Deconstructing Visualization: An Interview with Andy Kirk
Data visualization and infographics are a hot commodity, but viral graphics don’t simply materialize fully formed from the digital ether. Companies that don’t understand their goals, or fail to plan accordingly, can sink many hours and many dollars into a project that never quite comes together. What are the key steps project teams have to take before and during the visualization project? What are the critical questions designers should ask of clients? Should teams aim for a firm objective or an emergent process?
Enter Andy Kirk.
Kirk is a UK-based data visualization designer and consultant who leads training sessions across Europe and the United States. His blog, Visualising Data, is a huge repository of information, including monthly curated lists of the best data visualizations on the Internet.
Kirk recently took time out of his busy schedule to answer some questions for Mindjet about just what goes into a good data visualization.
MJ: OK, so a team sits down and decides it wants to make an infographic as part of a project. What should their first step be, before anything else?
Kirk: I’ll throw in two steps, merged into one. Before they do any design work or even get into the data they need to identify what the reason is they are creating the visualisation in the first place. What is its purpose? Determining the reason, the purpose, the motivation for creating the piece in the first place is vital as it sets the design tone and the term of reference for what follows creatively. Alongside this the designer needs to be completely clear about the parameters that exist around and within the project. Who is the intended audience, what is the desired format, what timescales are they working with, what are the resources, capabilities and technical restraints etc. All these factors have a huge influence on the shape and direction of the visualisation design.
MJ: In recent years, the technology available has made the possibilities for data viz nearly limited only to one’s imagination. This is great, but it can also be daunting or distracting. What advice do you have for designers trying to get a grip on all of it?
Kirk: On the technology side, it is a very difficult challenge harnessing and developing an expertise across all key tools and programming languages. In that sense, the field is very fragmented with no single killer tool to achieve all stages of data extraction, cleaning, preparing, sketching, designing, executing and publishing. So any visualisation designer needs to be able to build up a broad range of technical skills to cover these vital stages before they can realise their own creative ambitions. With regards to the absence of limits to what one can achieve if they have these skills, that is probably true. The advice for designers overwhelmed by this prospect is simply to let the problem context and the data do most of the talking. Understanding the purpose of your project, developing and refining your specific curiosities and analytical angles and then immersing yourself in the data will give you the best chance to reduce the vast array of choices you may have at the outset and eventually give you a clear path down which you can execute your designs.
MJ: What are some good resources for designers and teams looking to expand more into data visualization?
Kirk: I’m going to take a very self-indulgent path towards answering this one and mention a few things I do on my blog. Firstly, I have a live collection of the “essential visualisation resources” which includes details of all the important tools, software and programming environments for creating visualisations. Soon it will also include details of the key data handling tools and the places to get hold of data. It also has links to over 60 books that I believe are the most influential texts for my journey through the subject. Also, on a monthly basis, I collate and curate the best of the visualisation web. This is the online content published or discovered during the previous calendar month which I believe represents the most important, interesting and influential content, whether it be new designs, articles, developments, new sites etc.
MJ: What’s the most common question you’re asked at your training sessions and seminars, and what is your answer?
Kirk: Two questions really. First is “what is the most important tool I need to learn” for which there is, unfortunately, no simple or indeed single neat answer, as explained by the diversity of tools and programming environments out there. The second one is a question that seems to be what people arrive at the sessions with and is “can you teach me how to make my charts sexier”. Once again the answer is multi-faceted and often disappointing because visualisation is not about creating sexy charts from unsexy ones, it is about deploying the most efficient, effective and impactive visual variables into an intuitive design that will both attract the potential reader and help them feel smarter about a given subject. Sexy graphics can be an outcome, but only as a by-product of an effective design process, not the objective in itself.
MJ: What question do you wish you were asked more?
Kirk: “I work for a very cash rich organisation who would like you to come and train our organisation about visualisation, would that be possible to arrange?” That’s probably the best question which I’d like to hear more frequently, otherwise I can promise you the delegates are very forthcoming in asking the full spectrum of interesting, challenging and worthwhile questions.
MJ: What’s the most common mistake teams make? Like, if you could parachute into a meeting and say, “Whatever you do, don’t …” How would you finish that sentence?
Kirk: “…use terms like ‘cool charts’ or ‘fancy graphics’ to describe your design requirements. Let’s quickly clear up the mess I’ve made with my parachute landing, sorry about the coffee spill on your new tie, and let’s start from the top with your requirements, the purpose, your parameters, your key analytical questions, your data set etc…”
MJ: You highlight the best of data visualization on your website, but what about the not-so-good stuff? Is there a trend or type of data viz you’re just plain sick and tired of seeing?
Kirk: Well for a while I did throw on to the site brief criticisms of pieces that I really hated, but I’ve now tried to move away towards focusing on the positive. The thing is you never really know or are able to appreciate the brief and the conditions a designer was working under. This means a superficial, face-value only judgment can be unfair and uninformed. Sometimes you do see work coming from the likes of Google or other big data/tech organisations and you think “they should know better and be able to do better”. So to a certain extent I think they’re fair game, but I now prefer to leave the mature critical analysis to the likes of Robert Kosara’s Eager Eyes website, Kaiser Fung’s Junk Charts and Bryan Connor’s The Why Axis as they set the bar for forensic, balanced evaluation of data visualisation designs. If you were to push me though, I would have to say I personally hate the junk infographics that are so saturating the web these days.
MJ: Is there anything I should have asked, but didn’t?
Kirk: You could have asked me where my training sessions are next scheduled for! My answer would have been, “thanks for asking, I’m shortly going to be inviting my readers to inform me of where they would like me to take my training courses next. Once I establish sufficient quantities and clusters of interest I will arrange my next set of dates and locations.”