Tanya Shapiro is a self-described "part data geek, part business detective." As an independent data consultant, she helps businesses unlock the value of their data.

Shapiro specializes in data visualization and dashboard design, streamlining reports, delivering business intelligence solutions, and more.  She applies a data product management mindset to help clients tackle a variety of problems within their analytics pipeline.

After attending last year’s Observable Insight conference, Shapiro’s interest in using Observable to work with data was piqued. Since then, she’s created some of the most unique and popular notebooks on Observable.

Shapiro’s notebooks on Observable are popular, in this author’s opinion, because they do a few things. First, Shapiro uses Observable to its full potential — leveraging the ability to pull in multiple data sources fast and get from exploration to viz that tells a compelling story in a few clicks. Second, her notebooks are…fun. Read on to learn more.

For Tanya Shapiro’s map of RuPaul's Drag Queens around the world above, data was scraped from Drag Race Wiki. Shapiro says, “It's wild to see how RuPaul's footprint grows as the show adds different international franchises. There are over 350 queens on this map from more than 10 franchises!”

How did you get started with data visualization? How did you start using Observable?

I came into the data visualization world in large part thanks to the online R community. Over the past few years, I’ve met some exceptionally talented people — both online and in real life — including one of my data viz heroes, Allison Horst.

My curiosity about Observable was already piqued after attending the Observable Insight conference. I distinctly remember having that “oh wow” moment when Maya Gans started talking about polyglot workflows with R and Observable.

I was inspired by many of the talks at the event, and motivated to find different ways to build my own interactive visuals. Then there was this fortuitous overlap of events. Allison and I connected the following week to hang out and geek out about data viz. We talked about our respective journeys, the power of learning through community, and the different tools we work with. 

With her encouragement, I started experimenting with Observable. Since we both share an R background, her ability to translate JavaScript and Observable concepts has been very helpful. She has a gift for making complex data topics accessible to beginners. I still bug her with my Observable questions today!

When I found out about Observable notebooks, I was ecstatic — I didn’t have to worry about the setup — I could start experimenting directly in the notebook and see the outputs instantly.

- Tanya Shapiro, Independent Data Consultant

I see you have a Drag Queens Netwerk Diagram on Observable, as well as on your site, built using other tools. Can you talk a bit about the process of working on Observable vs. other tools? 

Yes, I’m a big fan of RuPaul’s Drag Race! I think it’s always easier to learn new things when you’re working with a topic you’re interested in.

For Tanya Shapiro’s Drag Queens: Netwerk Diagram above, she modified code from Guillermo Garcia's notebook and Raven Gao's notebook to experiment with network diagrams. Netwerk Diagram shows drag queens who were featured on different franchises and seasons of the TV reality competition series, RuPaul's Drag Race.

Last year, I used it as an opportunity to learn more about web scraping and data wrangling with R (rvest). I might be biased because I’ve been working with R for a few years, but I think data wrangling and pre-processing in R is easy thanks to libraries like dplyr. I also created a couple of visuals in R to explore the data, including a static network diagram with ggplot2 to show how all the contestants were connected throughout the different seasons.

This year I’m pushing myself to create more interactive visual content. I’ve tried a few different libraries in R to bring some of my visuals to life, but I started noticing a theme: everything comes back to JavaScript. A lot of the interactive visual libraries in R are actually JavaScript wrappers. If I wanted to customize something beyond the wrapper’s ability, I’d end up having to hack a solution by injecting more JavaScript in R. Which led me to the conclusion - I should probably try these ideas out with ACTUAL JavaScript!

I think one of the biggest hurdles for a JavaScript beginner to get started is the setup process (eg. download an IDE, install node.js, create html and css files, etc). When I found out about Observable notebooks, I was ecstatic — I didn’t have to worry about the setup — I could start experimenting directly in the notebook and see the outputs instantly.

One of the first ideas I wanted to try out was a revamped version of my Drag Race network diagram. I recycled the data I wrangled from my R project — and was immediately impressed by how easy it was to upload csv files in Observable.

As a JavaScript padawan, I found the D3 coding part behind the graphic a little bit trickier. But luckily I had a lot of help via the Observable community. I’ve learned a lot from exploring other people’s notebooks, and when I’m stuck, there’s always someone on the  Observable Slack channel willing to help me troubleshoot.

I think it’s always easier to learn new things when you’re working with a topic you’re interested in.

- Tanya Shapiro, Independent Data Consultant

For the grand finale, I tested my graphic in R Quarto. I often use Observable notebooks as my data viz playground. When I’m satisfied with the end product, I like translating it to Quarto, which supports Observable and R code. I can do the data wrangling in R, pass the data into an Observable code chunk to create beautiful interactive graphics, and then knit it all together to produce the html file. I’m excited to play around with these workflows more. R and Observable is easily my favorite dynamic duo.

You've partnered with groups like R Ladies and Women Who Code to share your knowledge of data viz design and to help others develop their technical skills. What advice do you have for people interested in getting started with data viz?

We all start somewhere, the trick is to just start… and then keep going! I’d also add don’t be afraid to experiment and explore new ideas even if you think they’re awful, you’ll learn something new in the process, I promise.

The advice I give myself a lot these days now that I’m picking up a new coding language: be patient with yourself. Progress isn’t always linear, I’m far off from where I want to be, but I’m falling in love with the journey.

How can people learn more about your work and ask questions about working with you?

Learn more about my work and reach out about working with me here. Thank you!