If you've missed it in early Feb 2023, there was a Eurasian data conference organized by Alex Kolokolov called the data2speak. I certainly learnt a lot from the talented speakers and learning more about the struggles between business dashboard and data journalism.
Do check out their official page and YouTube for tutorials for how to build effective business dashboards. While it's Power BI-focused, the principles and UX design tricks are definitely tool-agnostic!
There were two main categories - Business Dashboards and Data Art & Journalism; with over 40 submissions. Each submission were graded in the following criteria, which measures the effectiveness and design.
Selecting the winners must have been really tough, because the quality of the submissions were really on-point. It's also refreshing to see so many different takes on the same dataset, which shows how framing creates different narrative.
Felipe's piece is really inviting, with so much interaction built into it. While it was built for a specific group of audience, his dashboard has that UX polish that allows the users to quickly understand the insights (Even for non-football fans).
A clean and minimalistic design that capitalizes on the Gestalt principles for the layout. I'm a huge fan of negative space, so this really captures my eye. Also, if you look carefully, you would notice that Sofia has created that narrative around hiring for this dashboard, so it could be something that would interest say.. talent management team.
Similar to Sofia, there's very good Gestalt principles in place. However, you may also notice that the storytelling here is very different. Natalie's focuses on employee engagement and satisfaction instead.
A really cool piece that highlights social impact, specifically on the Bridges to Prosperity initiative for rural communities. The narrative is well thought out and brings us along to show how connected to critical destinations (healthcare, food, employment opportunities) and how it is distributed demographically and geographically.
Nicole's piece focuses on the grassroot initiatives that move us towards a more sustainable planet. The visual hierarchy is so easy to follow and the viz goes to show the overall picture, the geographical distribution, which energy sources are being innovated, and going thru the different sustainability goals.
Another piece on sustainability, but focused on climate change. While the first two talked about what we are doing, this talks about what’s happening now and where the worst areas are. This is storytelling.
Sankey diagrams are captivating bcoz it’s exciting and tackles both the overview and granularity (part-to-whole) like why some people love pie charts. There is an increasing trend of use of sankey trends, but the best practices hasn’t been followed (like with pie charts). What Martina did, was to provide clear instructions and visual aid to help in that. The orange hue represents 2019 being better than 2018, and with that you can immediately see where the highlights and lowlights are. There were a lot covered in the workshops and panel talks, but here’s a summary of what we learnt. While much of the content was structured around business dashboards and data journalism, you can still apply these principles to other forms of information design, like a pitch deck, or a post-even analysis, or even a status update.
The rest of the conferences was filled with amazing speakers who talked about everything data, design and building that data culture. You can still check out the recordings over at data2speak but here's three main learnings my team and I have personally gained from it!
So the first thing for any visualization is the purpose. Is it necessary and does it add value to your conversation. For you to address this, it is always good practice to start with questions, and the answers you are trying covey using DV. It is imperative that you keep this at the back of your mind at all stages of the process to ensure that you are not deviating from the end result.
The next is to understand how you can deliver the right message to the right audience using the right medium in the shortest amount of time.
Lastly, you need to consider the data literacy of your audience, and the emotions or call-to-actions you wanna invoke, which determine how much granularity or interactivity is required.
The next important piece (as presented by Sarah Bartlett) is accepting the fact that you are not the user, and the importance of understanding who is going to use the dashboard, what are they trying to monitor/understand, how will they use this information, how often will they use the dashboard, how will they accesd the dashboard and how familiar are they with the data.
Reiterate early and often is also a crucial step as it ensures you always have the right view of the data for your stakeholders. One of the popular design processes called out was the Double Diamond, which is a continual loop of feedback and iteration.
One of the ways you can further guide your audience in a User Experience Design (UXD) approach, is starting with the overview, zoom and filter and details on demand.
What's data visualization without talking about design? While there are many ways to visualize data, there are still some guidelines we can adopt to make out message clearer and reduce cognitive load. The best way is of course... "stealing like an artist" - don't reinvent and reuse ideas that have proven to work, and add your own personal touch to it. And we learnt about the Squint test too, if you squint your eyes at a data visualization piece, you can immediately pick up all the important elements that "pop", and you can use this trick to modify how you want certain elements in your viz to interact.
We can also take a page from data journalism, which breaks traditional data visualization guidelines to tell a story. The main ideas they talked about is Breaking News, which uses 1 or 2 data points to craft the whole visual around, or News application, which tells multiple data points around a central theme, or Visual Investigation, which is cross-sectional view of the data in detail. Similarly in a business setting, that would be your KPI numbers, monitoring dashboard and operational dashboard.
The last piece on design was about semantics, and how ideas, icons and colors play an important part in tapping into unconscious bias. The International Business Communication Standards (IBCS) was coined in 2015 to help with this, and you can definitely reuse this checklist in your everyday dashboard building!
There's also a fun presentation on using colors, especially when there's an innate color associated with that data point. Like using Blue for Blueberry, Yellow for Banana and so on.
And there you have it, a quick recap of the data2speak conference. Be sure to check out the full recordings if you wanna learn more! Hope you found this insightful or interesting... if not you know what you have to do 😉