Seeing and Understanding Data

A conversation with author and Tableau Technical Evangelist, Andy Cotgreave

Andy Cotgreave is co-author of The Big Book of Dashboards, Technical Evangelist at Tableau, and host of If Data Could Talk.


Technically, what is a Technical Evangelist?

TCE18 Square.jpg

In a nutshell, I get paid to make people enthusiastic about our mission to help people see and understand data. I fell in love with data communication and Tableau around 13 years ago, and have been lucky to have a job that allows me to share my passion. I write booksbylines, make videos, and do presentations. I think that being able to do this job well is a result of me, somehow, having a great intersection of skills. Anyone seeking a job like this needs to be experienced and skilled in their field. They also need to be an excellent communicator, able to write and speak. One needs to be creative, too. Creativity is important in order to make new connections, generate new ideas, and think of novel ways to promote your passion.

Hear about upcoming conversations:

Creativity is important in order to make new connections, generate new ideas, and think of novel ways to promote your passion.
— Andy Cotgreave

How much farther can we innovate with dashboard technology? What's coming next?

Dashboards evolve, just like any design-driven artifact. Cutting edge dashboards from 2012 look extremely dated now. Even some of the dashboards in our book could do with an update! This is partly driven by design: tastes change, so what looks appealing today will look clunky soon. A big tension I see today is how to move dashboard design to mobile. The real estate is completely different. Desktop dashboards need to fit one landscape screen. On mobile, dashboards can be tall and thin: people don’t mind scrolling. Deciding the right level of information density for mobile is also vital. Designers need to consider how to pare back the experience to make things easily understandable, without removing too much information. Interactivity can be more challenging on a mobile dashboard, too. As tools become more powerful, “dashboard technology” moves towards complex application design. The UX considerations cannot be overlooked.

 
RM4_0262.jpg
 

Is it enough to have expertise in Tableau to be competent at data visualization? What other skill sets (e.g., knowledge of programming languages, additional software, etc.) are required?

For this answer, I’ll point you to three important articles/tweets. Andy Kirk’s Eight Hats of Data Visualization was one of the most impactful posts I read when developing my own skills. To be good at data storytelling, you need need diverse skills, from engineering, to artist, to journalist, to project manager. Data visualization is an art as much as a science. Alberto Cairo tweeted that if you think writing, design and speaking are “soft skills” it shows you do not understand the field of data analytics. Finally, Kim Rees described how data visualization is a means to convey an opinion or an argument. Each of those examples drive home the fact that “knowing how to use software” is a scratch on the surface of being great at data visualization.

...”knowing how to use software” is a scratch on the surface of being great at data visualization.
— Andy Cotgreave

What's the most challenging concept to communicate to clients about visualizing their data?

GSKO2016_Seattle_Day2_Gibbons-873.jpg

First it can be hard for people to know what they actually want. This is through no fault of theirs. Projects often start with “We want to be able to see our data.” Adopting the famous 5 Whys technique is valuable: WHY do they want to see the data? I find that if you ask why around five times, you’ll actually understand what it is they want. It can be hard to get people to understand the tension between complexity and clarity. There is a temptation to try and answer all the questions – but then things become too complicated. So one might try and keep things simple – but then there’s not enough detail available for any constructive insight. I have never found a “right” answer to the complexity/clarity conundrum: it depends on each client’s needs.

Will there be a "Little Book of Dashboards"? I.e., how is mobile changing best practices for design?

We have no plans yet, but we do keep seeing many new dashboards appearing every few months!


In general, do you think the public has been well served by covid-related data visualizations? Has it done more to educate, or to mislead? 

I think covid-19 has been one of the most data-driven stories of our lives. To understand the spread of the disease, people have had to learn to understand complex concepts such as log scales and exponential growth. They have had to grapple with the every day uncertainty in data. They have had conversations based on numbers. Media organisations have largely been very sensitive in the way they have presented data, and been highly skilled in how they have told data stories. I do believe covid-19 has been a positive for the field of data visualization.

To learn more about Andy Cotgreave, visit gravyanecdote.com.