• December 17, 2015
• Civic Engagement

This paper is part of a series published by Data-Smart City Solutions, a project of the Ash Center at Harvard Kennedy School. The series explores data-related facets of civic engagement in today’s cities.

INTRODUCTION

Code for America is a San Francisco-based non-profit that believes government can work for the people, by the people, in the 21^st century. Over the last five years, we’ve partnered with dozens of local governments around the country to improve service delivery—in the health, safety and justice, and economic development areas—through technology. We also focus on improving the public’s relationship with government by creating innovative spaces and channels (sometimes digital) where government and residents can meet.

From this work, we’ve developed a strong perspective on what makes “21^st century civic engagement.” In order for governments to be open, responsive, and engaging, community members must feel they are active participants in building government, and that their participation can influence decisions about issues that affect them. We believe that technology can be a powerful tool for community engagement. It can help local government expand their reach to a broad cross-section of the community (and clearly see who is being left out), help people understand the issues at hand, encourage productive actions, and demonstrate how those actions are creating positive outcomes.

Typically, Code for America partners with cities through our flagship Fellowship program. The Fellowship embeds a team of software developers, designers, and product managers into a city department for eleven months to help improve a city service. However, in January 2015, we piloted a new type of “Senior Fellowship” program, a six-month partnership where one alumni Fellow worked with the City to develop new approaches and tools to support deep and broad community engagement.

During this project, the City of Boulder and Code for America partnered on “Housing Boulder,” the community engagement process that would inform Boulder’s 2015/2016 Housing Action Plan. While this case study documents our work on a housing-related project, we believe our engagement tactics are relevant to a much broader audience. As a result, this case study also offers a series of recommendations to help governments begin using 21^st-century civic engagement strategies that creatively combine in-person and digital channels.

PARTNERSHIP OVERVIEW

Background

Boulder, CO is a city in southeast Colorado located at the foothills of the Rocky Mountains. It has around 100,000 residents, is a fairly liberal-leaning city, and is the home of the state's largest public university, the University of Colorado. Because of its high rankings in health, education, art, food and beverages, cycling infrastructure, and proximity to outdoor recreation, Boulder is a very desirable place to live. All of these factors have contributed to increased growth in the City, and challenges around keeping it an affordable place to live. Like many U.S. cities, housing is a particularly sensitive issue and central to the affordability issues. For Boulder, key housing issues include:[1]

• A shrinking economic middle class (for the City of Boulder, this is a $65,000 to $150,000 annual household income).
• Detached single-family homes are increasingly only affordable to the wealthy.
• Attached condos and apartments are more affordable, but less appealing to families.
• Almost 60 percent of Boulder workers live in surrounding communities.
• Shifting demographics, especially aging, and changes in housing preferences.
• A need for more diverse housing options in existing residential areas.
• Lack of a strategy for guiding redevelopment in the City.

Housing Boulder was launched in 2013 as a comprehensive housing strategy to help determine community priorities for the expansion and preservation of diverse affordable housing choices in Boulder. Two central characteristics defined the Housing Boulder effort: broad community outreach to ensure an open, balanced, and fair process; and convenient and meaningful opportunities for residents, those who work in Boulder, and other interested parties to inform the planning process.

The City of Boulder had put together a comprehensive engagement strategy that focused on both in-person and virtual engagement strategies including large events, small working groups, posting information to the city’s website, and using tools and surveys to gather community feedback. While they were executing on this strategy, the City wasn’t completely satisfied with the results of the outreach.

The City had a commitment to involve members of the community that may not have participated in city planning processes in the past, and engaged with Code for America to complement the existing approach by testing new practices and tools to broaden their reach and ensure that underrepresented audiences were also participating in the dialogue.

Project Team

Code for America hired Becky Boone, a 2014 Code for America fellow in Denver, as a full-time Senior Fellow in Residence to live in Boulder and work with the city for six months. Her work was supplemented with strategic oversight and support from staff at the Code for America headquarters in San Francisco, CA.

In the City of Boulder, work was primarily concentrated in the City’s Communications Department and Department of Planning, Housing and Sustainability. The team also consisted of the Housing Boulder Process Subcommittee, which included members of the City Council and Planning Board. The Housing Boulder Process Subcommitee also partnered with working groups, which were comprised of members of the public interested in the City’s housing issues.

Code for America’s Approach

Over the past five years, Code for America has worked with local governments across the country and developed a number of tools designed to increase participation and engagement. We have seen that there are five key practices of effective 21^st-century community engagement, codified as the Code for America Engagement Standard:

1. Reach: Defining the constituency you are trying to reach, with an emphasis on identifying those whose voices aren’t already represented.
2. Information: Providing relevant information that is easy to find and understand, and speaking with an authentic voice.
3. Spaces and Channels: Making use of a diversity of spaces, both online and offline, that meet people where they are.
4. Productive Actions: Identifying clear, concrete, and meaningful actions residents can take to reach desired outcomes.
5. Useful Feedback Loops: Making sure the public understands the productive impact of their participation, and that their actions have value.

The project was organized around this set of practices, and tailored to meet the needs of the City and residents of Boulder. To understand these needs and set a baseline to measure impact, Code for America conducted user research with the community and government staff.

User Research & Project Goals

Over the month of January 2015, Code for America conducted interviews with staff and community members. In speaking with the Community Planning & Sustainability Department, we observed that:

• While the department frequently asked for public feedback, they didn’t regularly let the community know how their participation and feedback led to productive outcomes.
• The department expressed concern that they were only hearing a select few voices that participate in public meetings, open houses, and media platforms; and these voices may not be representative of the “community at large.”
• The information that was being provided on the project website was difficult to navigate and understand.

From the community interviews, we saw that:

• The community didn’t have a good understanding of how planning processes work, or the function of City Council.
• People did not trust the existing process, they felt that the decisions were already made, and that their participation would not change or influence the outcomes.
• Many people said they were too busy to participate in in-person meetings.

Taking the project scope and user research into account, the City of Boulder and Code for America worked together to define the goals of the project, which were to:

• Stimulate interactive dialogue and collaboration with a cross-section of the community.
• Increase participation and move from one-way communication to multi-way dialogue.
• Create a new approach to community engagement that can be replicated on other projects and in other communities.

While our partnership with the City of Boulder specifically focused on building inclusive community engagement into a housing strategy, we believe these practices apply to community engagement efforts more broadly. The following are the approaches and tools we used to meet the City’s goals, and a set of recommendations for governments to deepen their own community engagement efforts.

I. EXPANDING REACH

Our first step was to define the constituent groups who were currently participating, and set goals for increasing engagement with the population that was not participating. For Code for America, participation meant actively engaging in Boulder’s housing conversation through in-person or online channels.

Because the City was not formally collecting demographic data about meeting participants, we needed to develop a baseline. We collected this data using clicker polls and surveys, and found that majority of the people participating in the in-person events were homeowners between the ages of 56 and 74.

However, using a tool called Census Reporter, which allows users to easily sort through U.S. Census data, we got a better understanding of the demographics of the City. We found that:

• 65% of the population of Boulder is under 40.
• 52% of the population of Boulder rents their home.

This discrepancy in representation suggested a need to target younger residents in the City. The Pew Research Center’s internet user demographics indicate that 97% of adults ages 18-29 years old, and 93% of adults ages 30-49 years old use the internet, we felt that driving online engagement would help us reach the City’s underrepresented demographic: renters under the age of 40.

Outcome

At the end of the partnership, public participation in online channels began to better reflect the City’s demographics. Thirty-six percent of participants in online forums were under the age of 36, while only 21% of those who attended the in-person meetings were in the same age category. In addition, 45% of those who participated online were renters, compared to 28% at in-person meetings.

Recommendations to Expand Your Reach

1. Understand who makes up your community

The goal of your engagement strategy should be to reach a representative sample of your community, not just to get more people involved. You can use data from the Census Bureau to understand the demographic makeup of your community and identify who you need to reach. Free tools like Census Reporter make it easy to find and understand this information.

2. Measure your effectiveness

If you've collected data about who is taking part in your process so far, compare these figures against Census data to understand who isn’t participating yet. Use Census data to set a baseline and check whether your results are over or under that figure. This will show which groups are currently under or overrepresented.

If you aren't measuring the reach of your engagement efforts yet, here are a few simple ways to find out who is taking part in existing forums and channels:

• Sticky dot exercise. During a public meeting, label poster boards with categories for key demographics you are interested in, such as age ranges, education levels, or gender. Have dot stickers available for people to place on the boards to show the groups they identify with.

• Paper surveys. At the end of a public meeting, conduct an attendee satisfaction survey that asks participants to provide basic demographic information and feedback on the quality and effectiveness of the meeting. This creates an important feedback loop and helps you measure who you are reaching.
• Clicker surveys. Handheld electronic polling devices, known as “clickers,” can be distributed to audience members and used to gather responses to questions. The meeting leader asks a question of the group, audience members key in their response using the clicker, and the answers can then be instantly displayed to the group. Ask demographic questions to find out who is in the audience at the beginning of the meeting. Ask for reactions to a particular issue before and after the meeting to track how opinions have changed. Clicker surveys and other digital polling services have varying costs, but average around $1,500 for 24 clickers and accessories. There are also options to rent out clickers for specific events. Some of the best clickers actually come from academia, where educators can poll their students during class.

• Online surveys. Online surveys are a quick and easy way to reach new audiences. Be sure to include questions that collect demographic information so you can understand who your survey is reaching. When designing your survey, use a tool that supports mobile-responsive surveys such as Typeform or SurveyGizmo. Many residents now rely on phones or tablets to access the web.

3. Set clear goals

Once you understand who you’re reaching, set concrete goals about how you would like to expand your reach. Communities that are underrepresented in your efforts to date should become priority groups for your future work. Focus on your biggest gaps first.

Well-defined goals state who you want to reach, how much you want to increase participation, and by when. Setting goals like this will allow you to clearly measure your success.

Examples of good goals include:

• By the end of July 2015, increase participation among the targeted constituent group by 50%.
• Achieve representation from people under 40 in at least three city-hosted events by June 2015.
• Before the proposal goes before Council, hold neighborhood meetings in three underrepresented neighborhoods with at least 50 people attending each event.

4. Build relationships with key groups in the community

Regardless of tactics and technology, an effective engagement strategy is built upon a strong foundation of relationships. When trying to engage underrepresented residents, creating partnerships and working with other groups who already engage with those residents is an important step. For example, if you have a goal to engage more residents under age 40, consider working with student organizations or young professional associations.

Take the time to personally meet community leaders, attend their group’s meetings, and show genuine interest in their respective needs. Doing this will help you build a coalition of community groups that will share information and invite their members to participate in your work.

5. Regularly measure your progress

Throughout your process, regularly measure your progress against your goals and the baselines you set at the beginning. If you’re trying new tactics or approaches but not getting the participation you were expecting or need for your goal, be ready to adjust your strategy. Keep track of all your data in one place and make notes about what worked and what didn’t for future projects.

Signs of Success

When your outreach strategy is successful:

• Your participants (both online and in-person) are representative of your community’s geographic, ethnic, age, income, and other demographic distributions.
• You’ve made demographic data collection part of every engagement tactic (public meetings, online surveys, social media, etc.).
• You’re building relationships with local community groups that represent some of your harder to reach demographics.
• You are regularly measuring and checking who you are reaching, and taking steps to improve your reach.

II. PROVIDING RELEVANT AND USEABLE INFORMATION

During our research, a primary theme was that information on the City’s website was hard to find and hard to understand.

The primary access point to information about Housing Boulder was its website, www.HousingBoulder.net. Doing a full-scale revamp wasn’t possible during the six-month partnership, so our goal was to make the content on the existing site as easy to understand and act upon as possible. We followed guidance on content strategy from project advisors from the award-winning Government Digital Service in the UK, whose website GOV.UK is a leader in writing clear and actionable content for their citizens.^[2]

On the www.HousingBoulder.net homepage, we gave the user immediate information about what was going on with the initiative, provided them the opportunity to participate in an online survey and answer questions on Boulder-related housing issues, and gave them three choices: get involved, learn about Boulder’s housing story, and learn about possible housing tools and options. Each of those options took the user to another page where they could learn more.

Using Google Analytics, we analyzed web traffic before and after the revamp. The bounce rate—the percentage of users who enter and then leave rather than continuing to another page—on the Housing Boulder landing page was reduced by 10%. When looking at the bounce rate across devices, we saw the biggest decrease was from mobile phones (22%).

By redesigning the content on the pages, we were also able to reduce the average page load time. Although relatively unchanged for mobile users, desktop and tablet traffic noticed roughly a 40% decrease.

Recommendations on Providing Relevant and Useful Information

1. Write user-friendly content. Residents come to your website to get specific answers. Most of them don’t want to scroll through a long list of project documents and outdated press releases. Do research to understand what your users need. Then provide actionable and understandable answers to those needs at the beginning of your website. Once you have a working prototype, do more user research and testing to check that your website is easy for people to use and understand.

To understand what residents are looking for:

• Talk to residents before you start writing content. Find out what they currently know and what questions they have about the issue.
• Use Google Trends to understand the words people are searching for related to your issue.
• Do regular user testing sessions to check in with people on how they are using your content and how they navigate and understand it.
• Use user-centered design strategies to develop a clean, consistent visual layout, navigation and content strategy.
• Use Code for America’s free City Analytics Dashboard tool to see how users navigate your website, in real time.

2. Write for the web. Good web content is accurate and easy to read. It is also displayed differently than a technical document or report. Writing for the web means:

• Keeping information brief and to the point.
• Using plain, simple, jargon-free language.
• Structuring pages to work clearly across a variety of platforms and devices.
• Publishing information in the most commonly spoken languages in your community in addition to English whenever possible.
• Publishing information in open digital formats, like HTML, instead of proprietary formats like PDFs. Open digital formats make content more accessible across more platforms, including mobile devices, which are often the primary way residents access the internet. They also make it easier for others to share your content.

Learn more about content strategy: The 18F Content Guide helps writers create content that’s easy to use. The Code for America user needs playbook offers ways to focus city websites on real users’ needs. Learn from the UK’s Government Digital Service about how users read content.

3. Keep content up to date

It is important to ensure that your content is timely, relevant, and accurate. People get frustrated and lose trust in websites if content is out of date or wrong, advertises events that already happened, or has broken links.

To keep your content up to date:

• Regularly check your analytics data to understand which pages become more or less popular over time.
• Remove information that is no longer accurate. Remove old, irrelevant pages regularly.
• Publish content on current issues; monitor social channels, local blogs and news sources to see what relevant conversations are happening and join in.

Signs of Success

The information you publish is useful and relevant when:

• Your written content is produced in HTML by default.
• Your content is accurate, timely, and regularly reviewed.
• You are using analytics to help decide how your website is organized
• You have clear criteria around when to update, archive, and delete content.
• Your content is free of jargon, written in plain language, and at an 8th grade reading level.
• You are publishing content in the most common languages spoken in your community, in addition to English.

III. USING NEW CHANNELS EFFECTIVELY

Through our user research, we learned that many people weren't taking part in traditional public meetings because they were held in spaces or at times that were not convenient or welcoming. To invite participation from everyone, we worked with the City to open up a variety of spaces and channels so that we were meeting people where they were, both online and in-person in an accessible, safe, and welcoming way.

In Boulder, we identified spaces where our target constituencies were already meeting in order to gather feedback where it was convenient for them. Additionally, we deployed Textizen, a tool built by Code for America Fellows in 2012, as a way to bridge both digital and physical channels for communication to reach as many people as possible. Textizen, which creates text message (SMS) surveys and analyzes their results, was used to pose a consistent set of simple housing questions to attract interest and drive people back to the Housing Boulder website. Turnkey access to Textizen can cost from $4,000 to $35,000, depending on the complexity of the implementation. For cities with more in-house technology capacity, the open source code is available on Github.

Additionally, Housing Boulder held public meetings to participate in the housing conversation and supplemented them with digital tools so residents could participate remotely. During these events, the City used Twitter to promote and communicate with residents, Periscope to livestream the event to Twitter users, and SurveyGizmo to conduct two polls, one before the event and one after.

The biggest event during the series of meetings had roughly 200 people attending the panel in person, while 636 live viewers logged onto the Periscope livestream. As a measure of engagement, 225 users “hearted” the Periscope stream. As indicated by the survey results, the demographics of participation online were closer to Boulder’s actual demographics than the attendance at the in-person event.

By diversifying the spaces and channels available for residents to participate in the housing conversation, the City of Boulder realized a significant increase in participation among constituent groups that weren’t participating before. We received:

• 523 responses to the housing survey.
• 50 responses through paper forms.
• 55 responses through Textizen.
• 418 responses online through SurveyGizmo.

All of these responses otherwise would not have been included in the City’s public feedback on the Housing Boulder project.

Recommendations on Using New Channels Effectively

1. Use digital channels as well as offline meetings. Digital channels are important complements to offline meetings because:

• Residents can participate whenever it is convenient for them.
• People who can’t physically attend meetings (because of mobility, geography, or work and family constraints) can participate.
• Young people (who are disproportionately heavy users of digital channels and often underrepresented) are more likely to participate.

Broadcast public meetings and events on digital channels that are easy for your residents to access. Use Twitter to provide real-time text-based updates. New tools like Meerkat and Periscope make it cheap and easy to broadcast live video to the public from a mobile device. Early adopter governments like the City of Philadelphia have already used Meerkat to broadcast events. People who can’t attend the event can still watch the meeting and stay informed.

Supplement online video streaming with digital participation tools so people can be informed and provide feedback. Promote your livestream with a hashtag for people to use on Twitter or Facebook. Having people participate with the same hashtag will create a conversation that can be found in one place.

2. Use tools to meet people where they are. Not everyone will want to attend a public meeting or watch a livestream. Meet people where they are in their physical environment and make it easy for them to take part when it’s convenient for them. In addition to Textizen, we suggest Code for America’s CityVoice, a place-based call-in system to collect community feedback on places (like vacant properties or public parks) using the telephone.

3. Work with community partners. Hearing a message from a trusted voice rather than a stranger or a distant authority figure can make the message much more resonant. Consider developing neighborhood outreach programs like Philadelphia’s 311 Neighborhood Liaison program or Somerville, MA’s ResiStat program. These kinds of programs create partnerships with trusted community figures to deliver authentic and relevant messages on your behalf.

Signs of Success

You are using channels and spaces successfully when:

• You are engaging with your residents using online and in-person channels.
• You understand which channels your residents use, and meet them on those channels.
• The kinds of feedback from online and in-person channels is consistent and can be compared.

IV. ENCOURAGING PRODUCTIVE ACTIONS

The central focus of this project was to create ways for residents to meaningfully participate in the Housing Boulder planning process and express their opinions at meetings.

A significant part of the partnership between Code for America and the City was to strengthen the Code for Boulder Brigade, a group of civic volunteers who create digital tools and employ design thinking that can improve city services. The Code for Boulder Brigade is one of over 100 such volunteer chapters supported by Code for America.^[3] It is relatively new, and the goal for the Brigade over the six months of the project was to help them become a strong partner to the City, who could support efforts to engage residents after the contract period was over.

As a starting point for this work, Code for Boulder hosted a large event to convene residents around the housing topic and began to prototype tools that residents thought would be useful to inform input on the topic. For example, they began working on a tool to make the City’s permit data more accessible and help residents sign up for location- or topic-based notifications and comment on active development review projects. The Code for Boulder Brigade and its subsequent projects is a key vehicle through which the public can contribute to housing solutions for the City. More importantly, it allows residents to use their hands to build useful tools in addition to using their voices to shape decision-making about housing in Boulder.

Recommendations on Encouraging Productive Actions

1. Use a resident survey. Surveys are easy ways to collect data and feedback from residents that can be easily combined to inform decisions. Surveys may not seem like a productive action, but well-designed surveys produce targeted, actionable feedback. They are also a good way for residents to start participating because they take very little time, can be done at any time or place, and don’t require people to identify themselves. Done well with regular feedback, answering a survey can lead to more involved participation.

Tips for deploying good surveys: Use digital surveys to reduce the time spent collecting, combining, and typing up results. Use online survey platforms that are mobile responsive (i.e. can be taken via mobile phone or tablet) such as Typeform or SurveyGizmo. Share survey results immediately with your participants, if you can. If not, collect their email address to report back. Ask for clear and specific information, rather than vague or open-ended feedback. For example, ask residents to rank a set of actions to help the City prioritize. At the end of a survey, make sure participants: are thanked for their time understand how their input is being used have ways to stay involved in the project Work with community groups to promote your survey and make sure people know about it. Collect demographic information in your survey to check whether or not you are reaching a representative sample of your community. You will be successful in conducting a representative survey when you: Collect responses from at least 1 in 1,000 residents, with geographic, ethnic, and age distributions similar to your city’s census data. Ask questions that help you prioritize or identify actions to take. Write in the major languages spoken in your community, not just English. Have an outreach strategy to make sure people know about the surveys. Share the results with the public. Learn about designing online surveys in Code for America’s Resident Survey Guide.

2. Let people participate often and regularly. Let people participate throughout a project, not just during certain stages. This will make it easier for people to participate and continuously provide feedback, producing more meaningful work. Think about how people can participate from the beginning and not just react to decisions that have already been made.

3. Work with community groups. Digital technology has created many opportunities for people to productively get involved with designing and building government services. Code for America facilitates a network of volunteer groups called Brigades who work with their local governments to use technology, design, and data to improve their communities. Working with your local Brigade allows you to use your community to accomplish your goals while also creating ways for residents to participate in local government.

4. Do research with residents. Get into the habit of doing civic user research, where residents help you understand how easy a government service is to use. Feedback from these research sessions can be used to help make services easier to use, responsive to people’s needs, and ultimately, more successful. The act of testing a service or tool is a very direct way for citizens to feel like their feedback will be used.

Civic User Testing (CUT) groups exist all around the country, including Oakland, Chicago, and Chattanooga. The Smart Chicago Collaborative has a guide on how to implement a CUT group.

Signs of Success

You will be successful in encouraging productive actions when:

• Government officials ask residents to get involved, and what they are asking for is clear and specific.
• Interactions between the city and the public are positive and cooperative.
• Government officials can easily understand public feedback and use it to help make decisions or implement processes.
• There are many, continuing ways to meaningfully engage with the city’s work.
• Residents understand how their involvement will change their community in the long term.

V. CREATING USEFUL FEEDBACK LOOPS

In Boulder we found that many people didn’t trust the existing process, felt like decisions were already made, and didn’t think their input would make any meaningful changes. Our goal was to keep residents informed of the status of the project, as well as aware of how their input was being used to shape city decisions.

To do this, we collected email addresses of Housing Boulder participants and created an email list to provide information recapping the project and providing ways to stay involved. This was a practice the City had utilized in the past, allowing them to communicate with stakeholders who self-identified as having an interest in particular topics.

We also released an online survey that asked residents about their biggest housing challenges and needs in Boulder and included an instant report out. When the participant finished the survey, they could see the overall results and learn how their answers compared to other residents.

In addition, we developed a practice of surveying community members about their experience while attending a public meeting. While this feedback did not necessarily inform the policy goals of the Housing Boulder project, it did help the City better understand the effectiveness of their outreach tactics.

Finally, the installation of a citywide website analytics dashboard (which collects real-time data around how people use and interact with the City of Boulder website) will help city employees understand what information residents are looking for. This data can be used to prioritize what information gets shared back to residents and how to organize their website’s information architecture.

Recommendations on Creating Useful Feedback Loops

1. Express appreciation and communicate next steps. Nobody wants to feel like their input has gone into a black hole. At the end of each interaction, clearly outline what you plan to do with the feedback and any next steps. Acknowledge and appreciate the time given to participate.

2. Show things instead of talking about them. When a resident takes part in building or making something, they can immediately see the result of their action. For example, residents who participate in a civic user research exercise see their feedback incorporated into a web page or service redesign and understand that their participation was meaningful and a good use of their time.

3. Keep communication open. Find times to send residents news and information about the topics you’ve discussed with them. Ask residents if they’re willing to give their email address or phone number so you can keep them up to date. Tools like Textizen let you send follow-up messages to residents who take surveys. You can also send emails with tools like Mailchimp that come equipped with analytics you can use to understand how people interact with the emails you send.

4. Collect feedback and make it public. Share feedback publicly so residents can see how their whole community responded. Being transparent like this builds trust and builds consensus about how the community’s input helped you make your decision.

CONCLUSION

By focusing our efforts in Boulder and using 21^st-century engagement strategies, we were able to meet our project goals:

• Stimulate interactive dialogue and collaboration with a cross-section of the community.
• Increase participation and move from one-way communication to multi-way dialogue.
• Create a new approach to community engagement that can be replicated on other projects and in other communities.

As a result, higher proportions of underrepresented constituent groups participated in the Housing Boulder conversation.

At the same time, an equally successful outcome is the new knowledge and competency we’ve help the City of Boulder develop around implementing inclusive engagement strategies. From this partnership, the City has learned how to engage under-represented constituencies, to communicate effectively with the public (both through easy-to-read content and by using multiple spaces and channels), as well as to create meaningful feedback loops between government and the public so residents feel that their efforts are worthwhile. While these skills were originally developed around a Housing Plan, our hope is that these practices spread to other initiatives and city departments.