How We Built It: Getting Spooky with Splunk Dashboards

Dashboards are not just tools for businesses and other organizations to monitor and respond to their data, but can be a method of storytelling. All of our data has the potential to be crafted into compelling narratives, which can easily be accomplished with the help of Dashboard Studio’s customizable formats and advanced visualization tools. 

We can take a series of disparate datasets and bring them together in one place if they share a common theme — in this case, Halloween. We aggregated open-source data into Splunk about candy production and box office records for scary movies, as well as some fun mock data we created for testing purposes about trick-or-treating at certain addresses and what candy they gave out.

How we organize this data on a dashboard requires deliberate and thoughtful decisions in and of itself. We already covered this in the first post of our series on dashboard best practices here. The goal of this dashboard is to be eye-catching and memorable. We want the theme to match the season, and style our dashboard accordingly. Our second blog post in our best practices series tackles color choice, but in our case, we already know what we want to do for our dashboard. Halloween is typically associated with black and orange, with other sharply contrasting colors like green and purple associated too.

We want to tell a story about Halloween in 2022. With the data, we’ve learned that candy production is up (good for sweet tooths and dentists alike), scary movie box office sales are down (horror movie haters rejoice), and there’s a winner for the hottest haunted house on the block for giving out the most candy. Here’s what the dashboard looks like with just the data visualizations laid out together:

To create a sense of grouping for each of these datasets, we added semi-transparent dark rectangles to the dashboard canvas and organized the visualizations by the dataset used. This creates a hierarchy of the information, with each grouping titled to give the viewer context as to what information is where.

For some of the more advanced configurations on the table, we used the UI to do everything we needed. Formatting numbers, dynamic background coloring for cells, and changing the ranges to match our data.

We replaced the series colors for each of the column charts with a mix of orange, purple, and green to give a more festive feel: orange for the box office revenues on the left and purple for the more candy-oriented data on the right. To make it extra clear that one is a historical trend and the other a future forecast, we made the annual trend line overlaid on candy production each month of the year in orange, while the ML-powered prediction range for the trick-or-treat data was made green.

For the final touches, we added images to help tie together the dashboard. There are tons of great resources online for backgrounds and icons, such as Pexels or the Figma Community, and your choice of them can help make or break the look of your dashboard. With the wrong choices, a Halloween dashboard can easily veer too into middle-school-poster territory, but with the right ones, it helps elevate your dashboard into a state of perfect polish. We tried to keep it to one related image per group of visualizations, with the exception being the little potion bottle for the Trick-or-Treat dataset since it contained the most visualizations, and helped bring more orange into an otherwise purple and green heavy area of visuals. 

For the fun title text, we used an open-source font that we rendered outside of Splunk, then re-uploaded it as an image. This is a workaround that can be easily done for any text that might need to follow branding guidelines. We increased the height of our dashboard to accommodate the title text and a better view of the background, filling some of the extra real estate in the corner with a spiderweb. 

There we have it! A fun, festive dashboard dedicated to Halloween, built using concepts that can easily be applied to many other situations that are not as seasonal. We brought together data from different sources together to create a visual story that can capture the attention of an audience, enabled by the easy editing capabilities available in Splunk Dashboard Studio. If you’re ready to take your dashboarding skills to the next level, check out the rest of our blogs and stay tuned for even more!

Helpful Resources

If you're new to Dashboard Studio, here are some resources to get you started: 

• Dashboard Studio Tutorial
• Dashboard Studio Tech Talk
• Splunk Dashboard Studio Documentation
• Splunk Ideas - Dashboard Studio for feature or enhancement Requests
• Examples Hub - Find the Examples Hub from the Dashboards page in Search & Reporting
• Splunk Community - Dashboards & Visualizations for questions 

This article was co-authored by Matteo Palarchio, Product Manager for Dashboards and Lizzy Li, Senior Product Manager for Dashboards.