Catching Up With Giovanni Magni of Accurat Design Studio
Accurat’s head of data visualization spoke to Nightingale about the importance of asking questions in data viz, collaborating with data scientists, and more
I’ve been a longtime fan of the Milan-based data visualization studio Accurat. Founded in 2011 by Giorgia Lupi, Gabriele Rossi, and Simone Quadri, Accurat brings together aesthetic sensibilities with functional visualization. They have been making award-winning data viz for the past nine years with clients like Starbucks, IBM, Google, and MOMA. After joining Giorgia Lupi’s team at Pentagram, I was introduced and connected with many of the designers and developers who work at Accurat.
Giovanni Magni is currently the head of data visualization at Accurat, and has helped lead some of their most interesting projects. Coincidentally, he also runs one of my favorite data viz accounts on Instagram, Data Glitches. We met over a video call a few months ago, and I have loved Giovanni’s design-centric perspective on data visualization. He responded to my questions about data viz and design through a barrage of Slack messages.
Sarah Kay: How did you find your way into data viz? Was there a chart or piece of work that inspired you?
Giovanni: I don’t think I have a specific project or work that led me into data viz but there were two moments back when I was studying that got me closer to the field.
First, I spent a semester in Brussels, Belgium, where I studied at the LUCA School of Arts. One of the classes was held by professor Johan Van Looveren and the assignment was to redesign the periodic table of elements. That was (without knowing at the time) a pure exercise of information design and my first contact with the discipline. Professor Van Looveren continues to push the university towards information design.
Once back in Italy, I attended the Density Design class at Politecnico di Milano, taught by professor Paolo Ciuccarelli, who is now a professor of design at Northeastern University. That use of data viz got me fascinated and I started getting into the field more and more, until I decided that it was what I wanted to do.
SK: How did you end up at Accurat?
G: After graduating I tried to keep working in academia and applied for a PhD but unfortunately didn’t get the scholarship, so I tried different things.
I collaborated with the Density Design Research Lab for a little while. I made some freelance projects while also working with the Statistical Division of the Municipality of Milan.
Accurat was looking for interns, so I applied and I got in. I still remember the call interview I had with Gabriele (one of the founding partners of the studio). Since then I’ve been working as a data viz designer.
SK: You started as an INTERN!?! That’s amazing! I’m really impressed!
G: I did! Good old times.
SK: Tell me a bit about your current role at Accurat. What are your broader responsibilities? And what does your day to day look like?
G: I’m currently the Head of Data Visualisation. It’s my responsibility to guide the dedicated data viz team. I also oversee the overall Accurat production of visual models and data viz applications in order to guarantee our standard of quality and innovation.
One thing that I really like is that I have the chance to speak for Accurat: to bring our idea of design out of the studio to other companies and events. My days are mainly split in two main activities, active projects that I work on and managing the team.
SK: You’ve told me before that a data viz is only as good as the questions that are being asked. Can you explain why you think that is? And what questions are you interested in right now? What questions aren’t being asked?
G: Data visualisation should solve problems: We are not artists, we are designers. Don’t get me wrong, this does not mean that we don’t need art in projects. Accurat’s work has clearly shown the importance of aesthetics in data visualisation.
What I’m saying is that clients and people have problems, and they come to us for solutions. Problems might be clear, specific, and business oriented (“We want to make more money with our platform”) or experimental and ethical (“Let’s find a new way to convey this message”), but we have to focus on finding the solutions that help solve the problem.
The questions that aren’t being asked are: What was the context? What was the goal? I see a lot of people debating on how charts might or might not respect all the statistical rules of charts and interpretability of visual models, while really few understand that every single rule—even the strongest ones— could be shaped and adapted if it will make the project more effective. It is the same with complexity, but that’s another story.
SK: How can data viz practitioners become better at asking questions?
G: We should take as much as we can from positive design critique. One of the things that we are trying to do in Accurat is foster self-critique and take internal moments when our work could be discussed and our choices doubted.
Thanks to the work and approach of Amin Al Hazwani, our director of design operations, we are working on making this possible. It takes time to point out the multiple possible solutions for a problem, to discuss them, to understand the context and the designer’s choices. Hopefully this will make the team more oriented to self-critique and our projects will keep on getting better and better.
SK: There aren’t many data viz studios in the world, and I think Accurat one of the largest ones. How does Accurat’s structure of designers and developers work? What does that collaboration look like? Are there unique outcomes from that setup?
G: When I started working at Accurat there was only one developer, but right now there are more developers than designers. I think it’s the normal evolution of things since we mainly work with digital products and applications. Even when we create something different (such as exhibitions), due to the nature of our work we always have a more technical side in all of the projects where we need the support of coders.
We try to work together from the beginning; it’s not always simple but it’s important, and it even happens that great ideas also come from developers.
Recently a thing happened that made me glad of our process. We were working on a project with basic chart models (bar charts, line charts, etc.) and within a discussion about a specific task the developers explained to the designers the best logic given their experience with similar projects and due to typical chart behaviour. The roles were reversed and that was brilliant! We have been able to create a data visualisation culture on the whole team, and I’m super proud of this.
SK: You approach data viz from a design perspective. Do you think data viz designers need to know how to code?
G: I am a designer with a very little background in coding. I can generally understand concepts when I talk with developers and I can build my own little prototypes (with the worst code ever written).
That said, I am a fan of designers who are able to code. I think it’s helpful and gives you a lot of potential, but at the same time I will always want designers who cannot code on our team.
What I have noticed is that when you have an idea on how things are done, you automatically are influenced by the limits of the technology that you are going to use, since you know what you can and can’t do.
Designers who can’t code are the ones bringing ideas to the table that might seem impossible to realise. But those are the ideas that make developers push and find solutions, hopefully creating innovation. I like seeing the expression on a developer’s face when somebody proposes something totally unexpected.
SK: There’s a growing interest in data viz coming from designers. How can an interested designer break into the data viz field and/or acquire data viz skills?
G: I think there are an overall set of rules and practices that need to be known, and designers can start from books and online repositories to learn them. A good way to get familiar with the topics is to follow the right people on Twitter, where I think most of data viz designers still publish and discuss projects.
I think a lot will come with experience, so my best suggestion is to try to do stuff, use your own data, force yourself to work with data and deal with the issues that arise. I still think that working on real projects is the best way to get confident.
SK: You have a little hobby of collecting data glitches. Why the glitches? What do you find interesting about them?
G: I think they are beautiful and I started collecting them as a tribute to data visualisation. I find them interesting, especially once out of context, they have some sort of aesthetic that fascinates me. I enjoy this concept of unexpected beauty.
I find it interesting that this could happen, that when working with data we can generate elegant and beautiful images with mistakes, breaking things out, without even having that as a goal in mind.
SK: You’ve had the chance to work with Giorgia for quite a while. Is there an insight you can share that you learned from working with her?
G: I’ve always been impressed by her ability to motivate every single choice she makes. You might think that having an approach that looks less statistical than others implies an attention only to aesthetic but it’s quite the opposite, every single element is positioned and customised accordingly to choices she makes.
One other thing that I love about Giorgia is her humanity. She has reached a point in her career when she is clearly one of the top designers out there, but she keeps on caring for people, is totally open and always available for anybody who might need advice or help. She is a good person, and in these days that is relevant. It still happens that she asks for my feedback on what she’s doing—this makes me laugh sometimes.
SK: I totally can see that humanity in her too! She’s one of the most down-to-earth and kind humans I know.
G: Indeed!
SK: Do you worry about people misinterpreting data? Or worry about data collection transparency and handling that happens before visualization?
G: Totally, but most of the time you can’t do a lot. Typically you get on a project once the data has already been collected by the client or it’s structured in a way you can’t change. Our job it’s to always keep this in mind and to design accordingly.
SK: What topics in data viz fascinate you right now? Where do you think there is room for the data viz community to grow? What do you see on the horizon that is interesting?
G: It looks like data viz has become mainstream, everybody talks about it, everybody does it. Given the attention we have on the field, I think it’s interesting to see where we can go. Maybe people are ready to experiment with new visual models, maybe we are confident enough with charts and numbers that we can now push forward and innovate in the field. I really don’t know where we will end up, but I’m sure this is a quite interesting decade to work in this field.
