Election Officials, Spotlights / November 9, 2017

Denver Publishes Election Data in Interactive Chart

This story was featured in our ELECTricity newsletter in November 2017. Sign up to receive more success stories from election offices across the country. And how is your election office using technology to run excellent elections? Tell us about it by emailing [email protected] — we’d love to share your story!


If you follow the Denver Elections Division on Twitter, you know that it’s an election office that loves to tell its story to the community. In its Tweets, the Division profiles voters happily inserting ballots into drop boxes, shares behind-the-scenes photos of office operations, and shows how running elections is a year-round effort using the informative (and funny) hashtag #WhatWeDoBetweenElections.

You can also see this desire to enlighten the public in the way the office publishes election data.

By using interactive Microsoft Power BI dashboards, the election team is able to provide useful information to its audience instead of just raw numbers. It’s a great example of how an office can deliver data in a format that’s user friendly for data geeks and the general public alike.

Publishing data proactively

In recent years, Colorado has become synonymous with election innovation, and Denver has been at the forefront. A growing, energetic city, Denver is home to over 453,000 registered voters who cast ballots by mail, at drop boxes placed throughout the city, and at convenient Voter Service and Polling Center locations.

It’s easy to vote in Denver, and the Denver Elections Division also makes it easy to use the data that it produces.

That’s because, according to Communications Specialist Joe Szuszwalak, the Division understands data as a resource, and he and his colleagues make data publicly available so more people can benefit from it.

Joe Szuszwalak, Alton Dillard, and Amelia McClain
Communications Team members Joe Szuszwalak, Alton Dillard, and Amelia McClain. Photo courtesy of the Denver Elections Division.

“Our team is very data driven,” says Joe. “Our policies and processes are outcomes from thorough analysis of related customer and internal data. By publishing these dashboards publicly, we hope to provide our customers with the same opportunities to track trends and gather information on the elections process.”

Now, anyone who’s curious can explore data on things like registration, turnout, and election costs in a series of dashboards on Denver’s Election and Voter Data web page.

Staff members collaborated to make the dashboards. Data Architect Paul Huntsburger provided the data, Ballot Operations Coordinator Stuart Clubb and GIS Analyst Steven Sharp created the visualizations, and the Division’s Communications Team tackled web publishing.

From the perspective of the staff, a major benefit of the dashboards is the ability to push data out to the community instead of making people ask for it, explains Joe.

“By proactively publishing this type of information, we have reduced data requests to our office and increased transparency and engagement with our community partners and customers.”

Making data interactive

Denver’s dashboards are actually built on two different data visualization platforms: Tableau and Microsoft Power BI. The reason is that the Division sees benefits of both.

Joe says that the data team likes to use Tableau — which we highlighted in our May 2017 spotlight on Greenwich, Connecticut—to visualize geospatial data, while it favors Power BI for making charts.

“Power BI is great for making data displays of non-geospatial data, with robust chart and graph capabilities,” Joe observes. In addition, as a Microsoft product, it can be included in an election office’s licensing agreement, and for people who use a lot of Microsoft products, it may be more intuitive than Tableau.

Joe Szuszwalak and Stu Clubb
Joe with Ballot Operations Coordinator Stu Clubb, who created the Power BI visualizations. Photo courtesy of the Denver Elections Division.

To make a dashboard in Power BI, you start by building a report. You connect to a dataset, choose which data fields to display, and then select the type of visualizations you want.

You can put several visualizations together, and once you’re happy with it, you publish it to the Power BI site, where you’ll transform your report into a dashboard for public access. As the final step, you can embed your finished dashboard into your website.

On the user’s end, what’s most important about the dashboards is their interactivity. “The ability to interact with data in a visual manner,” Joe explains, “provides unique opportunities to ‘dig deeper’ into the data.”

For instance, let’s say you’re curious about election costs for Presidential Elections. When you go to the election costs dashboard, it presents you with costs for 17 races. But, to see only Presidential Elections, you just click on a checkbox to filter the data.

Now, you can immediately see three data points on Presidential Election costs, and the chart shows costs decreased from 2008 to 2012 and then 2016. Nice job, Denver!

Chart shows costs for Presidential Elections: $6.51 in 2008, $4.61 in 2012, $4.15 in 2016
By filtering the chart, you can compare apples to apples.

The interactivity also highlights relationships between variables.

“By selecting certain options, one can begin to build a better picture of how some variables may affect others,” Joe explains. “And making these connections in a visual manner instead of looking at raw data really solidifies the comprehension of the information being conveyed.”

For example, if you want to see how many registered voters are both unaffiliated and inactive, you can find the number (22,463) with just one or two clicks.

In the party chart, Unaffiliated is selected, while in the status chart, Inactive is selected, showing 22,463 voters are both Unaffiliated and inactive
Selecting more than one variable allows you to drill down into data.

Clearly, finding answers like these would be much more difficult without the Power BI dashboards.

Achieving benefits for everyone

Since launching the dashboards, Joe and his colleagues have heard encouraging feedback from members of the public as well as other election offices.

“Our audience values the convenience of having the data published in a visual, interactive format that is continually updated,” he says. “We have also heard from other offices making use of the statewide data that we have available.”

All told, the dashboards have been beneficial for everyone. The election team has achieved greater transparency and public service while reducing the need for data requests. Meanwhile, community members can access data quickly and easily, finding answers to their questions without needing to wade through spreadsheets.

Heather Heyer with Joe Szuszwalak
Voting Process Administrator Heather Heyer with Joe at a community event. Photo courtesy of the Denver Elections Division.

Joe acknowledges that making data visualizations can be complicated, but he encourages election officials to try making a dashboard with Power BI. He says that since most officials already know their data, all that’s needed is to get familiar with the platform.

“Power BI is simply an additional tool available to make the data more consumable to our customers,” he suggests. “And there are many resources (YouTube tutorials, etc.) available on the internet to help anyone looking to get started with the Power BI application.”

In fact, the Power BI site has plenty of good content to help you get started, including this run-though of basic concepts for Power BI. And for people who want to check out the tool, there’s a free version of the platform that includes much of the functionality of the pro version.

In addition, if you have questions about Power BI or about Denver’s dashboards, you can email Joe Szuszwalak and he’ll do his best to steer you in the right direction.


What solutions has your office come up with to provide data in user-friendly formats? Share your best practices by emailing [email protected].