We recently launched a website on the Impacts of COVID-19 on the Georgia Economy, gacovid19.org, in which students from Georgia Tech are covering a number of different industries and producing weekly reports as well as a podcast.
Inforest’s contribution to the project, in addition to the website, was to recommend and support a program for producing interactive graphs and charts. It turns out that there are a number of different tools available, including programming API’s, desktop programs and Web based services.
We ultimately chose Google Data Studio (GDS) as the primary graphing tool for the project. Most Web developers and marketers are probably familiar with GDS since it can be used to pull in data from Google Analytics and combine with other data sources to provide fancy reports to your clients.
Key advantages for GDS include that it is free to use, allows users to create graphs without programming knowledge, (although the interface takes a bit of getting used to) and integrates with a variety of data sources, especially the online spreadsheet tool, Google Sheets.
It’s really the Google Sheets/GDS combo that sealed the deal for us. GDS can be set to periodically refresh its data set, so as new data is added to the Google Sheet, your GDS graph or chart is automatically updated/redrawn. This makes maintaining graphs really easy.
GDS also does a good job working with responsive web design techniques. If you design your graph layouts to fit within a certain aspect ratio, such as 16:9 or 4:3, you can successfully resize your graphs to fit within mobile and desktop layouts. Older tools, such as Tableau, have responsive layouts, but they are somewhat clunky and don’t transition on the fly when the screen is resized.
Key disadvantages of GDS compared to other programs and services is a lack of map displays, limited image thumbnails and it can be a bit buggy. GDS is limited in presenting data in map formats. For example, showing COVID-19 statistics in a Georgia county map is not yet possible in GDS and had to be done in Tableau.
GDS also does not provide a thumbnail preview as the graph loads, and the static preview thumbnails provided are of very small size and quality. Thus, in order to provide preview images for the homepage, we had to write our own node.js script to generate thumbnails with Chromium. We’ve also noted that sometimes graphs will generate errors on Safari browsers, supposedly from bugs cropping up as the GDS software develops.
However, all together Google Data Studio has worked well for for the gacovid19.org website. All of the researchers were able to develop their own graphs without needing a whole lot of support from Inforest and the graphs look and perform relatively well. While GDS is still first and foremost a tool for extending Google Analytics, it proved a good choice for integrating interactive graphs in a website.