After last week’s smashing results with endless great vizzes and submissions for Makeover Monday, we have another strong week behind us with many excellent visualizations built on a dataset about sustainable public transport in cities around the world.
This week, I have four lessons, including one that is Makeover Monday specific, so let’s get straight to it…
LESSON 1: COLOURS
Colours are a great way to highlight important data points, to differentiate dimension members or the magnitude of a problem. Colours can also be very effective in communicating emotions.
Experimenting with colours, whether for the background of a viz or by creating a colour palette for a particular topic is a good way to improve the way we communicate information.
It’s easy to go overboard though and simplicity should still prevail, especially when we focus on data and information over art.
If you want to communicate clearly and effectively, less is often more and following this week’s submissions here are some questions and suggestions to consider:
- Does your viz need a background colour? The default is white for a reason. If you change it, make sure your viz is still easy to read and comprehend. Keep in mind that colour perception of your chart colour changes as background colours change and a red and grey bar chart that ‘pops’ on a white background, might look dull on dark grey or jarring on a yellow background
- Does every dimension member need to have a colour? It depends. If you pick a colour for male/female which repeats throughout your viz, that’s perfectly fine. If you have 100 cities and each of them has their own colour, that’s not so fine. Who will differentiate between 100 different colours? And Tableau will repeat colours or have very similar shades (if you use the rainbow palette). Also, if you’re colouring every bar in a bar chart in a different colour, is that necessary? The individual bars are already the differentiator
- Is your diverging colour palette easy to comprehend? When you apply diverging colours to a measure, make sure you pick a range with distinct ‘min and max colours’, i.e. end points in the continuum. If you start with red and move to purple via orange and pink, these shades can all look very similar in your viz. It’s better to pick stronger contrasting colours, such as orange and blue, red and blue, grey and purple, etc.
Keep the above suggestions in mind when developing your viz, especially when you have a clear message that you don’t want to distort.
LESSON 2: ATTENTION TO DETAIL
I can’t say this often enough, so I’ll repeat it again :-). The best way to make your vizzes stand out is to pay attention to detail. This relates to the content, the design and the chart choices you make to tell your data story.
Of course we should aim to use the most effective viz for the data and the story we’re trying to tell. But the best bar chart in the world is going to still disappoint our audiences if we don’t focus on the small stuff as well. The tooltips and how they’re formatted, the alignment within the dashboard, the choice of borders and gridlines etc.
There are bigger potential issues as well and I suggest you check these once your viz is published to make sure it looks just as great on Tableau Public (or other platforms you choose) as it does in Tableau Desktop (or your tool of choice).
- Are all your titles and text boxes easy to read, i.e. nothing is squished, cut off, in a weird illegible font?
- Have you ‘explained’ your colours and/or symbols somewhere either within a description or a legend etc.?
- Have you given your data space to breathe?
- Have you removed unnecessary ‘clutter’ that could distract your audience from your key message?
- If you have used interactive elements, are they easy to understand?
- What is the overall impression of your viz? Does it draw your attention?
- Is there anything you can remove to improve the clarity of your story?
Ask yourself these and other, similar questions and see if it helps you improve your viz. Don’t be afraid to get feedback from us and the community along the way and seek out others whose style you like or admire and find ways to adopt some of their skills.
LESSON 3: MAPS – DO YOU NEED THEM?
Andy and I both know that when we provide datasets with City or Country names, someone will create a map. Actually, lots of people create a map. And maps are great. They’re can be excellent way to visualize data and to get people’s attention, because your typical audience is very familiar with maps. Maps provide nice context and they can also be made very interactive.
However, maps are not always necessary, and just because there is geo information, doesn’t mean that we should include a map in our viz. Why?
- Maps take up space and if the information communicated through the maps is very light, then they could simply be a waste of valuable screen real restate that could otherwise be used for insights and telling a more powerful data story
- Maps can easily distort your story, especially filled maps, because they make countries with large landmasses look more significant compared to countries with smaller landmasses. Let’s take the random example I’m just making up on the fly, of bicycles per capital in the Netherlands compared to Russia. I bet the percentage is higher in the Netherlands. A filled map will make Russia look more significant because the whole country will be filled with colour, while the Netherlands may be ‘squished’ and hardly visible among the other European countries because they’re all pretty small in comparison.
- Despite the great interactivity, maps that include many countries or even the whole world, can make it difficult to click on and interact with individual countries, so filtering needs to be well thought-through to enable optimal interactivity
I’m a big fan of maps myself and it’s tempting to use them often. I’d suggest, however, that we ask ourselves when building our data stories, whether a map is the best way to present the information at hand or if a simple bar chart could do a better job.
We’ll be hosting a webinar in two weeks on November 30 together with Mapbox, where we’ll cover all these ideas and suggestions, give you some live demos and recommendations.
You can sign up for the mapping webinar here.
LESSON 4: SOME MAKEOVER MONDAY SPECIFIC STUFF
We have welcomed a number of new people to our community lately and it’s great to see the enthusiasm with which this project is being embraced by everyone. To make our lives easier and so that we can provide the most value for everyone, here’s a quick reminder on the easiest way to get started, get heard and get feedback:
- Download the data from www.makeovermonday.co.uk
we post datasets usually on Sunday afternoon (GMT)
- Read the accompanying article and check out the original viz
- Create your own viz in whatever tool you like
- Tweet your submission, including
- A picture (this will drive interactions from other people and will help us check out your viz on mobile)
- A link to the interactive/online version of your viz
- Tag us (@VizWizBI and @TriMyData) in your image or tweet, which makes it much easier for us to find your submission (on mobile we typically just look at those notifications, rather than searching through the #makeovermonday hashtag)
- If you want comprehensive feedback from us, also include the hashtag #MMVizReview and register for the weekly Viz Review webinars on our BrightTALK channel (Please, please, please include this hashtag when you first post your viz so we don’t have to scroll through a whole conversation to find the link to your viz when we do the webinar)
- Participate in the conversation on Twitter
- If you received feedback during Viz Review, feel free to iterate on your viz and post an updated tweet with your new version. We love seeing people improve their work throughout the week
If you have questions along the way, you can always ask us. Don’t ever be afraid to send us your questions, chances are, someone else is wondering about the same thing, so by asking you’re helping more people understand, get clarity and receive more information.
Thanks to this great community for your participation and your continuous enthusiasm. It’s been an absolute blast and we really enjoy working and collaborating with you all to create better data stories – one week at a time.
What I like about it:
- Simple, uncluttered design that is easy to understand
- Encourages user interactivity by giving the viewer a way to change the reference city
- Great colour palette, which shows clearly which cities rank above and below the chosen ‘reference city’
- Michael learned something new by iterating on his viz after Viz Review
What I like about it:
- At the risk of repeating myself: Klaus always surprises me with his designs and the approach he takes
- Another unique viz and this time he links it to the SDG action campaign work he did in September by connecting the topics, which is great
- Great interactivity! Well formatted tooltips and check out the icon in the top left corner, which changes as you hover over different cities
- The design is really cool and the blue points guide the viewer, while giving the ‘skyline’ structue
- Great description for context
What I like about it:
- Great colour choices and really nice design
- Mobile design, it looks great on my iPhone!
- Interactivity is encouraged and is easily achieved through large boxes to be selected
- Nice labels on the bar chart
- Flows well from top (high level) to bottom (more detail)
What I like about it:
- Fantastic interactivity!
- Great design – this is more than a simple Makeover but I really appreciate that Mike has taken the time and a different approach to the topic and took apart the index to help all of us learn more about it
- Great annotations, which help users understand the triangle layout AND give additional information
- Well designed tooltips and really helpful context provided in the explanations at the top
- The attention to detail makes my heart sing!