The coronavirus variant tracker, learning on the job, and a membership program

A few weeks ago I pitched an idea to my team at Axios to do a coronavirus variant tracker. I had read several news stories about these new variants sweeping the country, and wanted to know more, so I went searching one night, and the only variant tracker I could find was incredibly complex. It was very impressive in the amount of information provided, but was long, difficult to find the info you needed, and was clearly written for a scientist. I thought we could do better, so I set out to design something that would be easily approachble for lay people, and would summarize the key info quickly, and at a glance.

Now that it’s all published, I’m really happy with how it turned out and the excellent reception it’s received. But it was definitely a stressful experience to build on such a tight deadline. Here’s a few of the things that went right (and wrong) leading to the project being successful.

• When I first pitched this, I was planning to use some data the CDC was publishing about case numbers in each state. But it turned out that for a few reasons, it wasn’t great data. Luckily, right around that time, the outbreak.info API came out with some major upgrades that gave me exactly the data I needed in a super easy and accessible API. This was the first time that someone saved my ass on this project

• A few days before the deadline, I was trying to figure out how to make this area chart that would show overall prevalence of the variants in the U.S. over time. The API had this data, but it would mean me coding up another chart in Javascript (not super easy) and the data needed a good bit of wrangling, which I didn’t know how to do easily in Javascript. I could have figured these things out, but at this point I was already neck-deep in last-minute issues and had a giant pile of TODOs, so I just couldn’t handle doing a new thing at this point. Luckily, my coworker Sara was testing this new cool thing called ‘git scrapers’ (blog post upcoming, hopefully). Turns out I could write a quick R script to pull the data and wrangle it into the format I need, then run that on a schedule and save it to a database that would only update if the data changed. Then I could just make the area chart in datawrapper and link it to that live-updating data, and voila! The second time that someone saved my ass on this project.

• When I started this project, we didn’t have a way to publish projects on Axios outside of embedding in our normal CMS. Obviously for this project, that wasn’t what I wanted. The thought of squishing my beautiful project into a 700 pixel wide iframe made me want to cry. I had brought up the issue with our tech team, but they were initially not super keen on slotting in a last-minute request like this. Then, when I tested my prototype in our CMS embed, things got even worse… turns out we have an issue with certain types of embeds that was causing the iframe to size improperly and cut off my content halfway down. We were so close to saying “we have to nix this project, it just won’t work”, and then at the last minute, Max Lever from our tech team reached out and basically just got it done. He figured out a way for us to publish this on our website as a separate full-width page and for the third time, completely saved my ass.

I’ve been plugging away on the information design course I’m developing, which is now officially called “The Glamour of Graphics”. So if you enjoyed my talk at rstudio::conf a couple years ago, you’ll want to keep an eye open for this course. It’s a full course in design, focused on data visualization, with all the stuff I wished I could’ve fit into my talk but couldn’t.

I haven’t talked about it much, but for the past year I’ve been part of a small group of fellow dataviz designers and freelancers. We help each other work through freelancer stuff, give advice, feedback, share resources, celebrate each others wins, and just generalyl help each other to be better at whatever we do. Everyone involved is incredibly talented and lovely, and getting to be a part of such a special community was without a doubt the best thing about last year for me.

Now, some of us have decided that we want to band together and offer a membership scheme for others that are aspiring to freelance or change careers into data visualization. We’re still discussing the precise offerings, structure, costs, etc. but the gist of it is that you’d get access to a whole group of people that have been at the top of their field doing dataviz in various industries and as freelancers. We have experts in design, UX, coding, storytelling, and we’d be producing heaps of awesome content, answering your questions, and helping you to meet your goals.

We’re planning to launch with a small set of beta testers, where in exchange for your feedback, you’ll get a special beta-tester discounted subscription rate which you can keep forever once we launch officially. If this sounds like something you’re interested in, please respond to this email and let me know, what would you want out of a membership program like this? What would make it something worth paying money for? And would you potentially be interested in being a beta-tester?