In my last post (Next-Generation Engagement Platforms, May 12), I covered three of the most powerful open source next-generation public engagement platforms. These platforms are compelling because they invite residents and citizens to do so much more than vote and submit complaints to governments. Each platform I covered invites its users to co-create their society through meaningful actions like voting on budget priorities or determining the physical makeup of forthcoming public spaces.
In the month since that post, Black Americans and their allies have led a long-overdue national reckoning on racism. The movement was sparked by yet another horrific police killing of an unarmed black man, George Floyd, but it quickly grew to encompass racism in all its systemic permutations, at the individual and institutional levels.
One of the most concrete demands of this massive, decentralized movement is to de-fund the police. In many cities, police budgets have continued to grow by the millions while crime sharply decreased, crowding out spending on health and human services and other community. Black Lives Matter activists are determined to use the power invested in them by the protests to confront racism at the systemic level this time around, in the hopes of addressing some of the root causes behind the racist policies, decisions, and hierarchies affecting their lives before attention shifts.
This organizing work includes concentrating public pressure on municipal budgets. In CityLab, Laura Bliss writes about the Peoples’ Budgets blossoming in cities across the country. These proposals offer alternatives to the double-edged status quo organizers see, of over-funded police departments harassing underfunded communities. As Bliss notes, these budget alternatives follow in the spirit of participatory budgeting. Participatory budgeting is where the residents of a city get to develop local improvement projects and then vote on funding them with public budgets. It’s a popular form of civic engagement because instead of just voting for a representative, residents get some degree of direct say in what their tax dollars will support. (We track over twenty participatory budgeting projects in the Civic Tech Field Guide, if you’d like to learn more).
Now, with a nationwide protest movement testing the levers of participatory democracy to see if they still work, we may see a shift from tactical urbanism’s parklets back to its social justice roots. There’s an election this year, and a huge amount of energy will necessarily go towards enacting change at the ballot box. More immediately, however, participatory democracy platforms can help give the public a meaningful place to engage. With this context established, I’ll review three more leading examples of digital engagement platforms.
To B-Corp, Or Not to B-Corp
All three of the following engagement platforms are operated by for-profit companies. In the civic tech sector, that often means the companies are social ventures, pledging to work toward public interest missions and reinvest profits back into their products and services. At least, until they get acquired by a larger govtech vendor, which, as we’ve seen, happens fairly regularly to civic tech startups. Still, an organization choosing a for-profit tax structure doesn’t make it inherently less virtuous than a non-profit organization.
There’s a halo effect that comes with open source projects, because they re-invest in the commons in a way that could benefit everyone, not just individual organizations. But choosing a private company’s engagement platform isn’t always a bad idea. To the extent that many of these engagement platforms are offered under a Software-as-a-Service (SaaS) model, that comes with real benefits in freeing up your team’s time from managing technical minutiae. Installing, hosting, maintaining, and updating an engagement platform demands technical skills, time, and often, money, whether in the form of salaries or hosting fees. Government bodies and organizations without technical capacity to spare probably should choose a SaaS platform, which are hosted for them. The monthly fee is often less than the cost of a single software engineer in many places.
These platforms are free for resident-users, because they’re paid for by the government or institution hosting the public engagement process. Their pricing models are designed to reward, not punish, high levels of public engagement. For example, if a resident’s online proposal attracts 300,000 votes and comments from their neighbors, it won’t cost the host any more than if the proposal only attracted 100 votes and comments. Instead, the hosts pay for the features and level of consultation they desire.
In this post, we’ll go over:
Neighborland is a for-profit company based in the United States
Price: $1,000 per month for the self-service platform
Example installations: Berkeley’s Civic Center, San Francisco’s Central Market
Neighborland is most often used in public planning processes by local governments. It’s best for reaching a wide breadth of participation over the course of a time-bound plan-making process.
The company was recently acquired by Nextdoor, a VC-funded startup that has raised over $450M to become one of the largest place-based social media platforms. Nextdoor is active in the majority of US neighborhoods. This growth hasn’t come easy; the US population unfortunately includes some people obsessed with crime in their area independently of low crime rates, and some “civic” apps grow their user bases by serving this demand rather than designing their feature set to moderate such behavior (we see you, Ring). Some Nextdoor users developed a reputation for racially profiling their neighbors, forcing the company to respond with a series of in-product features aimed at reducing profiling, including text analysis.
Neighborland, the public engagement platform, is still active and will be kept online for the foreseeable future. The platform is best used for public planning, especially when it’s deployed early in the process, before there’s a determined course of action. Its features mirror those of popular social networks (like comments and feeds), and enable a large number of people to take basic participatory actions (“64 neighbors want to improve the Caltrain Station Entry Plaza in Dogpatch“).
In the US and Canada, the company has worked at every level of government and with a broad mix of civic organizations. Over 200 agencies have used it, along with the Rockefeller Foundation’s 100 Resilient Cities program, and universities, foundations, and nonprofits. More than three million US residents have participated on Neighborland-hosted projects. Neighborland’s partnerships with the cities of Mesa, Raleigh, and Los Angeles reached an average of 65,000 residents each — a high number compared to traditional and digital engagement processes alike. Its founder, Dan Parham, says about 10% of those residents actively participated on the platform on a more ongoing basis.
Spectrum of Engagement
As we did in the last post, we’ll evaluate each platform on the IAP2 Spectrum of Engagement. Again, it’s important to note that this scale isn’t meant to suggest that every platform should seek to fulfill all five goals.
Compared to some of the open source platforms, Neighborland offers a relatively limited set of direct democracy features. Residents can do things like submit ideas, vote up others’ ideas, comment, and join forum discussions (check out the full range of Neighborland’s configurations). Platforms like Decidim and CONSUL offer deeper forms of civic engagement, like collaboratively crafting policy proposals, voting on binding proposals, and visually allocating budgets. For this reason, Neighborland is better optimized for planning stage participation than an ongoing municipal platform. Other platforms are more tailored for ongoing public engagement, whereas Neighborland is primarily designed around public consultation within time-bound planning processes.
Neighborland’s strength is engaging meaningful portions of the public through digital means. Parham says they help their public sector clients “reach [between] ten and one hundred times the level of participation compared to traditional outreach methods, at one tenth of the cost.” This estimate is based on an analysis by the Metro Nashville Planning Commission, which found that residents could be engaged for $47 per person at events, $10 per person at public meetings, and $1 per person on the web. Given that events and public meetings are currently off the table, Neighborland’s experience successfully conducting large-scale digital outreach is worth keeping in mind.
Neighborland pro-actively works to ensure that public participation is representative for the locality in which they’re active. Motivated advocacy organizations can sometimes have an outsized influence in open planning processes. Neighborland helps partners achieve representative participation levels and mitigate town hall squeaky wheels by conducting outreach until they achieve statistically valid and demographically proportionate representation. They do this with earned media, targeted Facebook ads, representation quotas, and other outreach methods.
This offering was one of the bigger differences I found in comparing private sector companies like Neighborland with other options. Public engagement administrators can certainly run this kind of digital outreach themselves, and some do, but not all will successfully engage the public at this scale. By providing this service, Neighborland ensures far wider scale engagement than a planning website will ever secure on its own, and lends the processes’ results more legitimacy.
Parham claims a 100% approval rate for their clients’ plans, even with many projects completed. He stresses that this stat doesn’t mean the platform is a rubber stamp for officials: it’s not uncommon for plans to undergo significant alterations as part of the public engagement process. This was the case with a neighborhood plan in Atlanta, as well as the master design of Dorothea Dix Park in Raleigh, North Carolina. In the park example, the public engagement process helped identify public pushback on the allowance of revenue-generating development (like food businesses) in the park. This element of the plan was tabled to allow “further study.”
Because Neighborland regularly achieves high volumes of participation in a given locality, they use Google’s Natural Language Processing API to help governments sort and cluster the large amount of feedback. Parham says this “allows administrators to quickly gauge positive, neutral, or negative sentiment in the public’s response to a concept, scenario, plan, or policy that an administration shares.”
Neighborland also built their own topic modeling and counting tech to summarize participant-generated text. You may recall from my previous post that CONSUL has explored a similar approach. This application of NLP allows administrators to quickly identify the predominant themes that emerge from large volumes of text input from participants. Parham says that despite this time-saving feature, most of Neighborland’s clients still prefer to process public feedback manually, and cluster and synthesize it into themes themselves.
As a hosted SaaS platform, the company rolls out continuous product improvements, technical maintenance, and security updates behind the scenes, without action required from project administrators. These updates are included in the regular fee. The company promises “government and enterprise level security and reliability (99.99%+ uptime)” as well as “consumer grade site performance (2 second or less page load times).” Neighborland also offers a JSON API, which makes exporting project data easy for developers. The API allows the development of custom applications or integrations to extend the platform’s functionality, or connect it to other services.
Training consists of help text built into the administrator view of the portal, how-to videos, and developer documentation of the API. Hands-on training is available for an additional fee. The company says it is a market leader in meeting accessibility and equity requirements (including handwritten data, voice, SMS, and Twitter integrations). It also meets Americans with Disabilities Act and W3C accessibility requirements.
As a proprietary software solution, the code behind Neighborland is closed-source. Judging from previous startup acquisitions, the future of the platform post-sale isn’t guaranteed. Should the platform be folded into Nextdoor, you could later need to turn to another solution. Neighborland’s focus on time-bound engagement processes and the data portability they offer help partially allay these concerns.
Neighborland’s experiences to date are almost entirely located in the United States. They have hosted a project in Toronto, Canada, and are open to projects in other countries, “for a small setup fee.” Hosting multiple projects, using a custom domain name, CRM integration, and setting custom privacy agreement and terms of service, each require a bespoke contract agreement (over $2,000 per month).
Neighborland’s data license requires that data collected be shared between the public agency and the company itself, which they say is to ensure governments don’t censor citizen feedback or input. Neighborland also maintains an independent ‘ideas’ page for each city on the platform, and does not give governments the ability to delete the ideas placed there. This ensures that public input is kept online and accessible, and that even out-of-scope participant input is maintained rather than discarded. Project data is openly available for anyone to extract, analyze, and interpret. This allows contention from third parties like journalists, researchers, and opposing political factions.
Parham emphasizes that their platform is best used to achieve legitimate public buy-in, rather than demonstrate the appearance of buy-in. Because their process is more expensive and time-consuming than traditional survey methodologies, the company generally only receives interest from government bodies that are genuinely interested in gathering representative public input.
Overall, Neighborland stands out as a best-in-class platform for community engagement. They offer feature parity with similar platforms, a highly configurable offering, and the advantages of a hosted Software as a Service platform eliminates many of the technical headaches of managing an online service for smaller or less technically-equipped institutions. Neighborland’s success with a broad range and deep roster of public sector customers to date indicates that the platform is well designed for government users, as well. It is a good solution for engaging a large number of people to engage in a time-bound planning process where project administrators mediate participants’ input.
To get started using Neighborland:
A for-profit company based in Brussels, serving primarily northern Europe, as well as Canada and Chile. (Not be confused with the University of Toronto’s cybersecurity-focused research center, The Citizen Lab).
Price: CitizenLab’s pricing is based on the population of the city and the degree of features and services required. Prices range from $5,000 to $50,000 per year, but usually fall between $5,000 and $15,000.
Example installations: Youth4Climate, Grand Paris Sud (which consists of 23 municipalities)
CitizenLab is designed to help groups inform, consult, engage, and co-create with constituents. It’s best applied at the local level, where people have strong feelings and ideas on their immediate environments.
Spectrum of Engagement
(Depending on installed features — the most basic installations won’t include the features most likely to facilitate true empowerment).
Even though CitizenLab hasn’t focused on the US market, I’m including them in this report because the team has put considerable energy into improving feedback loops on citizen input. In the product, that translates to nudges, internal assignments, notifications, email reminders, and other alerts that keep government officials engaged and responsive to constituent communications.
This strategy appears to work — Marketing Manager Coline Cuau said in a 2019 interview that 65% of their registered users actively contribute to consultations. That amounts to an average of 3 comments, 2 ideas, and 8 votes submitted per person, a really high ratio for this kind of digital engagement platform.
CitizenLab also promotes people finding and connecting with one another through their recommendation engine for related ideas. This way, not all innovation and empowerment needs to take place in relation to the authorities administering the engagement process. Neighbors advocating for similar things can discover one another, a missing link with most social media platforms.
CitizenLab prides itself on helping admins parse large amounts of citizen input. Its multilingual Natural Language Processing (NLP) feature automatically classifies large volumes of constituent comments into topics. Administrators get data visualizations and dashboards that map constituent feedback geographically and categorically. The company has partnered with Nesta, a leading civic technology foundation based in the UK, to research the effects of governmental use of NLP in public engagement. As we consider the role (and potential damage) of public sector AI, early research like Nesta’s will be critical.
Participation on the platform varies depending on a city’s outreach methods, but according to Cuau, generally about 15% of inhabitants visit the CitizenLab platform at least once. Only 2% of inhabitants in a given CitizenLab instance, however, created accounts and interacted on the engagement platform. This is a lower rate than some of the other platforms were able to demonstrate.
CitizenLab’s platform does not yet integrate with other government tech solutions, like the Constituent Relationship Management systems that many government and civil organization offices use to track communications with citizens (though they say this feature is a priority for future product development). In the meantime, CitizenLab’s API does allow data exports, meaning that a software developer could write a program that exports information from the platform to integrate with an office’s existing tools.
Their capability to provide demographic data depends on project administrators collecting it. While the company has noticed less usage from elderly residents, there’s currently no back-end capability to augment online participation with data from offline outreach and workshops. Government employees can manually add these consultations to the platform.
CitizenLab is explicitly designed to encourage strong feedback loops between engaged citizens and civil servants. The platform is set up to ensure people know what’s being done with their ideas and contributions, even if they aren’t being implemented. The platform is also designed to keep people engaged on next steps and implementation milestones. More than other platforms, CitizenLab encourages administrators to establish and communicate the parameters of their public engagement with constituents, including the criteria for idea selection, how votes are counted, and how directly democratic the process will be from its onset. This includes clear deadline dates on next actions, and reliable evaluation frameworks that people can understand.
On the admin side of the equation, CitizenLab’s own internal Key Performance Indicators focus on encouraging moderators to reply more often, more frequently, and in more depth. Moderators can assign constituent input to relevant colleagues and departments, who receive weekly summary and nudge emails to respond. They can participate within existing workflows (like email) so that the addition of an engagement platform doesn’t act as friction on more communication.
To increase public participation, CitizenLab recommends that engagement administrators reserve budgets for marketing and communications of their open input processes. Too often, the initial investment in the digital platform consumes the entire outreach budget. Because restricted budgets are the norm, CitizenLab prioritizes search engine optimization and email outreach campaigns. For engagement process administrators with their own email lists of constituents, email performs quite well for CitizenLab consultations: 40% of recipients open the email and complete the registration process on the platform, a high conversion rate for email marketing. Postcard mailer campaigns are an effective tactic (especially with live events on hold), but they are expensive and require more advance time.
CitizenLab prides itself on the core belief that public participation processes don’t exist to validate existing plans. The company says that it encourages project admins not to bother launching a public engagement process if they’re not willing to adequately reply to the public and take their ideas seriously.
Give a million residents a million keyboards, and officials will have enough support to cherry-pick their preferred comments and ideas. CitizenLab works to encourage public bodies to focus instead on achieving more active communication loops with their constituents. This, in turn, can beget further participation. When constituents see their input implemented, or even considered, it boosts trust and underscores that not all engagement exercises are participation for its own sake.
To get started using CitizenLab:
A for-profit company based in the United Kingdom, Australia, and New Zealand, serving primarily Westminster-system parliamentary governments.
Price: $9,000 – $37,000 per year subscription, depending on government size and integrations required. Their full rate card is available on the UK Digital Marketplace.
Example installations: UK Department of Education, City of Bristol, the Scottish Government
Lastly, I want to quickly mention Delib.net. It’s really designed for parliamentary democracies, but it’s still worth learning about here because of a few important engagement features.
Delib offers three products. Citizen Space helps public officials lead public consultations. Simulator allows citizens to try their hand setting priorities in resource-limited scenarios. And Dialogue engages residents in structured conversations.
Delib is also interesting from the founders’ journey perspective because the company’s platforms were co-developed with the United Kingdom government. The UK government funded 50% of the development costs for the first version of the Citizen Space product, and Delib, the private company, funded the other half. This development period included 6 months of research within the British government identifying user needs and assessing common problems within the governmental agencies. Delib spun Citizen Space out from this government incubation period as an independent company, and has spent the past ten years iterating and building out a broad base of subscribing agencies.
Delib is unique in explicitly inviting the public in to understand the trade-offs officials must make in their decision-making. Delib’s Simulator product allows weighted resource allocation and priority-setting that begins to reproduce the conditions government officials face in their own decision-making. There’s one thing that public officials and roleplaying gamers share in common, and that’s a deep understanding that you can’t have full strength, agility, and dexterity all at the same time.
Delib is also unique in its longevity, and as a result it has seen several digital engagement trends come and go. Delib’s Citizen Space platform, live since 2010, has been used in over 50,000 public consultations to collect over 5 million pieces of response data in public involvement processes. These processes have been hosted by over 400 government organizations, including national governments and their departments, local councils, public service agencies, and pan-European bodies. Delib’s products are primarily used in the United Kingdom, Australia, and New Zealand. The Scottish Government used Delib to consult 24,000 people on their independence referendum. US President Obama’s first open government website, Recovery.gov, used Delib to crowdsource ideas for transparently tracking the public funds.
Spectrum of Engagement
Delib has pioneered — and iterated upon — several open governance trends, from crowdsourcing in the 2000s to the recent excitement surrounding liquid democracy (which empowers voters to participate directly as well as through their preferred delegates). With each generation, Delib has moved the goal post a bit closer to meaningful online participation. This is reflected in their mature products, like Simulator, which take on more arduous goals of involving citizens in priority-setting and instilling empathy for the tradeoffs civil servants must routinely make.
Delib engagements regularly attract tens of thousands of public responses. When the BBC 6 music channel was in danger of cancellation in 2010, 50,000 responses on Delib products helped save the station. Engagements with the Department of Education and Transport for London have also attracted over 100,000 public responses. According to Ben Fowkes, Commercial Director at Delib, their flagship Citizen Space product is used for 5,000 consultations a year. But Delib says their focus is on fostering systemic cultural change within government. The team prides themselves on achieving the qualitatively important impacts more than engagement metrics. When the Mayor of Liverpool was faced with the reality of severe budget cuts, Delib’s Simulator brought the public into the course of action in a meaningful way by letting them experience what it was like to try to fund all of the deserving government programs without adequate resources.
The Delib platform uses open standards, and government agencies can easily export all of the structured data, of which they are the owners. The prominent “We Asked, You Said, We Did” section comes as a standard configuration on the Delib platform. It’s a compelling, easily-stealable way to demonstrate government responsiveness to citizens, and in turn, encourage further participation.
Delib’s strategy is to change government culture from within by convincing civil-servant-level staff to operate openly (and retain subscriptions to their platform, which facilitates such work). The Delib products are geared toward ongoing usage and the company’s goals are to drive widespread governmental adoption of their platform. For civil servants to continue working openly with these tools, their government agencies must retain subscriptions to the service. This is also true of many of the other platforms included in this report, but some of the other platforms are designed for more narrowly time-bound public engagements. Open source solutions like CONSUL and Decidim are equally committed to government culture change, but approach the challenge by building platforms the government can infinitely own, adapt, and improve on their own.
Delib transfers government engagement processes to open-by-default software platforms. The company prides itself on facilitating improved response times from government. Their government agencies commit to returning outcomes to engaged citizens, and the feedback and aggregation tools help civil servants reach ambitious targets. Transport for London, for example, returns with outcome results to engaged citizens within 90 days, considered a major win in the context of government infrastructure decisions.
Like many others, Delib’s products promote citizen ideation and crowdsourcing. Their platforms stand out for the slower, discursive journeys that they bring citizens through. The practice of democracy isn’t necessarily about the final decision reached, but rather the sometimes arduous collective processes of debate, dialogue, and decision. Delib’s Simulator product is designed to support these workflows.
The other platforms I’ve presented here also facilitate ideation, debate, and online voting. Delib’s Simulator goes a step further, elevating the citizen to the bureaucrat’s level of understanding so that they can gain a more sophisticated understanding of the tensions and tradeoffs inherent in governance decisions, as with the City of Liverpool example. Delib’s products demonstrate respect for a more enlightened relationship between citizen and government, where people with different perspectives must come to a common solution. Like Pol.is, Delib’s products strive for a more enlightened relationship between citizen and government, where people with different perspectives need to come to a common solution.
To get started using Delib:
If you’re looking for a US-focused solution and are yearning for more options to explore, check out Bang the Table and Public Input. There are also lots more digital public engagement platforms, public and private, in the Civic Tech Field Guide’s Engagement Tech section.
Whether we like it or not, civic tech is increasingly a business sector. Even this one subset of the field, public engagement platforms, is teeming with companies alongside open source projects. The recent creation of the Association Civic Tech Europe, a trade group for public engagement platforms, is further evidence that the sector is maturing. With so many options, it’d be easy to get lost.
Some common success factors enabling web platforms to connect online public engagement with actual results from analog political institutions are:
- A driving political force that compels the government to share some modicum of real power with digital participation platforms. This can originate in external pressure from protest movements, as in the Spanish and Taiwanese cases, or budget hardships, as in Liverpool’s case, or, in many cases, internal motivation in the form of progressive government leadership willing to experiment with new methods.
- People-powered institutions won’t rely solely on tech. For efforts like vTaiwan and e-Democracia (Brazil), significant teams of volunteers, consultants, and/or paid staff help synthesize large volumes of digital engagement back to the interpersonal level with public officials. This additional human labor connects digital citizen feedback to real-world political systems that might otherwise proceed without it. No matter which engagement tools you choose, you should allocate time and resources for this trans-mediating step.
- Some of the most exciting citizen engagement platforms are designed to operate at municipal scale. Cities have emerged as a locus of action when national legislatures are otherwise blocked. Mayors and other local politicians are often too focused on keeping services running to spend too much time on entrenched political debates. It turns out that citizens are likewise less polarized on local policy issues, too. This dynamic appears to play out with web platforms, too. Civic platforms that attempt to shepherd polite conversations about nation-level issues eventually come to rest in the Civic Tech Graveyard. The successful platforms covered here instead focus on local level issues like housing, transportation, local economies, education, and quality of life factors like noise and air quality.
The public engagement platform genre of technology also presents risks:
- Digital divides remain in who feels comfortable using technology and who can afford it. Rebecca Rumbul’s seminal mySociety study on the topic, Who benefits from civic technology? Demographic and public attitudes research into the users of civic technologies (2015) found that in the United States and the United Kingdom, digital engagement platforms primarily attracted older and less ethnically or genderly diverse user than the overall population. These divides are further complicated by broader socioeconomic barriers, where some groups of people feel more comfortable advocating for their interests than others. We need to work hard against this status quo to achieve anything resembling representative participation using technology.
- For these reasons, most public engagement experts insist that digital consultation processes be augmented by a commitment to offline outreach like workshops, meetings, surveys in places where less-connected populations spend time. Our ability to run face to face workshops is hampered at the moment, and only some digital engagement platforms make it easy to integrate feedback from offline consultations with the rest of the data. Equitable, effective digital outreach has become more important.
- Looking forward, developers of public engagement platforms are incorporating emerging artificial intelligence solutions, like the aforementioned topic clustering of citizen feedback. It is not clear what effect these technologies, widely recognized as imperfect, might have on citizens’ perception of whether their government is listening to them, or on governments’ perception of citizen voices. We should pay attention to research like Nesta’s on this topic, and choose platforms that show us their code when possible.
- Power. The traditional gatekeepers of power aren’t inclined to respond to the public voice as it is expressed online. The majority of digital engagement processes conducted on the platforms I covered here were embarked upon voluntarily, by governments pro-actively seeking improved public engagement. Or as one platform founder put it, “Hundreds of agencies have used our platform. That leaves over 35,000 that haven’t.”
Political force has been a key factor in how digital engagement platforms like Decide Madrid, Decidim Barcelona, or g0v Taiwan secured enough power to deliver meaningful outcomes from people’s engagement. It will be interesting to see if the Black Lives Matter protest movement can shift power back to digitally-organized communities as we’ve seen in Spain and Taiwan.
The deleterious effects of the pandemic’s sudden termination of many of our physical channels for civic voice, be they town halls, public workshops, or office visits with representatives, has not yet been determined. But in the face of extralegal murders at the hands of those entrusted to protect us from violence, people aren’t waiting. They have brought the anti-racism conversation to the streets in defiance of social distancing rules, and to the board room in defiance of racist social norms that prevent us from having uncomfortable conversations.
These digital platforms and engagement processes only have meaning if they are imbued with power. That power has to either be bestowed by those who already have it, which has historically happened in limited cases, or it has to be won.
If I were working inside of government right now (particularly at a local level), I would consider this a very important time to do more to hear from the people who have been hurting and without voice. That includes giving historically marginalized communities more ways to contribute to building the society they deserve to live in. Setting up a public engagement platform (digital or otherwise), making sure people have easy access to it, and mapping it to levers of actual, meaningful power in your institutions is one way to do that.
Everyone outside of government can organize, through campaigns and protests and democratic competition, and win the people the power to direct communal funds, collectively draft laws and policies, and reshape our cities. Digital engagement platforms are not going to win that power, but once secured, they can help focus it.