How are dataviz projects evaluated?
Experience with the international dataviz MYDS Award
Early this year in the framework of the international dataviz conference Make Your Data Speak with Data2Speak team we held our first international Dataviz Award under the same name! You can read the organizer’s general impressions and the results in this article.
The aim of our Award was to identify projects from two different areas of data visualization — Data Storytelling and Business Analytics! This is how we named our two categories: “Business Dashboards” and “Data Art and Storytelling”.
We initiated this competition to exchange best practices, judge great projects, and allow students to try their strengths. The winners could even receive money prizes according to place and category.
During the first months of the competition, students participated free of charge in our project application process — we held online reviews of participants’ work with our great judges, discussing with them the topic of data visualization. You can watch these video analyses on the Data2Speak YouTube channel.
When all the deadlines had passed — it was late 2022, early 2023 — our team selected a shortlist from all the projects, and then the juries started their work to evaluate the projects according to several criteria.
The criteria were as follows: Visual Design, Technological Innovation, Analytical Skills, and Narrative!
For the evaluation of the projects in this competition, we have managed to gather professionals from different fields, and different countries, with different preferences and skill stacks. Data journalists, data artists, business analysts, data storytellers, infographers, and data designers!
Our incredible judges did a great job and assessed several dozen projects and left feedback to all participants. We thank them very much for their hard work!
In addition, they left recommendations and comments on the projects, and we decided to share the most popular recommendations and comments with you. We can look at — which project parameters seemed meaningful and worth mentioning, both in a positive and negative way!
After all, what an interesting dataset it is! Unfortunately, it would take a lot of effort to categorize and analyze it properly. But for now, we have done a small analysis of the frequency of words and terms encountered, hoping that this will describe the main parameters that the jury looked at when sorting out the projects.
Based on the data processed and filtered from function words and symbols, we have put together a visualization called a word cloud as an illustration for this article. Now let’s look at the similarities and differences between the reviews in the different categories.
Let’s start with Business dashboards!
Even from our pretty but not practical word cloud for analysis, it is clear that the charts — their colors, formatting, and design — were the focus of our judges’ attention.
If we remove the common words from the data and combine synonyms, we get the following list of important points in the project:
And hopefully, after this ranking, you’ll never forget to design tooltips on your dashboard again! It turns out that such a minor interactive collaboration gets the special attention of judges, and probably future users.
Well, the use of color in dashboard design really makes a huge difference. We use color to design the dashboard itself, highlight data, and create charts.
Meanwhile, in third place comes the user, whose needs should not be forgotten. And the jury does not ignore it!
Of course, the dashboard topic places a lot of emphasis on technical points — making sure everything works correctly and interactivity as well is important, so don’t forget to check all the basics before publishing.
Here are our tips for you:
- Make sure that the colors are sufficiently contrasting and suitable for colorblind people.
- Ensure that the order of the elements follows the logic of how the user would use the dashboard. Remember the Z and F patterns.
- Check, that tooltips work correctly — whether they are neatly designed.
- Consider the layout of the filters and how they look.
- Make sure all elements of the same type are formatted the same way.
- Pay attention to the layout of your KPIs — are they sufficiently enriched with context?
- Don’t forget some technical details: alignment of elements, indentation, large header, and the presence of data labels on your charts.
Let’s move on to the Data Art and Storytelling category. Here the cloud of words turned out to be quite different.
It feels like more descriptive adjectives are used to describe data art and storytelling projects, more evaluation words. Well, this is really the kind of nomination where there are fewer laws and rules, and more given to creativity and personal perception. But if we look at the word leaders — after combining synonyms and removing common words — there are rules and preferences too.
As in the Business Dashboard nomination, the color leads the way. But further on, judges are interested to know the actual data presented, which is understandable, as these projects often present personal stories and events, not just business indicators.
Since these projects usually present a large amount of data — do not forget about the indentation and the empty spaces, and the cleanliness of the project, so that the data was not cluttered, that decorations and other visual rubbish don’t prevent the reader from reading the data.
So what advice can we give you when dealing with such different and unusual data art and storytelling projects?
Here are some tips:
- Make sure the colors are sufficiently contrasting and suitable for colorblind people.
- Check the data sources you are working with, and specify the source of the data on the project page.
- Check that the project is not overloaded with unnecessary details and embellishments and that elements are logically grouped together.
- Don’t forget the importance of fonts, make sure that all text reads well, is well aligned, and formatted in the same way. Better use fonts of the same family.
- Beauty matters, for a user to really want to explore such a complex project, it needs to look attractive.
That’s it! I hope you have had fun learning with us about the evaluation of dataviz projects. We are also looking forward to seeing you at next year’s Make Your Data Speak Award, which will take place already in this autumn!
Please write in the comments what you pay more attention to when creating or checking dataviz projects! Which mistakes do you encounter the most? Which ones do you consider as important, and which ones are minor and forgivable?
