Better Technology, By Design

4 approaches to changing the way digital services are designed in government

usability testing.jpg
 

By Rachael Carson

At Civilla, we’re on a mission to change the way our public-serving institutions work through human-centered design. In this effort, we’ve partnered with leaders in the Michigan Department of Health and Human Services (MDHHS) and their IT vendor to improve their technology. Together we’ve focused on redesigning the online portal that over one million Michigan residents use each year to access critical services like food assistance and healthcare.

Soon after we began the redesign, it became clear that in order to make significant improvements we would need to re-evaluate the traditional software development process. As we iterated, we gained new insights. Here are some of the lessons we learned along the way…

 

 
 

From conference rooms to kitchen tables

User voices transformed team culture

When we started working on redesigning online enrollment, most of the work occurred in conference rooms. Designers, project managers, program specialists, policy experts, and technologists all huddled to make decisions about the best options for site navigation or about which functionality should be prioritized. The voices of the people who would be be using the service were almost entirely absent.

Not long into the redesign, MDHHS leaders asked Civilla to implement a new way of working that made more space for the user’s voice. In response, we hosted a training and invited their technology team to learn some of the research methods that are central to our work. Soon after, they started joining us for user interviews, where we listened to people’s stories side-by-side to ensure that the site was designed to meet real needs.

Over time, the State’s technology team has become an extension of our research unit. Together, we’ve sat down with hundreds of individuals across the state. The process has shifted the culture of the team from one that was driven by personal opinions to one that is guided by user insights. Now, when the team huddles in a conference room we hear conversations start with “The user said…” instead of “I think…”.

 
fireflies.jpg
 
 

From development to prototyping

User feedback allowed changes to be made before code was written

Traditionally, MDHHS used a linear approach to build technology: they’d draft a proof of concept, articulate design specifications and technical requirements, and ship the product to the engineering team. Once code was written and tested, the final product would launch in one big bang. Unfortunately, feedback on the design often arrived too late to be incorporated, leaving applicants and staff frustrated.

During the redesign, our team worked with the State’s technology team to adopt a new way of working. Rather than building the product linearly, we would build in much shorter, faster cycles.

We began by designing wireframes and mockups, light prototypes that demonstrated how the new site would work, look, and feel. We brought each prototype into the field to test with applicants. User feedback enabled the team to rapidly iterate on the designs before a single line of code was written.

The idea of validating (or invalidating) ideas based on users’ feedback caught on. More than 18 months later, this approach has enabled the technology team to incorporate hundreds of hours of feedback on the software design.

 
daisy.jpg
 
 

From adding to deleting

The power of design often lies in knowing where to simplify

Technology conversations at MDHHS often focus on the promise of the new: new features, new pages, new functionality, and new tools. Like many institutions, the agency generates funding and momentum by adopting new ideas to address existing problems.

But new is not always better. Sometimes, less is more.

During the redesign, Civilla worked with MDHHS and their technology team to determine what should be built and, more importantly, what should not be built based on user feedback. For example, the team recently spent time brainstorming ways that the State could use text messages to communicate with applicants. We tested various ideas and heard consistently that users only wanted to receive text messages from MDHHS that contained case-critical information. This feedback informed design decisions that would ultimately strengthen the communication channel overall.

With time, the team has developed a stronger point of view about how the new enrollment portal can maximize value for its users. In turn, the team is more equipped to identify opportunities to simplify, streamline, and delete.

 
shell.png
 
 

From quantitative to mixed-methods research

Integrating data from multiple sources produced insights that neither type of data could’ve had on its own

In MDHHS, quantitative data is the dominant form of input for measuring user behavior. Quantitative data is very helpful in explaining what happened, but it often requires qualitative data to explain why it happened.

Once the new enrollment portal launched, Civilla worked with the technology team to pair quantitative data with qualitative insights to better understand where the site was falling short and how to improve it.

For example, data showed that very few people were signing up for text message notifications within the online portal. We spent time with applicants who were using the site to understand why. Their stories helped us see that people were in fact interested in receiving text message notifications, but opting in was unintuitive and required users to click more than seven times.

After identifying the problem, the technology team moved the feature onto the dashboard so that people could sign up with one click. Almost overnight, the number of people subscribing to text messages tripled. In this case and others, mixed-methods research provided the team with meaningful insights that continue to guide the improvement of the site.


Through our work in Michigan, our team has begun to set a blueprint for how human-centered design can be used to change the way government technology is built:

  • Building technology around user needs creates valuable products.

  • Testing technology in iterative cycles allows teams to adapt quickly to feedback.

  • Knowing what to build and what to delete saves time and money.

  • And pairing hard data with qualitative insights helps guide improvements over time.

Taken together, these shifts provide a powerful way for leaders to make more informed, thoughtful decisions on behalf of the people they serve.

 
 
bees.jpg
Adam Selzer