For those of us who follow social media on topics like analytics and data science, it seems like every day brings a newly popular data visualization. When well designed, these graphics are engaging and catch the eye, amongst so much text or photos of cats. But, I’m hearing that such stand-out is even more important in businesses.
We’ve written before about the need for Data Science teams, how to build them and the growing demand for specialist roles like Data Engineers. It seems as these teams mature a growing diversity of individual specialisms is emerging. As well as roles to provide consultancy to different business areas, we are hearing that Data Scientists are being complemented by Data Storytellers or Data Journalists.
But, in line with that growing focus in social media, the skillset we are hearing is growing in demand is that of a Data Visualizer or Data Artist.
Now, I don’t know how long ago your schooldays were, but my memories of those who chose to take art are very different from the kids who chose maths or science. Memories of abundant paint pots, clay and generally a creative mess – all feel a long way away from increasingly technology focused maths & science.
So, can the two coexist? From what we’ve seen, I’d like to suggest two options for filling this gap. Both have worked in different contexts.
Bringing an Arty-type into the team
Even before new libraries (in Python or R) made sophisticated interactive data visualizations more common, a few businesses saw their need for an eye to the aesthetics.
Listening to the leaders of analytics teams and some of their internal customers, it’s apparent that many organizations have been drowning in information. Multiple MI reports, BI dashboards, plus presentations from analytics teams. Despite all this data, too many leaders still express that it’s hard to “see the wood for the trees”. It sounds like more data does not equal more insight.
Seeing the need for better presentation of data, with succinct and compelling graphics, some analytics leaders have turned to graphic designers. Whether a new hire or borrowed from an existing Digital team, these artists have managed to improve the visual effectiveness of reports. Some appear to favor infographics, whilst other produce compelling graphs or even embedded videos.
When a good relationship is built with this designer, their help is really valued by the analysts or Data Scientists we place. After sweating to prove causation or develop an improved targeting model, most analysts crave impact and recognition. Having their output look so professional, engaging and quick to digest – sounds like it really helps achieve that response from senior leaders.
So, for some organizations, a separate role works well, but at the expense of doing nothing to improve the data visualization skills of the Data Science team. Given growing demand for these skills, that may limit your analysts.
Develop Analysts into Data Artists
The other positive response, we’ve seen in client teams, is to train up analysts or data scientists in data visualization. When this is done well, we hear very positive reports from both the analysts themselves and their managers.
It is perhaps a misconception that arty-types are very different from science-nerds. If you think about it, data analytics is a creative endeavor. Some of the most successful analysts or Data Scientists we have placed over the years have been very creative in their problem solving. More than knowledge of a particular programming language or package, a mindset that loves analytical problems and creating new opportunities, seems to distinguish the most effective analysts.
Given this demand, it is encouraging to see organizations with whom MBN partner are helping to fill this training need.
MBN has also really enjoyed partnering with The Data Lab and Scottish universities. Both to help develop the next generation of Data Scientists and to encourage active cooperation between employers and academia; ensuring students graduate with the skills employers need. So, it was encouraging to see The Data Lab host a Data Visualization masterclass to help both communities:
Using these resources and many more, a number of Analytics leaders have managed to develop data visualizers within their own teams. Some analysts have reported being surprised how much they’ve enjoyed this opportunity to release their inner artist and develop their visual skills. From what we’ve heard at MBN, we would certainly encourage this approach where possible.
What do your Data Scientists need to learn to become Artists?
So, if you’re an Analytics or Data Science leader who sees the need for improved Data Visualization skills in your team, where should you start? Here are some of the tips we have heard from our contacts:
For those who analysts are using software applications, like Tableau, SAS or IBM Analytics, further training on making best use of the existing functionality can be worthwhile.
If your analysts are coding in R or Python, it can be worth getting up-to-speed with the add-on libraries dedicated to data visualization capabilities. Apparently ggpolt/ggplot2, matplotlib, d3, rjsplot or bokeh really help.
Encourage personal development in this area. Some leaders have apparently provided time for analysts to develop these skills, others have run competitions or established internal communities. Recognition sounds to go a long way, as well as making it more fun for your team.
Consider training courses like those listed above. Time out of the office to learn & experiment is certainly valued by many analysts.
Having said that, more than one client has remarked that it’s not all about technical training. It seems design principles, aesthetics, learning what works for your organization and even understanding the science of the human eye – all matter. In fact, it may be more important for your analysts to learn the principles of good data visualization design, rather than just diving into coding skills.
So, as a final tip, here are some of the experts in this field whose books and websites we are hearing have helped analysts “get it”:
Keep up the good work, it is appreciated
Personally, I just want to commend the progress that has been made in this area. As someone who remembers green screens and ‘portable computers’ that came in bags resembling hiking gear, it’s amazing to see how far the industry has come.
Well done to all those who are producing data visualizations. Helping the rest of us to understand data faster & more accurately. Plus, making the world of Data Science just that bit more beautiful. Enjoy your art, we at MBN appreciate it!