For years there has been the belief that cities have to unveil new websites in the same fashion they present their plazas or parks: in one swift reveal. From the start, every amenity has to be ready, each piece of content added and the design — with the exception of minor garnishments — should be set and final.
Often, for city officials applying such tactics, an unwanted discovery was imminent. They might see to their dismay that the work wasn’t done. Site features didn’t work the way they envisioned. Navigation buttons went awry. The maladies just continued until leaders found themselves crossing fingers and throwing cash at the problems.
That was then, this is now, and luckily things are gradually changing. Cities have learned that websites — like any other piece of 21st-century tech — are always evolving. It’s why they’re embracing the idea of beta testing, whose advocates argue that the best thing to do before launching a website is to, in fact, launch a website.
Beta testing reshapes expectations for a “final product,” and instead, seeds the idea of an evolving service. Cities publish a prototype, gather public feedback, make adjustments, gather more feedback and so on, until a site is ready for rollout. The practice has gained appeal in Philadelphia, Los Angeles and other cities that have employed it to enhance services while avoiding launch day pitfalls.
To offer a few reasons why cities might consider such a strategy, Boston, a city of nearly 700,000 people, and West Carrollton, Ohio, a city of more than 13,000, join the federal digital service 18F in demonstrating three advantages of beta launches.
In Boston, the city’s foray into beta testing began in May 2015 when it sought to co-develop its new site with residents. The IT department built a blog to chronicle activities and sliced development into six phases. The first decided city partners for the project, the second entailed user research and the third included building the actual prototype. These phases were followed by the beta launch for feedback, the official relaunch of the site on Boston
.gov and a final phase dubbed “continued iteration” that theoretically never ends.
What might shock traditional government types was an inherent lack of hard deadlines. Considering the fluid nature of the project, Boston only set a rough timeline. This eased worry. Staff could take the right amount of time for quality development, while at same time, the flexible design took any causes for complaint and transformed them into constructive input via surveys, emails and open office hours. For government, the process was procedurally audacious, relatively foreign, and yet at the same time, safe and uniquely productive.
“The mentality we have around the website is that it’s not something that you build, and then launch, and then orphan or leave over time,” said Lauren Lockwood, Boston’s chief digital officer. “It’s something that should always be improved and maintained.”
The build-and-adapt philosophy lifts internal tensions as well. Lockwood said that from a project management standpoint, communicating function and design with a tangible product, as opposed to ideas and sketches, clarified a host of misconceptions at the start.
“It’s been really helpful internally as we migrate [data and content], because anybody in my position is going to be thinking about how to work with the teams internally to get all this done,” Lockwood said. “It also takes some of the pressure off once the site does go live because we’re not talking about putting things online for the first time.”
18F is an outfit in the federal government famously known for its capacity to solve problems. The group — which helps agencies build, buy and share software — aims to take complex policies and technologies and rework them into human-friendly services. 18F did this in 2014 when it helped save the Affordable Care Act’s health insurance platform, HealthCare.gov. It did it again with improvements to the Department of Homeland Security’s Immigration Application site, and applied the same sort of know-how at the U.S. Department of Veterans Affairs when it helped streamline the way veterans receive benefits.
18F members credit their knack for problem solving to modern design principles, concepts such as human-centered design and agile development — a collection of concepts that assess human behavior and embrace iterative methodologies to create user-friendly tech. In a statement to Government Technology, the group said beta testing is a fundamental component in these guiding practices. Similar to Boston, 18F jump-starts a Web project through a discovery stage that harnesses user research to target needs, which leads to an initial product via an alpha stage for solution testing. Shortly thereafter the group pilots a beta site to collect input for improvements before an official launch.
“Our design process starts by figuring out the right problem to solve and making sure that’s what we’re tackling,” 18F said. “Once we have an understanding of the right problem to solve, we use iterative testing, which includes any metrics analysis or usability testing, [and] we do this to continually validate that we are solving the right problem as best as we can.”
18F is in the process of relaunching the Federal Election Commission site, FEC.gov. This project, designed to improve citizens’ access to federal campaign finance data, serves as a prime example of 18F’s beta work. At the FEC, 18F did a deep dive into the world of arcane financial regulations and outdated data systems to create beta.FEC.gov. The site is set to replace its predecessor with a bevy of new features, additions the group is honing with the experience of its partners at the FEC and average citizens.
“When 18F started talking to stakeholders and users, we learned that users were often worried they hadn’t found the right information, all the information, or the most up-to-date information when navigating the site,” 18F’s Leah Bannon and Noah Manger wrote in a joint post.
Now with a fresh look, the beta site is off to a promising start with a suite of simple tools that specifically answer user concerns. Just a few of the features include a new navigation menu that lets visitors search for election finance information by candidate name, a geographic location, or within the executive and legislative branches of government.
While 18F doesn’t typically develop city sites, it does serve a variety of offices and departments on civic projects. When attempting to answer problems, 18F’s advice to cities is to gather not just feedback, but also representative feedback that reflects the local population. This can be done affordably too. Often high-tech algorithms or enterprise analytics aren’t required. Feedback can come from highly accessible platforms like Google Analytics, a stack of emails, a site widget prompting reviews, or just plain observation and note-taking.
“We perform both in-person and distributed testing,” 18F said. “The beauty of technology is that it allows us to both see the user in action with a product, and allows us to see and/or hear what’s happening with the user at the same time.”
West Carrollton is not New York City, Chicago or San Francisco. Like most cities and towns across the nation, West Carrollton does not boast an armada of IT workers or pretend to be the next Silicon Valley. It’s a small Ohio town with a close-knit community of suburban homes, wide lawns and spurts of interwoven roads. So when officials in West Carrollton sought to redesign their website at the start of the year, they approached beta testing as a way to build with community support.
Erika Mattingly, West Carrollton’s public relations coordinator, said the redesign began with guidance from the developers at ProudCity, a gov tech startup focused on digital services. Since much of the platform was already worked out, the beta launch was a brief two weeks starting on Feb. 9, out of which, Mattingly said staff and the ProudCity design team uncovered a surprising number of fixes and final adjustments.
“It was just a nice way to get feedback,” said Mattingly. “Not only from our residents and people using the website, but people internally within the city, who were essentially able to say ‘Hey, what about this?’ or suggest little things we may not have thought of when we first got started.”
The city placed a link to its beta site on its old home page, created a banner on the new site to draw input, while City Manager Brad Townsend reached out to key stakeholders for their impressions. The outreach yielded technical improvements to icons, content and search terms, but more significantly, Mattingly said the beta testing brought about a measure of community endorsement. Residents saw the public outreach effort on the site, found out through word of mouth, and then saw their notes and suggestions turned into site updates.
“It was a much easier process than it ever had been before when we had revamped our website,” Mattingly said. “I’m glad that we took the leap and tried something different.”