of as a combination of line chart and column chart together. 3:50 reason 3: they guide you to the insight, and often they do it subtly. 4:15 the grey line fades into background. 4:45 the information actually shown on the chart all the way there. 5:30 reason 4: consistency: 5:10 the red line here is interesting. 6:30 take a mess of data and turn into a compelling story for anyone to understand in a few seconds.
The title color block is a common tool for designers. Even though titles are generally larger and bolded, a block of color helps readers quickly understand the following is a new and distinctive section, by engaging primal sense of shapes and colors instead of language. You can even use different colors to signify different themes in the same page. Although you can apply the colors to the titles themselves, it can adversely affect text readability and visual consistency of the page.
yeah as a graphic designer, this is not unique at all. very common place. i do this a lot, i personally think titles look better underneath a line. but still, great video
One thing I noticed is that they put scale on the right. The "default" chart is one in the first quadrant of the XY frame (so data is above the horizontal axis and to the right of the vertical one). The Economist moved that into the second quadrant, meaning you get the scale at the end, where they wanted to get to. "something something reached X level" is easier to read if you have the graph there then the X value right at the end of the line. Mocing bar charts to the fourth quadrant is a similar effect.
Just found your channel and i love what you’re doing. I’m a CFP, CFA, and have personally built dozens of charts to teach financial education of tens of thousands of people. I only wish I found your channel while in that role!! Keep up the good work and I will look forward to every new video you put out!
I am so happy you made this video. I have been a subscriber to the economist ever since my undergraduate degree. I learned so much from their stories and their charts. I like you are spreading the message about their incredible work.
I used to be a subscriber, and my favorite of their charts were the complex ones, where they found three or even four data sets or relationships they wanted to bring together in one place. It would take time to process, but at the end I always had to admire how neatly and creatively they solved the design problem. Tufte would approve.
Amazing that being an The Economist reader myself and having in hands a task to develop a new report I was also looking at their visualizations exactly thinking why I find them so interesting, informative and compelling. Top! Thanks!
Takeaways: Simplicity: Choosing simple yet effective visuals with correct and minimal labels Matching the Chart to the message: Choosing the right visual Helping the audience to focus on the message: Highlighting the key category or value in the same colour as the colour above the chart title or bolding the value or highlighting the line/column of concern and subduing the colours of the rest
While this is interesting, a point you make several times quite correctly is that these charts exist to communicate the message in the Economist's analysis. However I would point out that this leads to a 'garbage in, garbage out' situation. Without critical analysis of the messaging communicated by the charts, is there not a risk of saying misleading charts are 'good' just because they communicate the distortion well? I think this dimension is important because you're critical of more complex charts and presentations while these may be necessary to communicate a more nuanced analysis than 'A goes up, B goes up'. Is a good chart not one which communicates truth transparently, tells an angle to the story you might not have considered, or one which is difficult to communicate in words? I'm not saying the Economist does low-quality analysis (I think in general they do quite well), but I think this is an important part of a conversation about presentation of data. I also say this because you frequently praised the skill of journalists telling a story with the data, but I disagree - I don't think it takes skill to pick a data trend that fits your narrative and graph it using a standard template. I think the skill we should value is using data transparently, consistently, and without agenda. I also think to some extent there is a risk of putting the cart before the horse here - maybe sometimes the journalist was handed new data and told to create a written narrative around it, in which case the skill is in the article - but again is this something worthy of praise if the data itself only tells one side of the story? We can manipulate data to tell whatever story we want; is it ethical to praise and platform it uncritically, without a deeper dive into the merit of the analysis?
Just as a journalist is not going to spend 20000 words debunking flat earth in order to get to the meat of a story on satellites, dataviz is always going to be a small slice of the story. The job of these charts is to illustrate the evidence and conclusions drawn upon in the article - of course there are going to be additional complexities and concerns outside of this but that's not the chart's job to communicate those. Data visualisation is very different within journalism vs a primary source with a null hypothesis and much more space for complexity. A chart that doesn't serve the "agenda" of illustrating the story doesn't deserve to be there, and the accuracy and efficacy of the reporting itself is where your criticisms rightly apply, not in the visualisations used. If the story is deceptive, the story is deceptive - that speaks nothing to the quality or lack thereof of the data visualization quality.
It’s notable that you did not mention the Graphic Detail feature whose charts vary from confusing to incomprehensible. The readership hours chart in the 24th Dec edition is a moderate example. I still love The Economist.
Great video. Also note how often they use color (of line series, bar, etc) to point out the data/trend of interest-- and they label series directly, as opposed to having a legend where one needs to look up which color line represents what. This dramatically increases rapid understanding by readers.
Thanks for this video! I started my first work at Nestle, and the entry position is doing reports and charts. This will help me to be better with my charts! Thanks!
Hi! I'm really enjoying your videos. Two things you could perhaps cover (or point me towards...): 1. slide animation and 2. a discussion of the trend to replace traditional "written" reports (e.g. in MS Word) with PPT slides that also serve as the "close-out presentation" and are therefore FAR to wordy.
Maybe a video on examples of unconventional charts that break the generally accepted principles of design but still are successful at conveying meaning in an efficient manner
Hi! Maybe a case study on a designated assignment from data collection, passing by data analysis and ending with the presentation or data visualization proposed to the client in order to share a bit the work done in between data collection and final visualization.
Two ways the pros use: Code it in R and make the chart with ggplot, or make it anywhere and redraw it by hand with Adobe Illustrator for complete control.
The chart on Taiwanese identity is only a slightly simplified version of the chart that the NCU uses to present this data. They do it a bit more colorful, with another category (no response) and points/small circles for the individual data points of the surveys. The version of the Economist is good, though, to boil it down to the essentials and present it to a new audience.
Out of interest, you should have a quick critique of the very complex Economist charts, because when they go complex they go extremely complex. They can get 4 or 5 dimensions on a chart. Sometimes I spend minutes trying to find the trend before realising should have just looked at the title. One thing they often do poorly is not use enough colour. They will have 3 different shades of red and three different shades of grey and the chart legend doesn’t help at all.
Hi Paul , your videos are really awesome and helping design ppt better, i have a question like if we are showing limited information then it's manageable, however if company having multiple business lines , let say 5 business verticals and even if we show only sales and profit and then comparison with last year and budget, its come too much information. Can u suggest any better way to show other than Excel grid. Thank you!
Serious Question : Do you have to pay The Economist to feature their charts in your videos? I think they have a rights management department that requires this.
What I don't like about them is the fact, that they just give as numbers next to line charts. Without telling us what they actually are. It's confusing, because my eyes read pictures way faster and than I have to go back to the text to actually follow up on what they represent. Ist it %? Is it amount? Just write the "%" symbol above the Y Axis. :/
That longest reigning monarchs chart is actually really confusing. The y-axis is centuries? Why isn't it just a chart of the longest reigning monarchs in descending number of years reigned?
A purely rank-based chart of reign length would place reigns from disparate historical periods immediately adjacent. This makes it just a bit harder to connect the names and historical period of each entrant with one's historical knowledge. By instead presenting the "running maximum" entry for each birth century, we get a more recognizable list because names appear in chronological order, and we _gain_ the trend over time. You can think of this chart as having been distilled from a straightforward sequential bar chart of _all_ reigns by editing out short reigns. Alternatively, think of this chart as a purely rank-based chart that has been sorted by birth century, and again showing only the longest entry of each century. This chart is indeed really well thought-through, showing long reigns across a millennium. Thank you for highlighting this example!
I don’t agree at all. Most people don’t know all the protons monarchs by just their name. Putting them in their historical context is very helpful to show through centuries how long the reigns have been.
2:22 That's an ugly one. It's hard to work out which line is which country. Which tick marks are for starts of years? Marking them using full vertical lines would make this clearer. And having a horizontal line every 3% is bizarre. 3:10 Another confusing tangle of lines. 2:41 Why not a top 10?
If there’s no sound try checking your settings in the video player. You should be able to select English under the Audio Track option. Let us know if that does the trick or if it’s something else!
Highly subjective. From the point of view of a statisticians, the graphs are primitive and often misleading. From the point of view of a propagandist, these are perfect. They tell you what to think, even if it's utter garbage. They leave out important aspects and context so they can promote their message. These charts don't convey information, they convey messages. You don't give these charts to someone who's supposed to make an informed decision, you give them to someone you want to make the decision you'd like to hear. And it doesn't require a degree in statistics to see through the ruse, you can only fool fools with that. Really unprofessional.
The Economist's charts are nice, but it's a shame how their reporting has lurched to the left in recent years. I subscribed for many years but I'll never give them my money again.
Literally always been like that. At least what we would call the left now (which they wouldn’t have then). They’re a self-described ‘Liberal’ (capital L) newspaper. They’re progressive. What did you expect?
BLACK FRIDAY SALE IS LIVE! Use code "blackfriday" for 20% OFF all courses at TheAnalystAcademy.com
good chart source: wallstreet journal, chartr, and consulting firms, the economist.
0:15 took detailed notes and compared them with some of the best chart makers in the business.
0:25 4 main reasons.
1:40 reason 1: simplicity. basic title, basic subtitle, and basic bar chart.
2:00 number 1 goal with charts: readability, not trying to impress the audience. if you scroll along their website and instagram page. compared to some consulting firm.
2:40 reason 2: they match the chart to the message.
3:15 expressed with a year over year percentage.
3:45 you can think of as a combination of line chart and column chart together.
3:50 reason 3: they guide you to the insight, and often they do it subtly.
4:15 the grey line fades into background.
4:45 the information actually shown on the chart all the way there.
5:30 reason 4: consistency:
5:10 the red line here is interesting.
6:30 take a mess of data and turn into a compelling story for anyone to understand in a few seconds.
The title color block is a common tool for designers. Even though titles are generally larger and bolded, a block of color helps readers quickly understand the following is a new and distinctive section, by engaging primal sense of shapes and colors instead of language. You can even use different colors to signify different themes in the same page. Although you can apply the colors to the titles themselves, it can adversely affect text readability and visual consistency of the page.
splendid explanation! kinda wonder why it fits so well
yeah as a graphic designer, this is not unique at all. very common place. i do this a lot, i personally think titles look better underneath a line. but still, great video
One thing I noticed is that they put scale on the right. The "default" chart is one in the first quadrant of the XY frame (so data is above the horizontal axis and to the right of the vertical one). The Economist moved that into the second quadrant, meaning you get the scale at the end, where they wanted to get to.
"something something reached X level" is easier to read if you have the graph there then the X value right at the end of the line.
Mocing bar charts to the fourth quadrant is a similar effect.
great insight
Just found your channel and i love what you’re doing. I’m a CFP, CFA, and have personally built dozens of charts to teach financial education of tens of thousands of people. I only wish I found your channel while in that role!! Keep up the good work and I will look forward to every new video you put out!
It's the tasteful thickness of it, that subtle off-white coloring. Oh my god, its even got a source.
Paul Allen's charts are the best!
I am so happy you made this video. I have been a subscriber to the economist ever since my undergraduate degree. I learned so much from their stories and their charts. I like you are spreading the message about their incredible work.
I used to be a subscriber, and my favorite of their charts were the complex ones, where they found three or even four data sets or relationships they wanted to bring together in one place. It would take time to process, but at the end I always had to admire how neatly and creatively they solved the design problem. Tufte would approve.
Amazing that being an The Economist reader myself and having in hands a task to develop a new report I was also looking at their visualizations exactly thinking why I find them so interesting, informative and compelling. Top! Thanks!
Takeaways:
Simplicity: Choosing simple yet effective visuals with correct and minimal labels
Matching the Chart to the message: Choosing the right visual
Helping the audience to focus on the message: Highlighting the key category or value in the same colour as the colour above the chart title or bolding the value or highlighting the line/column of concern and subduing the colours of the rest
I really like the content of this channel, and the Spanish dubbing gives it a particular and fun touch that helps me learn faster.
Another great video. Their simplicity is out of this world.
I used to look at their graphs and wonder what software they use, love this video, thanks a lot! 👏🙌
While this is interesting, a point you make several times quite correctly is that these charts exist to communicate the message in the Economist's analysis. However I would point out that this leads to a 'garbage in, garbage out' situation. Without critical analysis of the messaging communicated by the charts, is there not a risk of saying misleading charts are 'good' just because they communicate the distortion well?
I think this dimension is important because you're critical of more complex charts and presentations while these may be necessary to communicate a more nuanced analysis than 'A goes up, B goes up'. Is a good chart not one which communicates truth transparently, tells an angle to the story you might not have considered, or one which is difficult to communicate in words? I'm not saying the Economist does low-quality analysis (I think in general they do quite well), but I think this is an important part of a conversation about presentation of data.
I also say this because you frequently praised the skill of journalists telling a story with the data, but I disagree - I don't think it takes skill to pick a data trend that fits your narrative and graph it using a standard template. I think the skill we should value is using data transparently, consistently, and without agenda. I also think to some extent there is a risk of putting the cart before the horse here - maybe sometimes the journalist was handed new data and told to create a written narrative around it, in which case the skill is in the article - but again is this something worthy of praise if the data itself only tells one side of the story? We can manipulate data to tell whatever story we want; is it ethical to praise and platform it uncritically, without a deeper dive into the merit of the analysis?
Great comment!
Just as a journalist is not going to spend 20000 words debunking flat earth in order to get to the meat of a story on satellites, dataviz is always going to be a small slice of the story. The job of these charts is to illustrate the evidence and conclusions drawn upon in the article - of course there are going to be additional complexities and concerns outside of this but that's not the chart's job to communicate those.
Data visualisation is very different within journalism vs a primary source with a null hypothesis and much more space for complexity. A chart that doesn't serve the "agenda" of illustrating the story doesn't deserve to be there, and the accuracy and efficacy of the reporting itself is where your criticisms rightly apply, not in the visualisations used. If the story is deceptive, the story is deceptive - that speaks nothing to the quality or lack thereof of the data visualization quality.
Seems like you read How to Lie with Statistics.
It’s notable that you did not mention the Graphic Detail feature whose charts vary from confusing to incomprehensible. The readership hours chart in the 24th Dec edition is a moderate example. I still love The Economist.
Great video. Also note how often they use color (of line series, bar, etc) to point out the data/trend of interest-- and they label series directly, as opposed to having a legend where one needs to look up which color line represents what. This dramatically increases rapid understanding by readers.
Thanks for this video!
I started my first work at Nestle, and the entry position is doing reports and charts.
This will help me to be better with my charts!
Thanks!
The Guardian in the UK has the best charts IMO. They invested heavily into data viz many years ago and it's paid dividends IMO.
Hi! I'm really enjoying your videos. Two things you could perhaps cover (or point me towards...): 1. slide animation and 2. a discussion of the trend to replace traditional "written" reports (e.g. in MS Word) with PPT slides that also serve as the "close-out presentation" and are therefore FAR to wordy.
Thanks for watching! What video would you like to see next?
Maybe a video on examples of unconventional charts that break the generally accepted principles of design but still are successful at conveying meaning in an efficient manner
Hi! Maybe a case study on a designated assignment from data collection, passing by data analysis and ending with the presentation or data visualization proposed to the client in order to share a bit the work done in between data collection and final visualization.
Could you do a Video of converting a poor Excel spread sheet into a really professional one. I mean pure numbers, no graphics.
great video, financial times is a bit similar too (i feel they often have more complex data charts though that are beyond my capability).
Could you do more videos on how to create and format these types of charts please? Do you primarily use Excel or PP?
Excel is the best way to initially create it. Once you export it to PowerPoint, you can continue to simplify to align with your story
Two ways the pros use: Code it in R and make the chart with ggplot, or make it anywhere and redraw it by hand with Adobe Illustrator for complete control.
@@zachpw I highly doubt that in consulting context.
@@definitelynosebreather but it's cool
That time is better spent in other work
The chart on Taiwanese identity is only a slightly simplified version of the chart that the NCU uses to present this data.
They do it a bit more colorful, with another category (no response) and points/small circles for the individual data points of the surveys.
The version of the Economist is good, though, to boil it down to the essentials and present it to a new audience.
FT has the best charts of any newspaper globally
Amazing content and straight to the point, liked & subscribed
Out of interest, you should have a quick critique of the very complex Economist charts, because when they go complex they go extremely complex. They can get 4 or 5 dimensions on a chart. Sometimes I spend minutes trying to find the trend before realising should have just looked at the title.
One thing they often do poorly is not use enough colour. They will have 3 different shades of red and three different shades of grey and the chart legend doesn’t help at all.
What kind of graphic tablet are you using in this video?
Great breakdown, super useful
What red is that, its not a red red i can use in corporate
Hi Paul , your videos are really awesome and helping design ppt better, i have a question like if we are showing limited information then it's manageable, however if company having multiple business lines , let say 5 business verticals and even if we show only sales and profit and then comparison with last year and budget, its come too much information. Can u suggest any better way to show other than Excel grid. Thank you!
thanks for the clear explanation
Can you share slide examples for operating model redesigns in the IT sector?
How are those charts built? What software do they use?
wondering same thing here...
@analystacademy, can you help here?
Love this channel.
Serious Question : Do you have to pay The Economist to feature their charts in your videos? I think they have a rights management department that requires this.
Great video!
What I don't like about them is the fact, that they just give as numbers next to line charts. Without telling us what they actually are. It's confusing, because my eyes read pictures way faster and than I have to go back to the text to actually follow up on what they represent. Ist it %? Is it amount? Just write the "%" symbol above the Y Axis. :/
they have a cool newsletter about their data viz techniques!
That longest reigning monarchs chart is actually really confusing. The y-axis is centuries? Why isn't it just a chart of the longest reigning monarchs in descending number of years reigned?
A purely rank-based chart of reign length would place reigns from disparate historical periods immediately adjacent. This makes it just a bit harder to connect the names and historical period of each entrant with one's historical knowledge.
By instead presenting the "running maximum" entry for each birth century, we get a more recognizable list because names appear in chronological order, and we _gain_ the trend over time. You can think of this chart as having been distilled from a straightforward sequential bar chart of _all_ reigns by editing out short reigns. Alternatively, think of this chart as a purely rank-based chart that has been sorted by birth century, and again showing only the longest entry of each century.
This chart is indeed really well thought-through, showing long reigns across a millennium. Thank you for highlighting this example!
I don’t agree at all. Most people don’t know all the protons monarchs by just their name. Putting them in their historical context is very helpful to show through centuries how long the reigns have been.
A picture instead 1000 words.
💯
Old joke - an economist is the guy who, if you forgot your phone number, can estimate it for you
2:22 That's an ugly one. It's hard to work out which line is which country. Which tick marks are for starts of years? Marking them using full vertical lines would make this clearer. And having a horizontal line every 3% is bizarre.
3:10 Another confusing tangle of lines.
2:41 Why not a top 10?
So, how do they make it? I watched the vid without sound.
If there’s no sound try checking your settings in the video player. You should be able to select English under the Audio Track option. Let us know if that does the trick or if it’s something else!
Its kinda weird seeing this big dude being voiced by a teenager girl with the Portuguese dub option
Highly subjective. From the point of view of a statisticians, the graphs are primitive and often misleading. From the point of view of a propagandist, these are perfect. They tell you what to think, even if it's utter garbage. They leave out important aspects and context so they can promote their message.
These charts don't convey information, they convey messages.
You don't give these charts to someone who's supposed to make an informed decision, you give them to someone you want to make the decision you'd like to hear.
And it doesn't require a degree in statistics to see through the ruse, you can only fool fools with that. Really unprofessional.
Their charts are too simple, time series usually show only short period of time which can be often misleading (or easily manipulated).
O homem tá com uma voz de mocinha entojada!
Haha I know... made a mistake with the translation file!
I do not what happened with my TH-cam channel, but the video is in Portuguese. Please, correct it for English.
The video is dubbed into multiple languages actually! You can switch to English in the settings menu (on the playbar).
@@AnalystAcademy thank you
Nice video. My takeaway from these charts is that this administration sucks.
Guy makes linear estimation of food price inflation and you call it good graph because it is appealing? Who cares when it's wrong
The Economist's charts are nice, but it's a shame how their reporting has lurched to the left in recent years. I subscribed for many years but I'll never give them my money again.
Literally always been like that. At least what we would call the left now (which they wouldn’t have then). They’re a self-described ‘Liberal’ (capital L) newspaper. They’re progressive. What did you expect?
how dare they make progress
Lost me at "Russia climbing out of the recension" - that dump will cease to exist soon