NYC Social Service Atlases for Referrals: Best Free Databases

On January 17, 2024, New York City's public hospital system did something that quietly reshaped how social care moves through the five boroughs: it wired the Findhelp referral platform directly into its Epic electronic medical record, putting more than 7,000 community resource programs within reach of every clinician who logged into a patient chart at any NYC Health + Hospitals facility. Within six months, that single integration had produced an 86 percent closed-loop referral rate, and by the end of 2023 the system had already screened over 250,000 primary care patients for social needs. I want to spend some time with you on what that actually means in practice, because the moment we step back from the press release and look at the day-to-day mechanics, a much richer picture comes into focus — one in which Findhelp and Unite Us are only the most visible layer of a far deeper ecosystem of free referral and mapping tools that NYC nonprofit and community-based practitioners can lean on without writing a single check.

Closed-Loop Referral Platforms: Findhelp and Unite Us in NYC

When I sit down with a program director who is tired of handing a client a stapled printout of agency names and hoping for the best, the first thing I try to reframe is the concept of a "closed loop." A closed loop is simply the difference between sending a referral into the void and knowing whether the client actually got there, whether they were seen, whether services were rendered, and whether the referring organization hears back. In a city as dense and fragmented as New York, where a single client in Hunts Point might need a food pantry, a housing navigator, a Medicaid enrollment assistor, and a behavioral health provider all in the same week, that feedback cycle is what turns well-intentioned handoffs into measurable outcomes — and it is precisely what Findhelp and Unite Us were built to deliver.

Findhelp (formerly Aunt Bertha) entered the NYC conversation in earnest when NYC Health + Hospitals completed its system-wide integration in January 2024, but the platform itself had been quietly aggregating city-specific resource data for years, and that database now reflects well over 7,000 active NYC programs spanning food, housing, benefits navigation, workforce, and behavioral health. The integration matters because Epic is the connective tissue of the public hospital system, so when a clinician at Elmhurst or Jacobi identifies a social need during a visit, the Findhelp directory is already embedded in the workflow, the search is filtered to programs that actually serve the patient's zip code, and the referral can be sent with a single click. The 86 percent closed-loop rate reported within the first six months tells us something important: the bottleneck was never the willingness of providers to refer, it was the absence of an institutional feedback channel.

Unite Us operates on the same philosophy but with a different geographic and contractual architecture. The company acquired the Chicago-based NowPow platform in September 2021, folded it into its network product, and now operates Unite NYC as a closed-loop referral network that is offered free of charge to community-based organizations and Federally Qualified Health Centers. That detail — free for the receiving CBOs — is worth pausing on, because it inverts the usual SaaS economics. The payers and the health plans (including, as of March 15, 2024, Amida Care, the state's largest Medicaid Special Needs Plan) fund the infrastructure on the assumption that a tracked referral is cheaper than an unmet social need landing back in the emergency department, and the community organizations on the receiving end get enterprise-grade case coordination without a license fee.

A closed-loop referral is not a fancy email — it is the difference between hoping a client connected and knowing they did, and that single fact is what makes Findhelp and Unite Us worth every minute of onboarding.

In practical terms, what this means for the small executive director running a soup kitchen in East New York or a domestic violence shelter in Fordham is that, after a modest intake into one of these networks, every referral sent and received becomes a tracked, timestamped event with an outcome attached. If you have ever lost sleep wondering whether the family you referred to a housing program in December actually got a call back, this is the layer that ends that uncertainty.

Specialized Service Databases: Plentiful and Open Referral NYC

Not every referral is a behavioral health handoff from a hospital, and the NYC ecosystem is honest about that. Two tools in particular have carved out specific niches that deserve a place on every navigator's shelf, and they illustrate a complementary design philosophy: do one thing very well, and let the data stay open.

Plentiful is the one I usually start with when I am talking to a food pantry coordinator or a benefits enrollment specialist who keeps a paper reservation book behind the front desk. It is a free mobile reservation and referral database developed by the NYC Food Assistance Collaborative, and it does exactly what its name promises: it lets a client find a nearby pantry, reserve a slot, and walk in without waiting in a line that winds around the block, and it lets the pantry manage capacity, track visit patterns, and reduce no-shows on the supply side. The reservation layer is more consequential than it sounds, because in food insecurity work, dignity and predictability are part of the intervention. A family that knows they have a confirmed pickup window at 2:15 on Tuesday is a family that can plan around it, and Plentiful operationalizes that small but profound shift. Pantry operators also get aggregate data on visit frequency, which feeds directly into purchasing decisions and grant reporting — a quiet but meaningful efficiency gain for organizations that are often running on two staff members and a determined volunteer corps.

Open Referral NYC takes a different cut at the same problem. It is a local network of providers and technologists who have committed to the principle that social service data should be open, portable, and not locked behind any single vendor's login wall. Practically speaking, the network publishes free Airtable templates that an organization can clone, customize, and deploy to publish its own service directory in a standardized format, and it runs training for teams that want to take control of their resource data without outsourcing the schema. For a community organization that has ever lost institutional memory when a key staffer left, or that has been frustrated by the cost of maintaining a custom website, this is a remarkably grounded alternative: you own the data, the schema is interoperable with major referral platforms, and the price of admission is willingness to learn a spreadsheet.

Open Referral NYC treats your resource list the way open-source software treats code — shared, versionable, and belonging to the community that built it.

What I find encouraging about these two tools side by side is that they model different points on the same spectrum. Plentiful optimizes for a specific user experience in a specific service category and does so in a way that frontline staff can adopt in a week. Open Referral NYC optimizes for institutional sovereignty and long-term data stewardship. Neither requires a procurement cycle, and both are genuinely free at the point of use, which is the only price point that matters for the kind of grassroots infrastructure that holds the city's safety net together.

Data-Driven Mapping: NYC Community and Neighborhood Health Atlases

Every so often, a community-based organization is asked the same question by a funder, a city council member, or its own board: "Where are your clients coming from, and what does the neighborhood actually look like?" It is a reasonable question, but it is also expensive to answer well if you do not have a research department, a data engineer, or a graduate student on retainer. The two atlases I want to walk you through were built precisely so that you do not need any of those.

The NYC Community Atlas is maintained by the Center for Innovation through Data Intelligence, which sits inside the Mayor's Office, and it publishes two-generational and three-generational demographic and social asset profiles for all 59 NYC Community Districts. The 2Gen and 3Gen framing matters, because it means the atlas does not stop at counting households — it traces the ecosystem of supports available to children, parents, and grandparents together, which is closer to how families actually experience service delivery. If you are writing a proposal that needs to demonstrate that your catchment area has both a documented need and a documented asset base (which is increasingly what place-based funders want to see), the Community Atlas gives you a credible, citable starting point.

The New York City Neighborhood Health Atlas, managed by the NYC Department of Health and Mental Hygiene, does for health and social indicators what the Community Atlas does for demographics. It tracks approximately 100 measures across 188 Neighborhood Tabulation Areas, and the granularity is one of its real strengths — NTAs are smaller than community districts, which means you can zoom in on Brownsville or Mott Haven or Sunnyside with a level of resolution that helps you tailor outreach rather than just proving that a borough-wide problem exists.

These atlases do not process a referral or close a loop; they tell you where to send the referral in the first place, and that is a different and equally essential job.

I want to flag one distinction clearly, because I have seen it cause confusion in board presentations. Neither the Community Atlas nor the Neighborhood Health Atlas is a transactional referral platform. They are profiling and visualization tools — they map the territory, but they do not move a client through it. When a grant narrative asks you to demonstrate community need, use them. When you need to know whether a referral was completed, use Findhelp or Unite Us. Mixing those two functions is one of the most common strategic errors I see, and getting it right early saves a lot of awkward reporting conversations later.

Targeted Case Management Systems: ATLAS and the Atlas Evidence Database

The NYC ecosystem includes two more systems that sound almost identical by name and do entirely different things, and untangling them is worth a few minutes of your time because both have shown up in RFPs and referral workflows in ways that can confuse even seasoned practitioners.

The first is ATLAS — the Accessible Tool for Learning About Students — which is the case management system that NYC Public Schools uses to manage special education referrals and to track student cases. It replaced the legacy systems known as DTS and CAP, and the deployment timeline extends into January 2026 as the rollout continues across school sites. If your organization works at the intersection of education and social services — for example, if you run a school-based mental health program, a youth development initiative inside a public school, or a transition-age support service — you may find yourself needing to interface with ATLAS data for specific students, but it is not a general-purpose community referral database, and I would caution against trying to use it as one. It is purpose-built for the Department of Education's special education workflow, and outside that lane, it will not give you what you need.

The second is the Atlas Program, which is a partnership between the NYC Office of Neighborhood Safety and United Way of NYC that launched in 2020. Its operational core is the Atlas Evidence Database, or AED, which tracks referrals to free, community-based therapeutic services for justice-involved youth. This is a very different tool from a school case management system, and the distinction is worth preserving in your mental model: ATLAS (capitalized, an acronym) is about education; the Atlas Program (a name, not an acronym) is about violence prevention and youth justice.

Two systems with nearly identical names, two entirely different missions — knowing which is which is the difference between a clean referral and a frustrating dead end.

For practitioners working in the youth justice or violence prevention space, the AED is a genuine asset, because it gives participating community-based therapeutic providers a way to be discovered by the city's diversion and safety infrastructure without bearing the cost of building their own intake funnel. The unknown — and I want to be honest about this — is the exact software architecture underneath, so if your organization is evaluating a deep integration, ask the Office of Neighborhood Safety or United Way of NYC directly about API access, data sharing agreements, and any consent structures that govern how client information moves through the network.

Selecting the Right Tool for Client Tracking and Resource Mapping

After we have walked through all five layers, the question I most often hear is some version of "so which one do I actually pick?" The honest answer is that they are not competitors — they are different instruments for different moments in the same workflow, and the strategic move is to compose them rather than choose among them.

A few principles tend to keep organizations out of trouble. First, match the tool to the specific handoff you are trying to track — if your team is sending clients from a clinic to a community resource, you need a closed-loop network, not an atlas; if your team is writing a grant that requires neighborhood-level context, you need an atlas, not a referral network. Second, do not pay for what the city already gives away; every tool listed above is free at the point of use for community-based organizations, and any vendor trying to sell you a parallel product should be asked, very politely, why you would pay for a less authoritative version of data the City already publishes. Third, think about data stewardship before you sign anything — even when the tool is free, the data your clients generate is not, and Open Referral NYC's commitment to open, portable schemas is worth treating as a north star when you evaluate alternatives.

Finally, treat your choice as a living decision rather than a permanent one. The ecosystem is evolving quickly: integrations deepen, new funders enter with their own reporting requirements, and the line between a referral platform and a case management platform continues to blur. What I would encourage you to do, if you are starting today, is to pick one closed-loop network that matches your strongest referral partner, layer in one specialized database (Plentiful if you touch food, Open Referral NYC if you publish your own resource list), and use the two city atlases as your standing reference for any neighborhood-level narrative you need to tell. That composition is free, it is defensible, and it gives you a roadmap for adding capacity as your organization's footprint grows — which, in this city, it almost certainly will.