Why Collaborative Vizzes Rule
Why Collaborative Vizzes
Rule
Recently, I had the
opportunity to collaborate on a viz with the incredibly talented Lindsay Betzendahl (@ZenDollData). The viz was
extremely well-received by the dataviz community and ultimately became Viz of
the Day. But the result of the viz was not nearly as rewarding as the
process. This is our story, told in the
form of answers to questions, and written independently from each other.
How did the collaboration come to be?
Kevin: Tableau has changed my life. That
might sound like a huge overstatement, but it’s true. I talk to my identical
twin brother EVERY day, where historically it may have been once a month. I now work as a Tableau Developer under one
of the greats, Jeff Shaffer. And I’ve made a lot of really amazing friends
since starting with Tableau a little over a year ago. Tableau has literally
changed my life.
One of the best
friends I’ve made is Lindsay Betzendahl. We chat often, swap family stories,
talk about our jobs, but 99% of the conversations revolve around Tableau. One
afternoon, she started asking my thoughts on viz collaborations. She mentioned
the incredible collaboration
between Spencer Baucke
and Chantilly Jaggernauth
and started a rapid-fire sequence of questions. How did they do it? How did
they work on it at the same time? Who did the data collection? How did they go about collecting data? How
did they share the workbook? How long did it take? What were the challenges? We
probably discussed it for about 20 minutes or so and then I abruptly asked,
“Want to do a collaboration?” Her answer was a resounding “YES”.
Lindsay: It was sort of funny, in retrospect,
how this particular collaboration began. I didn't actually reach out to Kevin
initially with the plan to ask him to collaborate, but that definitely didn't
mean I didn't want to collaborate! I was probably subconsciously hoping he'd
agree to do a viz together since I really respect his work and find we have
some similar techniques. However, at the core, I was curious about how two
people would complete a public collaboration and the process. I even told Kevin
that my goal was to document the process so that others would have a map for
how to do a collaboration viz.
See, Kevin is
probably my closest friend in the Tableau Community. He was one of the very
first people I started to communicate with regularly before the 2018 Tableau
Conference. We spent a ton of time hanging out and discussing all things
Tableau at the conference together. This gave me plenty of insight into his
personality, his style, his strengths, and weaknesses. I was also aware of what
things we had in common and where we were likely to differ in both data and
design decisions. Honestly, this aided in our decision to do a viz together. I
think we both knew that we could handle whatever happened in the process and we
sort of knew what we were getting into. (There are a number of questions I
think you should answer before embarking on a collaboration with someone, but
those will appear in post #2 about this topic, which will be coming soon.)
So, when we started
talking, and he mentioned that perhaps we could do the collaboration together,
I already had answered all the "required" questions in my head and it
was an easy decision to say yes. There was no negotiation or thought, it just
happened.
How did you decide on what to visualize?
Kevin: As Lindsay said, as soon as we
decided that we would collaborate on a viz, we began talking about topics. Although our original conversation took about
20 minutes, the discussion on a subject took about 20 seconds (see the messages
below).
Lindsay: I'm not sure that the experience we had in choosing a topic for a collaboration viz will be overly common. We lucked out that we both were interested in animals, and I was lucky that he liked the cheetah idea. For others, I imagine this process could take a little more time, unless you have an idea that you intentionally want to collaborate with someone, you have similar interests, or a similar goal (perhaps a community project in which a data set or a topic is already chosen). Regardless, I'd spend some time here thinking about the subject matter if it doesn't come naturally. You want to have something you both enjoy and want to ultimately research and visualize together! I mean, you will be talking about it together for likely a few weeks!
Lindsay: I'm not sure that the experience we had in choosing a topic for a collaboration viz will be overly common. We lucked out that we both were interested in animals, and I was lucky that he liked the cheetah idea. For others, I imagine this process could take a little more time, unless you have an idea that you intentionally want to collaborate with someone, you have similar interests, or a similar goal (perhaps a community project in which a data set or a topic is already chosen). Regardless, I'd spend some time here thinking about the subject matter if it doesn't come naturally. You want to have something you both enjoy and want to ultimately research and visualize together! I mean, you will be talking about it together for likely a few weeks!
Truth be told,
aside from a very general topic idea (literally just the animal), I had no idea
what we were going to visualize or if we would even find data to be successful.
I think because Kevin and I happen to both be highly energetic and people
dedicated to success, we were likely going to make it happen one way or
another. And, thankfully, we did.
How did you collect the data?
Kevin: The original idea was that we would
independently search for data sets and other inspiration then come back to the
table to share what we found. However, about an hour later, Lindsay sent me a Google doc with about a dozen
links and another dozen infographics…I had done nothing, like absolutely
nothing. I must give 100% of the credit to Lindsay. When we talk about finding
data and inspiration; she did it all.
Lindsay: Since we had such an intense
discussion, I think we were both super excited. I immediately
started searching for data that would work in a visualization. This meant I was
looking for trend data, for the reasons why cheetahs were going extinct (what
were the possible causes), how many were left, what made cheetahs unique (so
why should you care that they are going extinct), etc.
My thought process
was that we couldn't go forward with a good viz if we couldn't actually find
data to make the viz! I won't deny it, this part took a lot of effort and time.
I was not just finding data, but finding data from credible sources, data that
linked up together (such as time frames), and various data points that would
cohesively tell a story.
In addition to the
data, I also researched other animal infographics and posters to use as ideas.
These images aided in our initial discussions about the design and overall
“feel” we would eventually decide on.
I think Kevin was a
little surprised how many resources and ideas I compiled right away, but I
think it helped get us up and running with our design ideas. Researching data
on cheetahs was something I really enjoyed, so I was more than happy to do the
"boring" work upfront. I created a Google doc so that I could keep
track of all the resources I found.
How did you get started?
Kevin: In professional vizzing, for me, it’s
all about exploring the data. But in personal vizzing, I typically like to
start with a design idea. So I didn’t spend a ton of time looking at the data
sources Lindsay provided (sorry Lindsay). I immediately went to Tableau and
started to design the header. She had found an incredible infographic about
elephants where the elephant was sort of protruding from the right-hand side of
the graphic. I loved that idea and decided to quickly mock one up for our
cheetah viz. This mockup looked like the following:
To be quite honest,
I don’t think she liked it very much. Her exact words were “cool”. But nothing else. Nothing. Nada. This image would be the source of many
discussions moving forward.
Lindsay: When we started, we worked
independently just brainstorming ideas and sharing them with each other. As
mentioned, the elephant infographic was pivotal in our design decision to make
our viz resemble an educational poster with plenty of text, engaging and clever
charts, plus some hidden interactive features. Part of making the viz look
educational, for me, was ensuring there were well designed and supportive maps
showing the cheetah’s reduced range. My early focus was to develop the maps to
look perfect.
How did you start building the viz?
Lindsay: In the beginning, I was highly
focused on creating the perfect maps and laying out the title elements. I tend
to build from top down and I had sketched out a few ideas before putting the
elements in Tableau.
I have always
admired the mapping abilities of Jonni Walker and wanted to emulate a spherical-looking map he had
created in his Safe Haven viz.
The main difference was that I also wanted it to be interactive.
I utilized several
techniques to create these maps. First, I used the Interworks polygon tool to
create the custom areas. Next, I created an image in PowerPoint with a
transparent circle in the middle to create the illusion of a circle-shaped map
and a second image of a circle with a shadow. Lastly, I needed the map to be
interactive (which it was not when the image was placed over it) so I came up
with the idea of floating another map on top of the circle image but make it
100% transparent!
This four-layer map
approach worked perfectly, and I was so excited that it was all I focused on.
Ironically, a few days later Simon Beaumont shared a viz with the same
technique and he wrote a blog post about it. I love how these
"discoveries" can happen in parallel.
So, while my head was down creating my map visions during week one, Kevin was sending me images
of cheetahs, ideas about chart types, and data to search for; oh, and paw
prints. He was really getting us focused on other key elements to the viz and
sharing some creative concepts.
In the beginning,
we were brainstorming, essentially, tossing ideas back and forth and feeling
out what the other person liked/disliked. I actually had the first draft of the
dashboard for five days before sending a first draft to him on February
20th. When I sent it to him for the
first time, it looked like this:
Kevin:
After I found an image and title that I liked, I started “doodling”. I
created a couple of chart ideas that I thought might work and I sent the
following to Lindsay:
How did you truly start putting the viz together?
Lindsay: We were throwing around a lot of
ideas, but in order to really ensure that we were in sync, we decided to jump
on a phone call to outline the sections of the visualization based on the data
collected and our ideas throughout the week.
That call was
crucial to our success as it allowed us to truly get on the same page. There
was one thing in particular that happened during that call that likely wouldn't
have happened otherwise. As we were talking through ideas, specifically Kevin's
"doodles" of the change in population. He wasn’t sure about it, but I
liked the concept, just needed to tweak the shape. I remembered Kizley Benedict had done a Makeover Monday visualization
that used 100 points in a circle. I immediately thought that it would be a
great resource to enhance the population chart that Kevin had already designed.
In the end, he came up with the idea of making each point a cheetah spot, which
I loved! Below is the original idea and the end result.
Kevin: After our phone call, we continued to
pass the workbook back and forth as new visuals were added. Lindsay started
building some charts and placing them into a dashboard. She passed it along to
me and I changed everything around and added a few charts. I passed it back to
her and she changed everything around. Truth is, we are both type A, both
opinionated when it comes to our work, both a bit OCD, and definitely both
perfectionists.
We had a few
discussions or perhaps I should call a spade a spade – arguments – about
certain aspects of the charts and the design. Nothing was heated in any way,
but we both thought we were right. It didn’t take me long (and I got the same
impression from Lindsay) that we had to make some concessions. At one point, I
said to her something along the lines of “you are really good at this stuff and
if we are going to collaborate together, I have to fully trust that you know
what you are talking about and your decisions will end up for the good.” Boy,
I’m glad I quickly came to that realization because, it’s clear, she knows what
she’s talking about.
Lindsay: We are both perfectionists and a bit
obsessive compulsive (so make note of this in case you ever want to collaborate
with either of us!). And because of this, we did have few “discussions” that
were somewhat difficult. But it was a refreshing exercise in having these
difficult design discussions. Plus, it was an excellent lesson in why iterating
and taking feedback can be so helpful!
For example, on
February 22nd, I sent back another version of the viz with a custom title. He
turned it around on the 24th and changed that title. I shared about my original
inspiration for the title and provided reasoning as to why it should this way.
He didn’t necessarily agree, but he trusted my decision. During that time, he
also had ideas about moving the main image, changing the background color of
the dashboard, reducing the size of the maps for more space, etc. I took those
edits and modified them again. During this particular editing process, I
realized we needed to use Kevin's original font for the title. I think he was
surprised when I said "we are using your original title idea."
I knew how
important this viz needed to represent both of our styles and those unique
talents each of us bring to the table. Kevin is known for awesome font choices
and catchy titles, so it was only fitting that his original idea was what we
used in the final viz. You can clearly see how powerful this iterative process
can be.
How did you move on from there to continue building the viz?
Kevin: From here, we exchanged hundreds of
messages back and forth and even had a couple more phone calls. We continued to
build charts and update the dashboard, then pass the data source and workbook
back to the other person. This back and forth continued for a couple of weeks.
I will tell you,
passing the data source back and forth was a bit tricky. In fact, at one point, we almost lost a
significant amount of data because we sent the wrong file and it was
overwritten by the other person. If I had to do it again, I’d probably go with
a Google sheet containing all the data so that we did not have to email it back
and forth.
Lindsay: After the first two weeks, we
certainly hit our stride. It became easier to think about the layout and data
elements to the viz as we were on the same page about the overall vision. Plus,
we had learned how to navigate communicating about our opinions about the viz a
well.
At this point, we
had already determined the sections of the viz and we sort of divided up who
would tackle each part first and then the other person would edit and provide
feedback.
After we got the
"fast facts" and "fading numbers" sections relatively
cleaned up, we moved on to the bottom half. Kevin added in a really cool line
down the center that gave a neat separation effect. I love how it wrapped
around the circle map! He also added in those paw prints, which I wasn't
originally keen on, but grew to like as the visualization took shape.
Finally, I added in
a radial bar chart to the map which showed additional data regarding the cheetah's
range in Southern Africa and another radial chart used to highlight key factors
contributing to the cheetah's low numbers in the wild. Kevin added in more
data about breeding, cleaned up my key factors radial chart with crisp icons
and tooltip details, he added a curvy timeline, more data about the illegal pet
trade, and an interactive paw print with details about how to help protect and
preserve cheetahs.
The last week of
work really came together relatively seamlessly. I'd say the first two weeks
were a lot of work, but things fell into place towards the end. Here are the last
few iterations. There were many little changes including small movements at the
bottom as we were cleaning it up and ensuring everything important would fit.
We changed some header names, resized various elements, checked spelling and
grammar, and ensured everything had a consistent look and feel.
What did you think about the final viz?
Kevin: When the viz was complete, I have to say that I thought it
looked beautiful and told an incredible story.
We agreed that we
would both publish it to our own Tableau Public pages. We also agreed that we
would unhide our vizzes and tweet them out the following morning at the exact
same time – 8:00 AM.
I remember the
excitement when we knew we would be sharing it via Twitter the following
morning. I knew we had something. When we tweeted it, we received the most
wonderful feedback from the community. It was a hit! Thank you to all that
commented on it and favorited it. And the cherry on top was the fact that it
was selected as Viz of the Day later that week.
I must say
something here, something very important. Prior to this viz, Lindsay had
several VOTDs…I believe the number was four. I had just one VOTD, one that I
got just one month into using Tableau. Through previous conversations, she knew
that I really wanted another one. I may have not said it publicly, but I really
wanted one badly. When we learned about receiving VOTD, she said that she hoped
they used mine as their link. Ultimately, they did. I don’t know if they just
randomly picked one or if she insisted to Tableau that they use mine, but I can
guarantee that if they asked her, she would have told them to use mine. Thank
you, Lindsay.
Lindsay: Honestly, I could not have been
happier with the results, and I could not have asked for a better partner to
collaborate on a visualization project with. It was so fun to work with Kevin.
Aside from the fact that he is a dear friend, he is wonderfully talented and
creative.
Receiving a VOTD
for a collaboration was immensely rewarding. It's exciting to share that moment
with someone who has contributed so much to the community and has been a powerhouse
of positive enthusiasm for so many. So, thank you, Kevin, for embarking on this
cheetah journey with me! Maybe there is another collaboration in our future.
What are your overall thoughts on the collaboration?
Kevin & Lindsay: If you are going to do
a collaboration, make sure to trust the other person’s skills, ideas, and
designs. You can hold to your opinions, but you must understand that the other
person also knows what they are doing and has a plan, even if it doesn’t
perfectly intersect with yours. Trust that person wholly and try to put that
person before yourself. In doing so, I believe you’ll enjoy the process much
more than you would otherwise.
So, there you have
it. If you’ve never done a collaboration with someone else, we both suggest that
you stop reading now, message someone, and get one started. It is an incredibly
rewarding experience and one that you get to share with someone else.
Here is our final
viz!
You can find this
viz on both of our public pages: Lindsay's cheetah viz and Kevin's cheetah viz.
The Slideshow
Kevin: Before we go, I’d like to share with
you a slideshow of our progression. Unfortunately, we did not capture as many
stages as we could have as this was a bit of an afterthought, but you can
certainly see the iterations on our work. Enjoy and thanks for reading.
Lindsay: Please tag Kevin and I if you do a
collaboration viz, we would love to see what other community members visualize
together!
Cheers,
Lindsay & Kevin
Kevin Flerlage, May 1,
2019 | Twitter |
LinkedIn | Tableau Public
great viz... thx for sharing this
ReplyDelete