Spots on the map Where the Opportunity Is
The Map Has Dark Spots.
That's Where the Opportunity Is.
Institutional mistrust is one of the defining civic issues right now — felt across every community, regardless of politics. Most programs read it as a reason to stay away. We read it as the highest-value opportunity on the map.
Companion piece: this is the technical diagnostic underneath RFUTR's Disinvested Communities practice. Where that piece describes the lab studying communities facing land pressure and institutional exit, this paper explains the instrument doing the reading.
Every institution serving a community has a map with dark spots — census tracts where engagement has historically failed and resources don't go. Those spots aren't dead zones. They're where institutional trust has collapsed the most, and where the institutions willing to enter first reach the people nobody else is reaching.
We Studied the Dark Spots
Community Intelligence (CI) is a behavioral trust-mapping platform built by RFUTR and PathSight Predictive Science. We measure trust at census-tract resolution, from roughly 350 tract-level indicators of which 140–145 are typically populated for a given tract, refreshed annually to five-year rolling averages — zero personally identifiable information, at any stage. We don't measure what communities say. We measure the structural conditions that shape whether they act.
We've found something programs consistently miss: the places where institutional trust has broken down the most are not the places to avoid. They're where Floor Identification — our tested, validated capability for finding the most disconnected places in a field — consistently surfaces the largest gap between what's assumed and what's actually there.
Why Two Communities React Differently to the Same Program
Two communities. Same program. Same message. Same outreach effort. One moves. The other doesn't. The demographic explanation doesn't hold. The income explanation doesn't hold. The geographic explanation doesn't hold.
The real variable is Trust Orientation — whether a community banks trust vertically, through institutions, or horizontally, through peer networks. Outward-oriented communities trust institutions, show up, and engage — but engagement isn't commitment, and without stewardship design, outward orientation produces initiatives that launch and don't hold. Inward-oriented communities route trust through neighbors and known community members. They don't respond to the mailer. They appear closed. But that's a routing requirement, not a refusal.
CI reads communities at three levels before any behavioral question is asked.
The relationships and channels that already exist in a place — measured by Institutional Leverage.
Whether that structure is oriented toward the ask being made — measured by Trust Orientation and Trust Transferability.
Whether there's enough margin for a community to move, and hold, once it does — the newest, least-tested layer.
The Diagnostic Toolkit
Four reads, one platform, built for civic and community-serving institutions working in exactly these conditions. Each one carries a stated test status — we don't blur what's validated with what's still in development.
Finds the tracts most institutionally disconnected in a field — where to start.
Tested & heldThe read points to where trust concentrates; naming the actual people is human fieldwork the read informs, not an automated output.
Read + fieldworkWhether behavior moves through horizontal peer networks or vertical institutional channels — confirmed to diverge from income across multiple cities.
MeasuredWhether behavioral change holds well past a program's launch — a structural estimate today, not yet tested against measured outcomes.
In the labThe person who shows up to every meeting but never commits isn't the floor. The person who hasn't responded to a single mailer but will move when their neighbor asks — that's what Floor Identification is built to find. And it finds the messengers who can reach them without collecting a single piece of personal data. No surveillance. No PII. The signal lives at the community level — which is also where it's most defensible.
You Already Know Something Isn't Working
More staff. More visits. Same result in the neighborhoods that matter most. The problem isn't effort. It's routing — the right message, the wrong messengers, in communities that don't trust institutional channels.
High-disconnection tracts get written off as low-engagement, low-value. Floor Identification measures institutional disconnection, not disinterest. The people are there. The channels to reach them have just never been mapped.
Every institution serving these communities competes over the same contested turf with the same tools. The highest-value, lowest-competition gains are sitting in the tracts everyone else marked as too hard to move.
Racine, Wisconsin: One City, Wildly Different Neighborhoods
Racine reads as the most institutionally closed of the Wisconsin cities we've studied — a citywide characterization every county-level dataset would agree on. What that same dataset can't see is what happens inside Racine itself.
Two neighborhoods, two miles apart, converting the same underlying trust into wildly different levels of institutional engagement — not because one community cares less, but because outreach was built for one pattern and never adapted for the other. Racine's overall Community Guardian orientation is the signal: this is a city that routes through relationship channels, not institutional ones. Most outreach here has been built backward.
The Entry Sequence: Connect, Protect, Recover
Connect, Protect, Recover is a three-stage civic re-entry model that matches the intervention to the actual condition of the community. In a high-disconnection tract, the sequence can't be reversed — the civic infrastructure required to carry a behavioral shift doesn't exist yet, and skipping ahead wastes the investment.
Re-establish trusted, non-transactional presence — not a program office, the persistent civic connective tissue a disconnected tract is missing.
Address real daily-life conditions eroding trust — safety, health access, housing. Presence around these conditions earns legitimacy no advertising can manufacture.
Convert restored trust into sustained civic participation — engagement that holds well past the program's launch.
This is what "entering the dark spots" actually means in practice. Not a bigger ad buy aimed at the same tracts with the same message. A sequenced entry — Connect before Protect, Protect before Recover — calibrated to what the read says each community can actually carry right now.
How to Engage Institutional Mistrust
Most advice about "hard to reach" communities is about tone — listen more, show up, be authentic. That advice isn't wrong, but it isn't the lever. Institutional mistrust isn't a mood. It's a routing problem — a community has learned, through specific experience, that messages arriving through institutional channels don't carry follow-through. Fix the routing and the tone problem mostly resolves itself.
The program's name shouldn't be the first thing a high-disconnection tract hears. The read points to where trust concentrates — naming the specific person is fieldwork the read informs, not something the instrument does on its own.
A program that opens with an ask in a Connect-stage community repeats the exact pattern that produced the mistrust. Sequencing isn't a courtesy — it's the difference between landing and being ignored.
Inward-oriented communities don't ignore mail and digital because they're unreachable. They ignore them because those are vertical channels in a horizontal trust environment. Trust Flow mapping shows which channel actually has permission to land.
Sometimes the honest finding is that a geography needs a longer relationship than one funding cycle allows. Long-horizon stewardship reads are still R&D-stage — not yet a tested prediction — but even the structural estimate is worth surfacing early, so the commitment matches what the community has been promised before and lost.
This is not a campaign tool. It is not built for one party, one candidate, or one election cycle. RFUTR does not build for, or share data with, political campaigns or party committees; that boundary is structural, not situational.
PathSight × RFUTR 501c3 created Community Intelligence as a behavioral data partner that acts as implementation insurance for high-stakes community investments and public programs.
Our co-founder Dr. Bob Raleigh built the behavioral science underneath CI over more than a decade of research. Kevin Starace built RFUTR from two decades of field work across nature, public health, and impact capital — with a sustained focus on disinvested communities.
Out of R&D and now operational. Our first step is always a Ground Truth Sprint — a behavioral diagnostic across your target geographies. The findings drive the engagement. Every engagement is bespoke.
Everything in this analysis is derived from aggregate behavioral signals. Zero personally identifiable information is collected, stored, or used at any stage. No individual is named, tracked, or profiled.
The unit of analysis is always the community — never the person.
In theory, we do not own the community data — nor do our partners. The community does. We believe this is the only ethical way to do this kind of work.
It is also, as it turns out, the way that respects what the instrument can actually claim: community behavioral architecture is what this reads, structurally and repeatably — not a judgment on any one person.
Best Use for Community
begins where communities already are.
We're a behavioral lab studying how communities build cohesion, defend their land, and shape what comes next — and an open invitation to learn, organize, and invest alongside the communities already doing this work.
What This Lab Is
Most land use decisions get made by asking one question: what is this site worth, to whom, for what return? Best Use for Community starts from a different question — what does this place need to become, for the people who live around it and the land and water that sustain them?
We call this a lab because that's what it is. We're studying how trust moves through communities facing land use pressure, corporate overreach, and institutions that have stopped listening — and we're building tools alongside the communities doing that work, not handing tools down to them. Community Intelligence (CI) is the behavioral layer underneath that study: 174 real-time indicators, ZIP+4 precision, zero personal data collected on anyone. What it produces is a read of how a community is organized, where its cohesion is strong, and where outside pressure is most likely to find an opening — so that the community sees it first.
Before any behavioral question gets asked, CI reads a community at three levels.
What the community is carrying — the civic, economic, and institutional conditions that shape how people receive information and act on it.
Whether the community's different groups — long-time residents, newer neighbors, local organizations — are moving in the same direction or pulling apart.
Whether there's enough shared civic fabric to sustain collective action — a campaign, a legal fight, a long defense — over time.
Where This Shows Up
The same behavioral understanding applies wherever a community is organizing around land, resources, or institutions that aren't listening. These aren't separate services — they're examples of the same underlying work, in different settings.
Understanding where cohesion already exists — and where it needs to be built quickly — before a zoning decision or land sale moves forward without the community at the table.
Legal strategy depends on a plaintiff community staying connected and aligned over months or years. Reading where that cohesion is strong, and where it's fraying, early.
When a city council or local board is no longer representing the community it serves, understanding how trust and information move outside those formal channels — and who communities already turn to instead.
Where industrial siting, waste, and emissions decisions land hardest on communities with the least formal power — and how those communities already organize around shared experience.
Every brownfield site carries two questions: what would it take to remediate and convert it, and what does the surrounding community actually want it to become. Best Use of Land × Nature asks both.
Some defenses take years. Understanding whether a community's organizing structure can sustain that — and where outside support is most needed to keep it from fraying.
Best Use of Land × Best Use of Nature
These are two indices we're developing — part of a growing body of work that communities, researchers, and partners are helping shape. Neither one is the point on its own. Together, they ask the question that usually gets skipped: what does this specific place, at this specific latitude and longitude, need — for the people who live there and the land and water around them?
A read on land use potential that starts from the community's own sense of what's needed — housing, gathering space, green space, economic use — rather than starting from what a developer's model says the land is worth.
A read on the ecological role a specific site already plays or could play — water, soil, habitat, shade, flood buffering — quantified using existing environmental data at the precision of a single location.
One place this comes together is a place-based partnership working through brownfield sites — understanding what's there now, what it would take to remediate and convert each one, and what the surrounding community actually wants it to become. That work is one expression of Best Use of Land × Best Use of Nature, and one of several active threads in the lab right now.
The indices themselves are still being built. That's part of the invitation — this is work that communities, researchers, and funders can help shape as it develops, not a finished tool being delivered.
This is a lab, and labs grow through partnership. If you're a community organization working through any of the situations above, we'd like to learn alongside you — understanding your community's behavioral architecture is something we can offer at no cost, because it's how the lab learns.
If you're a funder, researcher, or values-aligned partner, the growing body of work here — the Best Use of Land and Best Use of Nature indices, the brownfield partnership, the behavioral architecture studies — is work you can help shape and grow. We're early. That's the point.
Communities decide what "best use" means for their own places. Our work is building the tools that let that decision be seen, supported, and defended.
Everything in this work is derived from aggregate behavioral signals. Zero personally identifiable information is collected, stored, or used at any stage. No individual is named, tracked, or profiled.
The unit of analysis is always the community — never the person.
In theory, we do not own the community data — nor do our partners. The community does. We believe this is the only ethical way to do this kind of work.
It is also, as it turns out, the more accurate way. Individual data is noisy. Community behavioral architecture is stable, measurable, and predictive.
Friction Index™
+Spend/Avoid Map
The diagnostic that quantifies what conventional outreach wastes — and the decision map that shows where every dollar lands instead. From behavioral data to budget decision, in thirty days.
In Westover — Trust Orientation 26 — a standard institutional outreach effort reached 8 of every 100 residents. The other 92 were unreached. Not unreachable. Invisible to the delivery channel being used. The trust network reached 70. That 62-point gap has a dollar value. It always has. Now it has a name.
Where This Came From
Two years of behavioral lab deployment across health, civic, resilience, and conservation contexts revealed the same pattern in every geography: the communities where conventional outreach failed hardest were exactly the communities where trust-network delivery gained the most. The dollar was bouncing off the same neighborhoods, cycle after cycle — and nobody had named it, quantified it, or shown what it cost in real budget terms.
Friction Index™ names it. The Spend/Avoid Map translates it into a per-neighborhood budget decision before a dollar moves.
A campaign spends $50,000 on digital outreach, mailers, and phone banking across six neighborhoods. In three of them, conventional reach is below 20%. $30,000+ bounces. The trust network reaches 60–70% of those same neighborhoods. That $30,000 is recoverable — through different channels, different messengers, different sequencing. The Friction Index finds it. The Spend/Avoid Map redirects it.
The Six Variables
Each neighborhood gets a row. Each row tells the full story — from the behavioral condition to the signal to the dollars recovered.
The behavioral score anchoring everything. Above ~40: mixed channels work. Below 40: personal networks rule — only relational, person-to-person paths reach anyone. But TO alone is never enough. The fit type determines what the number means.
The share of a standard institutional effort that actually lands. In high-TO Hybrid neighborhoods this is substantial. In Guardian neighborhoods and low-TO communities it collapses — sometimes below 10%. The number that tells you whether your default delivery strategy has traction here at all.
The share reached when delivery goes through the trust network instead — Authentic Messengers, Bridge Nodes, faith institutions, peer navigators. Measured from the same behavioral substrate that produced the TO score. The trust architecture penetrates where institutions cannot.
Friction = 100% − Conventional Reach. The share of the conventional effort that is wasted because it never lands. Highest exactly where the need is greatest. The number that makes a budget director wince — because it has always been there, and nobody named it until now.
DELTA = Network Reach − Conventional Reach. The reachability the trust network recovers that conventional delivery left on the table. Largest where conventional delivery fails worst. The headline metric — and the whole argument, computed.
The decision the indices produce: SPEND (dollar lands, route confirmed), AVOID (dollar bounces, or worse — funds the opposition), CONDITIONAL (entry possible through a specific channel only). One signal per neighborhood. One trust reason per signal. The product is the decision, not the data.
| Neighborhood | TO / Fit | Signal | Conv. Reach | Friction ▲ | Network Reach | DELTA ▲ | $ Wasted | $ Recovered | Net ROI Shift |
|---|---|---|---|---|---|---|---|---|---|
| WestoverLow-trust · personal networks only | 26 | SPEND* | 8% | 70% | $45,800 | $30,640 | +$30,640 | ||
| EastdaleStrained · personal networks rule | 34 | SPEND* | 18% | 62% | $41,000 | $22,000 | +$22,000 | ||
| Lincoln-King · RacineDisinvestment Hybrid · TO 36 · TE 78 | 36/TE78 | SPEND† | 22% | 74% | $39,000 | $26,000 | +$26,000 | ||
| MillbrookHybrid · mixed channels open | 41 | SPEND | 38% | 58% | $31,000 | $10,000 | +$10,000 | ||
| HarborviewHybrid · high institutional trust | 52 | SPEND | 55% | 68% | $22,500 | $6,500 | +$6,500 | ||
| Lance · KenoshaGroup-2 Guardian · TO 58 — wall, not door | 58 ⚠ | AVOID | 8% | 70% | $45,800 | Funds opposition | Avoid entirely | ||
| NorthgateHigh-trust · institutional channels open | 68 | SPEND | 72% | 80% | $14,000 | $4,000 | +$4,000 |
* SPEND with network routing — not conventional outreach. † SPEND with sequencing — generative deposit before any ask. ⚠ Lance: DELTA is real but moves against you — Guardian network is organized resistance, not accessible opportunity.
The Media ROI Argument — A $350K Program, Mapped
Apply the Friction read to any real program budget and the argument becomes concrete. The following uses a $350K illustrative spend across seven neighborhoods — the same geography as the table above. These are directional figures; actual returns require CI deployment against the specific budget.
This is not a savings argument. It is an efficiency argument. The budget stays the same. The behavioral routing changes. The reach compounds. Every dollar that previously bounced now lands through a channel that has permission.
The Friction Index tells you what conventional outreach is wasting in each neighborhood. The Spend/Avoid Map tells you what to do about it — before a dollar moves. For every neighborhood in a geography, the map returns one of three signals and one trust reason. That is the decision. The behavioral engine underneath produces it; the program, campaign, or investment receives the call.
The critical distinction the Spend/Avoid Map makes that no other platform does: the same Trust Orientation score means opposite things depending on neighborhood fit type. Lance, Kenosha: TO 58 — the highest in the city. Every other platform flags this as high engagement. The Spend/Avoid Map calls it Avoid. Spending there doesn't move the community. It funds organized resistance. That distinction, applied before the media buy, is the whole ROI argument.
Racine and Kenosha — The Full Spend/Avoid Map
| Neighborhood | Signal | Trust Reason |
|---|---|---|
| UptownHybrid · Quiet Safety | SPEND | Hybrid inner layer aligned. Safety-investment is the generative-trust entry. High return. |
| Wilson · Columbus · LPHybrid · Progressive Steward | SPEND | Group-1 values under Group-2 surface. Collective Affirmation after foundation set. |
| Saint Peter'sGroup-2 · Faith-Anchored · TO 52 | CONDITIONAL | Parish is Bridge Node. Faith channel only — wrong door is a wasted dollar. |
| Bain Park · JamestownHybrid · dual signal · TO 35 | CONDITIONAL | Dual amelioration signal. Sequence only if cycle timeline permits. |
| LanceGroup-2 Guardian · TO 58 | AVOID | Highest TO = well-maintained wall. Spending here funds organized resistance. |
| Nash · McKinley · EndeeGroup-2 Guardian · TO 27–35 | AVOID | Retracted rigidity — closed even to itself. No ameliorating signal. |
| Neighborhood | Signal | Trust Reason |
|---|---|---|
| Prospect HeightsHybrid · Bridge Anchor | SPEND | Load-bearing anchor. Every other move sequences from here. |
| Lincoln-KingDH · TO 36 · TE 78 | SPEND† | Highest elasticity in the corridor. Sequence third — generative deposit before any ask. |
| West LawnGroup-2 · Inward · TO 42 | CONDITIONAL | Entry to deeper Racine. Identify whose word travels before broad outreach. |
| TO-39 Cluster (13+ nbhds)Group-2 Guardian · Uniform TO 39 | AVOID | Uniformity is the finding — one disinvestment pattern across 13 geographies. Not reachable this cycle. |
Racine's fragmentation into 30 neighborhood identities is itself a behavioral finding — disinvestment broke the civic connective tissue unevenly. The TO-39 cluster uniformity marks a single systemic cause across a wide geography.
Read the Friction table top to bottom: as Trust Orientation falls, Friction rises — and DELTA rises with it. The neighborhoods wasting the most on conventional delivery are also the ones gaining the most from trust-network routing. These are the same neighborhoods.
Read the Spend/Avoid Map alongside it: the AVOID signal on Lance is not a low-engagement finding. TO 58 is the highest score in Kenosha. It is the most expensive mistake in the geography — a dollar directed at a high-TO Guardian neighborhood doesn't bounce. It lands on the opposition side. The Friction read and the fit-type read together are what makes the budget decision defensible.
The implication for any program, campaign, or investment deploying across multiple neighborhoods: you are concentrating conventional spend in the geographies where it wastes the most. CI maps the routing that recovers it — and quantifies the recovery in budget terms before a dollar moves.
Dollar figures apply an illustrative $50K per-neighborhood and $350K aggregate spend. These are directional findings — not verified actuarial data. Actual waste and recovery figures require CI deployment against the specific program budget and geography. The behavioral relationships — Friction rises as TO falls, DELTA rises as Friction rises, fit type determines what TO means — are consistent across every deployment context to date.
Everything derived from aggregate behavioral signals. Zero personally identifiable information collected, stored, or used at any stage. No individual named, tracked, or profiled. The unit of analysis is always the community.
Five independent public behavioral streams: civic participation, health program utilization, economic behavior, electoral engagement, infrastructure use. All auditable. All observable. The direction is certain; the magnitude is anchored and improving.
An Avoid call is never a "do not help" call. Every neighborhood placed in Avoid for outreach receives a corresponding Serve commitment — unconditional civic investment keyed to actual community needs, not to how it votes.
Individual data is noisy. Community behavioral architecture is stable, measurable, and in the right hands — a genuine instrument of budget discipline, campaign efficiency, and community power.