Module 6: Economic Model and Sustainability¶
How do you fund civic infrastructure without compromising it? Revenue models, cost structures, and the strategic constraints that shape the economics of open civic coordination.
Prerequisites: Modules 1-5. You should understand the full system architecture, trust model, and jurisdiction design.
The Fundamental Tension¶
Civic infrastructure has a paradox at its core: the people who need it most can't pay for it, and the people who can pay for it have interests that might corrupt it.
If you charge residents, you exclude the people civic coordination is supposed to serve — the ones who don't have lobbyists, who can't attend Tuesday afternoon meetings, who've never been asked what they think about the zoning change happening on their block.
If you take money from developers, real estate interests, or political groups, you've introduced exactly the incentive misalignment you're trying to eliminate. The funder's interests inevitably shape the platform, even if only through what gets prioritized.
If you rely on government funding, you're dependent on the institutions you're trying to hold accountable. A city won't fund a tool that makes it easy for residents to organize opposition to the city's plans.
If you rely on foundation grants forever, you die when the grant cycle ends. Every civic tech startup that depended on foundation money followed the same arc: launch, pilot, lose funding, shut down.
So the question isn't just "how do we make money?" It's: how do you fund civic infrastructure in a way that doesn't compromise its integrity, doesn't exclude residents, and doesn't depend on any single funding source?
What Does Civic Coordination Infrastructure Cost?¶
Before talking about revenue, let's be concrete about costs. One of the structural advantages of protocol-based civic infrastructure is that per-jurisdiction costs are genuinely low.
Per-Jurisdiction Costs¶
| Component | Role | Monthly cost |
|---|---|---|
| Civic data storage | PostgreSQL + pgvector | ~$25 |
| Relay storage | Coordination database | ~$25 |
| MCP server | Serverless query handling | ~$2-5 |
| Relay compute | Serverless coordination | ~$2-5 |
| Vector indexing | GPU, on-demand | ~$5-10 |
| Blob storage | PDFs, audio | ~$1-3 |
| AI API calls | Resident drafting/chat | ~$5-10 |
| Total per jurisdiction | ~$65-80/month |
That's roughly $800-1,000/year per city. Not per user — per city.
Comparison With Existing Civic Engagement Costs¶
| Existing approach | Typical cost |
|---|---|
| Agenda management contract (Granicus/Legistar) | $20,000-80,000/year |
| Public engagement consultant | $50,000-150,000/project |
| Community survey | $10,000-50,000 each |
| Civic engagement platform (Bang the Table, etc.) | $15,000-40,000/year |
| Full-time community engagement staff | $80,000-120,000/year salary |
Protocol-based infrastructure at ~$1,000/year per city is two orders of magnitude cheaper than these approaches. This matters — it means the system doesn't need much revenue to be sustainable, and city adoption isn't gated by budget approval processes for large contracts.
The Real Cost: Data Ingestion¶
The real cost isn't infrastructure — it's data ingestion. Getting a new city online requires:
- Identifying and configuring the city's agenda management system (Legistar, Novus, PrimeGov, etc.)
- Building or adapting extractors for agenda PDFs
- Ingesting meeting transcripts (audio → text pipeline)
- Ingesting municipal code
- Configuring vector embeddings
- Setting up attestation for the jurisdiction
This takes weeks to months of engineering work per city. Once a city is ingested, the ongoing infrastructure cost is trivial. Data ingestion is the scaling bottleneck.
Revenue Models for Open Protocols¶
The protocol is open and permissionless. Resident participation is free. So where does revenue come from?
History has a clear pattern. The most successful open protocol businesses all follow the same model:
| Open thing | Free | Paid service |
|---|---|---|
| Linux | OS kernel | Red Hat Enterprise (support, managed) |
| WordPress | CMS software | WordPress.com (hosting, managed) |
| Elasticsearch | Search engine | Elastic Cloud (managed, enterprise) |
| Kubernetes | Container orchestration | GKE, EKS, AKS (managed K8s) |
| Email (SMTP) | Protocol | Gmail, Outlook (managed email) |
| DNS | Protocol | Cloudflare, Route53 (managed DNS) |
The pattern: The protocol enables the ecosystem. The managed service captures value by making the protocol easy to use reliably. You're not paywalling the protocol — you're selling convenience, reliability, and expertise.
For civic coordination:
Free (the protocol):
Nostr civic event kinds (open standard)
Relay software (open source)
Voice, attestation, subscription protocols
Paid (the service):
Managed relay hosting (turnkey civic coordination)
Data ingestion (structuring a city's civic data)
Multi-jurisdiction data access (for professionals)
Premium features (real-time feeds, analytics, bulk export)
Anyone can run their own relay, ingest their own data, and build their own tools. The protocol enables it. But most cities, newspapers, and civic organizations won't — they'll pay for it to just work.
Customer Segments¶
Different customers pay for different things because they derive different value.
Cities¶
Value: Structured civic engagement data — real-time constituent sentiment, attestation-verified voice counts, searchable meeting/decision history.
Why they'd pay: It's 10-50x cheaper than existing engagement platforms, and it produces defensible data. A council member who can say "42 attested residents support this — here are their voices" has a more defensible vote than one who relies on whoever showed up to a 7pm meeting.
Journalists and Media¶
Value: Structured civic data across multiple jurisdictions — voice counts, trending issues, anomaly detection, historical patterns.
Why they'd pay: Local journalism is dying because individual reporters can't cover the volume. A journalist with civic coordination data across 20 cities in a region can spot patterns no single reporter could: "Housing opposition is spiking in 6 cities simultaneously — same developer involved in all of them." That's a story that drives accountability.
Civic Organizations¶
Value: Issue-specific engagement data across jurisdictions and government levels. "How is housing sentiment trending across the county?" "How many people are committed to attending the hearing?"
Why they'd pay: Organizations spend enormous effort manually tracking civic engagement — monitoring agendas, counting attendees, coordinating member action. Structured civic data automates the intelligence layer.
Developers and Civic Tech¶
Value: API access to structured civic data and coordination signals. Build applications on top of the data layer rather than building data infrastructure from scratch.
Why they'd pay: Building civic data infrastructure is months of work per city. An API provides it as a service.
Residents¶
What they pay: Nothing. Ever.
This is non-negotiable. The moment you charge residents for civic participation, you've created a two-tier civic system. The protocol is free. The relay is free for residents. Subscribing, voicing, commenting — free.
Revenue comes from professional users who derive commercial or organizational value from civic coordination data.
The Moat Question¶
If the protocol is open and the software is open source, what stops someone from copying everything?
What's NOT a Moat¶
- The AI (anyone can use Claude or GPT)
- The relay software (open source)
- The protocol (open standard)
- The UX (anyone can build a better frontend)
What IS a Moat¶
Data ingestion per city. Weeks of engineering work to structure each city's civic data. This doesn't transfer — you have to do it for every city.
Attestation network. Attested residents are a network effect. Hundreds of attested residents whose voices, attestations, and engagement history live on a specific relay. Moving requires re-attestation.
Voice history. The accumulated record of who cared about what, over time. This is the civic coordination layer itself — and it's most valuable to the relay that has the most of it.
Institutional relationships. The city staff who trust this data. The journalists who cite it. The civic organizations that integrate it into their workflows.
The moat deepens with each city, each attested resident, each voice. It's a network effect, not a technology advantage.
Why Previous Civic Tech Failed¶
| Project | What they built | Why they died |
|---|---|---|
| EveryBlock | Hyperlocal news aggregation | No revenue model, acquired by NBC → shut down |
| Neighborland | Community input platform | Platform dependency, couldn't retain cities |
| SeeClickFix | Issue reporting | Acquired, narrowed to 311 tool |
| Change.org | Petition platform | Engagement optimization → lost civic focus |
| Nextdoor | Neighborhood social network | Ad-supported → optimizes for engagement/outrage |
Common failure patterns:
- VC-funded with engagement metrics → optimizes for the wrong thing (Nextdoor, Change.org)
- Foundation-funded with no revenue transition → dies when grant ends (EveryBlock)
- Platform, not protocol → users locked in, no ecosystem, single point of failure (all of them)
Protocol-based civic infrastructure avoids each failure mode: - Foundation-funded to start, but costs are low enough that modest commercial revenue sustains it - No engagement metrics to optimize (the relay has no algorithm) - Protocol, not platform — the ecosystem can outlive any single operator
The Sustainability Path¶
Civic coordination infrastructure follows a natural progression from grant-funded pilot to self-sustaining service:
Foundation bootstrap → Prove the coordination hypothesis with real residents. Costs are ~$1,000/year for a single city. Grants cover this easily.
Early revenue → Professional users (journalists, civic organizations) begin paying for multi-jurisdiction data access. Revenue doesn't need to cover all costs yet — grants still help with expansion.
Sustainability → Infrastructure costs covered by revenue from a mix of cities, media organizations, and civic tech API customers. Grants fund geographic expansion (new city ingestion).
Ecosystem → Multiple independent operators run relays for their jurisdictions. The protocol has independent life beyond any single organization. Revenue comes from managed service and data access.
The key insight: because per-city costs are ~$1,000/year, the revenue threshold for sustainability is low. A handful of paying professional customers covers the infrastructure for dozens of cities.
The Partnership Model¶
Civic coordination infrastructure has two distinct capabilities:
- Coordination protocol — relay, voice, attestation, edge intelligence
- Civic data ingestion — structuring government data per jurisdiction
These don't have to be done by the same organization:
Single operator:
Protocol + data ingestion per city
Scaling bottleneck: engineering time per jurisdiction
With data partners:
Operator provides: coordination protocol, relay, attestation, MCP
Partner provides: structured civic data across jurisdictions
Scaling bottleneck: removed (partner already has the data)
Potential data partners include civic data aggregators, state transparency organizations, and media organizations that already monitor government meetings. The coordination protocol is the novel contribution. Civic data extraction is valuable but not unique — partnerships that leverage existing data infrastructure let protocol operators focus on what nobody else is building.
The Anti-Revenue: What Must Never Be Sold¶
As important as what generates revenue is what must never be monetized:
Never sell resident data. The system is designed so that no entity has a complete picture of who participates and how. Even the pseudonymous data (which keys voice on which entities) is never sold.
Never sell algorithmic placement. There's no algorithm. The relay routes events and counts voices. There's no "promoted initiative" or "sponsored voice" concept.
Never gate resident participation. Voicing, commenting, subscribing — free at every relay for every attested resident. The gate is attestation (physical presence), not payment.
Never optimize for engagement. No metrics on time-on-site, click-through, notification open rates. These metrics create pressure to make the system addictive rather than useful. Their absence is a feature.
Why These Constraints Are Strategic, Not Idealistic¶
This isn't idealism — it's strategy. The moment an operator sells resident data or introduces engagement optimization, they become Nextdoor. The differentiation disappears. The trust disappears. The institutional relationships that make the moat dissolve.
The constraints create the trust. The trust creates the value. The value creates the revenue.
- Cities pay because they trust the system isn't manipulating their residents
- Journalists pay because the data is unbiased
- Civic organizations pay because the platform serves their mission, not an advertising model
See Also¶
- Relay Overview — How the relay works
- AI Proxy — Privacy-preserving AI access and billing architecture
Questions to Explore¶
On revenue: - "Compare this revenue model to Mastodon's funding model (Patreon/donations). Why might commercial service revenue be more sustainable?" - "What's the minimum number of paying customers needed to cover infrastructure costs for 10 cities?" - "How would pricing change with a data aggregation partner?"
On the moat: - "If the protocol is open, what prevents a well-funded competitor from replicating the data ingestion and undercutting on price?" - "How strong is the attestation network effect? Could attested residents easily migrate to a competing relay?" - "Compare this moat to Wikipedia's moat. Both are open, both rely on community contribution. What's similar?"
On sustainability: - "What happens to the system if the founding operator ceases to exist? How much of the value survives?" - "What's the risk that city government customers pressure an operator to add features that compromise neutrality?" - "How should an operator handle a situation where its largest funder disagrees with the anti-revenue constraints?"
On failure modes: - "Walk through the specific scenario where a civic coordination project follows the EveryBlock trajectory. What would cause it? How does the protocol architecture prevent it?" - "What if no cities are willing to pay? What's the plan B revenue model?" - "How does the protocol approach avoid the Nextdoor failure mode of optimizing for engagement over civic value?"