Skip to content

Post-v3 Operating Roadmap

Audience: maintainers and product planning. Confidential strategy reference.

This document tracks the areas that can remain missing after the v3 feature scope is complete — when the project shifts from “feature development” to productization, trust, distribution, sales, and sustainability. See the Roadmap, v2 → v3 backlog, and Innovation & Moat for everything that comes before this phase.

By v3, Roomful is technically strong: Flutter/Dart + web SDKs, self-hostable relay, presence, cursors, comments, locks, shared state, field-level collaboration, AI-agent presence, approval flow, history/replay, and a room inspector.

After that point, the product does not primarily need more big features — it needs production reliability, enterprise-sale readiness, cloud monetization, documentation, a demo/template ecosystem, customer validation, and continuous SDK maintenance.

Working ratio after v3: ~70% hardening / reliability / support / marketing · ~20% customer-driven improvement · ~10% strategic innovation.

QuestionObjective answerPractical consequence
Need new big features?Short term, no.Do a 3–6 month feature freeze.
Is passive maintenance enough?No.Need reliability, docs, demos, support, cloud beta, feedback loop.
Is marketing enough alone?No.Marketing must be backed by live demos, trust metrics, production references.
Biggest post-v3 risk?Trust and distribution.Developer-tool sales need trust, onboarding, and case studies.

The central question becomes: who runs this in production, how much do they trust it, how fast do they integrate it, can they self-host it, and do they pay for it?

Flutter/Dart + web SDK · self-hostable relay + basic production deploy options · presence, live cursors, viewport, selection, shared state · comments, locks, field-level presence, record locking · AI-agent presence, action timeline, approval flow · room inspector, history/replay, privacy/redaction base · documentation, examples, minimal SDK stability.

AreaWhy it can be missingRequired outputs
Enterprise Trust & ComplianceCompanies need trust, security, privacy, legal to deploySSO/SAML, SCIM, audit-log export, data retention policy, security model, SOC 2 prep, DPA/GDPR docs, vulnerability disclosure
Cloud & Monetization LayerTurning open-source core into revenueHosted relay, dashboard, usage analytics, API keys, billing, rate limits, team management, webhooks, SLA tiers
Production ReliabilityRealtime infra must be reliable and measurableReconnect recovery, latency metrics, load tests, chaos tests, backpressure, retry policies, event dedup, observability
Customer ValidationProof it creates value in real useFirst 5 production users, case studies, interview loop, feedback board, integration-friction measurement
Distribution & GTMGetting the tool to the right peopleLaunch plan, Flutter community, Hacker News / Product Hunt, outbound, content engine, template demos
Docs & OnboardingValue must be seen in minutesQuickstarts, SDK recipes, copy-paste components, troubleshooting, migration guides, auth examples
Ecosystem IntegrationsFit into existing developer stacksSupabase, Firebase, Auth0, Clerk, Postgres, Redis, Yjs, Tiptap, FlutterFlow, React Flow
AI Coding Agent AdoptionDevelopers set up SDKs via AI assistantsMCP server, Cursor/Claude Code instructions, install commands, generated integration plans
Open-source GovernanceSustainable community and contributionContribution guide, RFC process, issue templates, roadmap voting, maintainer policy, security policy
Support & OperationsManaging paying users’ problemsSupport SLA, incident playbook, status page, logs, escalation path, customer-success loop
  • Security model: which data lives on the client, relay, storage, and logs?
  • Threat model: malicious clients, room hijacking, replay attacks, token leakage, event poisoning.
  • JWT validation guide: issuer, audience, expiry, key rotation, permission claims.
  • Rate limit & abuse protection: per room / user / IP / token / event-type limits.
  • Vulnerability disclosure policy and security contact.
  • Dependency scanning, SCA, secret scanning, release signing.

Enterprise purchasing cares more about documentation and trust packs than raw features:

  • Data retention policy (ephemeral/durable split, history duration, deletion flow).
  • DPA / GDPR baseline docs; PII redaction guide (what never hits history, what gets masked).
  • Audit-log export format (JSON/CSV; time range, actor, room, action, result).
  • Data-residency strategy (EU, US, self-host, private cloud).
  • SOC 2 readiness checklist (a prep pack first, not a certificate).
FeaturePriorityValue toNote
SSO / SAMLP1Enterprise SaaS teamsRequired for Cloud/Enterprise plan
SCIMP2Large teamsUser lifecycle management
Audit exportsP1Compliance-heavy teamsShould merge with the AI agent action timeline
Admin rolesP1Team accountsOwner, admin, developer, viewer
Private cloudP2Security-sensitive customersSellable with self-host support
Data residencyP2EU customersCritical for managed cloud
Custom retentionP2EnterpriseAlso controls history/replay cost

Managed relay, dashboard, and observability for teams that don’t want to self-host:

  • Hosted relay: low-friction managed endpoint running in minutes.
  • Project & room dashboard: active users, event count, latency, errors.
  • API keys & environments: dev/staging/prod separation.
  • Usage analytics: rooms, MAU, events, storage, replay minutes.
  • Billing: free / startup / team / enterprise contact-us.
  • Webhooks: comment created, approval requested, lock acquired, agent action completed.
  • Logs and replay retention settings.
PackageContentTarget userRevenue logic
Open-source CoreSDK, basic relay, presence, cursors, shared state, locksDevelopers, indie teamsAdoption & community
Pro CloudHosted relay, dashboard, logs, team management, usage analyticsStartups, SaaS teamsSubscription
Enterprise CloudSSO, audit logs, retention, SLA, data residencyScale-up / enterpriseHigher ACV
Enterprise Self-hostPrivate deploy, support, onboarding, security reviewPrivacy-sensitive orgsAnnual support contract

Revenue-model risks: open-source only may yield no revenue; cloud too early dilutes team focus; cloud too late lets users self-host without paying; an unclear enterprise-support model loses security-sensitive teams after adoption.

Trust is not a feature in realtime products — it is the foundation.

Reconnect recovery (mobile network switch, tab sleep, flaky Wi-Fi) · event deduplication (idempotent re-delivery) · backpressure (cursor spam, high-frequency events, room fanout limits) · rate limiting (per event type, soft/hard) · persistence strategy (ephemeral/durable split) · replay safety (disable or redact for sensitive rooms) · load tests (10 / 100 / 1,000 concurrent rooms) · chaos tests (relay restart, Redis unavailable, network drop, token expiry) · metrics (p50/p95/p99 latency, reconnect count, dropped events, memory) · incident playbook (outage, data loss, latency spike, abuse).

Status page (cloud relay health, API, dashboard, webhooks) · customer-facing incident updates · internal runbooks (restart, rollback, mitigation, comms) · support tiers (community, pro, enterprise) · SLA definition (at least a response-time SLA).

Goal — the first 10 minutes: a developer sees presence/cursors between two browsers or a Flutter simulator within 10 minutes of installing.

  • Copy-paste quickstart: React, Flutter, self-host relay.
  • Hosted sandbox endpoint (run a demo without a local relay).
  • Clear error messages (token invalid, room forbidden, relay disconnected).
  • Debug overlay (connected, latency, participant count, dropped events).
  • Starter templates (collaborative form, CRM lock, comments overlay, AI agent status).
Docs categoryContentPriority
QuickstartsFlutter, React, self-host, cloudP1
RecipesField presence, record lock, comment pins, approval flowP1
ArchitectureProtocol, relay, persistence, auth, room modelP1
SecurityJWT, permissions, redaction, event privacyP1
OperationsDocker, Kubernetes, Redis, logs, monitoringP1
MigrationSupabase, Firebase, Ably, custom WebSocket, Liveblocks-like patternsP2
AI agentsAgent presence, action timeline, approval workflowP2

Flutter collaborative form (field presence, field lock, comments) · Web + Flutter same room (cross-platform protocol) · CRM record locking (real B2B data-conflict problem) · AI agent in-app collaboration (agent presence, proposed change, approval) · self-host deployment (Docker Compose relay + Redis + sample app) · Supabase/Firebase auth examples.

Product-language pub.dev/npm descriptions · Flutter community content (Medium, Dev.to, Reddit, Discord/Slack) · a GitHub README that shows value prop + demo GIF + quickstart + self-host link in one screen · a use-case-driven launch page (not an API list) · comparison pages (vs Firebase/Supabase custom realtime, vs Liveblocks for Flutter, vs Ably Spaces) · an outbound list of Flutter B2B SaaS teams and agencies.

IntegrationValuePriorityNote
SupabaseFast adoption for Auth+Realtime Flutter teamsP1Ship auth-claim examples
FirebaseOne of Flutter’s most common backendsP1Position as collaboration layer, not a Firestore replacement
Auth0 / ClerkSaaS auth integrationP2Start with a JWT guide
RedisSelf-host relay scaleP1Presence fanout and coordination
PostgresDurable comments/historyP1Needed for cloud and self-host
Yjs / CRDTAdvanced collaborative editingP2Post-MVP plugin
Tiptap / LexicalWeb editor collaborationP2Credibility in the web market
FlutterFlowFlutter low-code ecosystemP3Interesting distribution channel
React FlowCanvas/diagram use caseP3Pairs well with the AI-agent collaboration demo

Assume developers increasingly set up SDKs through an AI coding assistant, not only docs:

  • MCP server: Roomful docs, API reference, and examples served as context.
  • Cursor / Claude Code recipes: /add-roomful, /add-live-cursors, /add-record-locking.
  • Framework-specific integration prompts: Flutter, React, Next.js, Supabase, Firebase.
  • Generated integration checklist: SDK install, provider setup, auth token, relay config, production notes.
  • AI-safe docs: short, correct, versioned, copy-paste-runnable code examples.

This lowers adoption cost significantly — a DX advantage, not just marketing.

Firebase Realtime/Firestore presence → Roomful presence/locks/comments · Supabase Presence/Broadcast → Roomful room model · Ably Presence/Channels → Roomful protocol + component layer · custom WebSocket → Roomful relay/protocol · Liveblocks-like API mapping (web migration) · DIY cursor → LiveCursorsOverlay / Flutter overlay.

PlatformSupportTest coverageNote
Flutter iOSOfficialIntegration + reconnectMobile network matters
Flutter AndroidOfficialIntegration + reconnectBackground/foreground tests
Flutter WebOfficialBrowser matrixDOM-less Flutter overlay matters
React WebOfficialBrowser matrixWeb SDK credibility
Node.jsOfficialServer SDK testsNeeded for agent/system participants
React NativeFutureSpikeMobile-native expansion
Swift / KotlinFutureSpikeEnterprise/native path

CONTRIBUTING.md (local setup, tests, packages, release flow) · SECURITY.md (disclosure, supported versions) · CODE_OF_CONDUCT.md · RFC process for big changes · versioning policy (breaking change, deprecation, migration window) · maintainer guide (release owner, review rules, publish rights) · roadmap voting (community input, product direction preserved) · good-first-issues.

1 Flutter B2B SaaS app (field presence + record locks) · 1 internal dashboard (web + comments + live users) · 1 AI-agent SaaS (agent presence + approval flow) · 1 self-host customer (Docker/Redis/Postgres deploy validation) · 1 agency/studio (integrating Roomful into client projects).

MetricWhy it mattersTarget
Time-to-first-presenceOnboarding quality< 10 minutes
Time-to-production-demoReal integration effort< 1 day
SDK error rateStabilityLow and measurable
Reconnect success rateMobile/web reliability> 99% target
Docs completionSelf-serve adoptionQuickstart completion rate up
Support tickets per integrationDX problem signalDownward trend
Self-host setup successEnterprise readinessOne-command successful setup

A productization/growth cadence rather than new big features:

PeriodThemeMain outputsSuccess signal
Months 1–2Feature freeze & stabilizationBug fix, tests, reconnect, docs cleanup, self-host guideQuickstart and self-host flows are smooth
Months 3–4Demo & validation3 core demos, 5 design partners, case-study interviewsFirst production pilots
Months 5–6Cloud betaHosted relay, dashboard, API keys, logs, usageFirst paid-beta signal
Months 7–8Enterprise trust packSecurity docs, audit logs, retention, admin rolesCan pass a security review
Months 9–10Distribution scaleLaunch, comparison pages, content engine, outboundSteady inbound + GitHub/pub.dev/npm growth
Months 11–12Customer-driven roadmapCustomer-feedback improvements, support ops, pricing refineRetention and payment validation
  • [P1] Write Roomful security model and threat model
  • [P1] Add JWT auth guide with permission claims
  • [P1] Add rate limiting and abuse protection
  • [P1] Create data retention and redaction policy
  • [P2] Add vulnerability disclosure and supported-versions policy
  • [P2] Prepare SOC 2 readiness checklist
  • [P1] Build hosted relay project/environment model
  • [P1] Add API-key management and usage metering
  • [P1] Build a basic dashboard for rooms, users, latency, and errors
  • [P2] Add billing primitives and plan limits
  • [P2] Add webhooks for comments, approvals, locks, and agent actions
  • [P1] Rewrite quickstarts for Flutter, React, self-host, and cloud
  • [P1] Create copy-paste recipes for field presence and record locks
  • [P1] Add a troubleshooting guide for common relay/auth/reconnect issues
  • [P2] Add migration guides from Firebase, Supabase, Ably, and custom WebSocket
  • [P2] Add an AI-agent integration guide and approval-flow examples
  • [P1] Publish Flutter collaborative form demo
  • [P1] Publish Web + Flutter same-room demo
  • [P1] Publish CRM record-locking demo
  • [P1] Publish AI-agent collaboration demo
  • [P2] Create comparison pages and launch assets
  • [P2] Build an outbound target list for Flutter/B2B SaaS teams
  • [P1] Add Supabase and Firebase auth examples
  • [P2] Add Auth0/Clerk JWT examples
  • [P2] Create a Roomful MCP server prototype
  • [P2] Add Cursor/Claude Code integration instructions
  • [P3] Explore a FlutterFlow integration path
  • [P3] Add a Yjs/CRDT adapter spike
RiskImpactSignalMitigation
Wide feature set but no adoptionHighDemos watched, installs lowFeature freeze + onboarding + customer interviews
Cloud too early dilutes the teamMed-highSDK bug backlog growsLimited cloud beta; hosted-relay minimum first
Self-host is hard to set upHighDeploy-focused support requestsOne-command Docker Compose + K8s guide + logs
Enterprise trust barrier not clearedHighStuck in security reviewTrust pack, audit logs, retention, security docs
Web competitors close the gapMediumLiveblocks/Velt Flutter or self-host moveStrengthen AI-agent + self-host + Flutter-first message
Maintenance load risesMediumIssue response time growsGovernance, test automation, community triage

After v3, the big feature gaps largely close. What remains is trust, productization, distribution, cloud, enterprise readiness, documentation, customer validation, and support operations — not more technical features. The correct model is not passive maintenance; it is stabilizing the product, easing self-serve adoption, winning production users, opening a cloud beta, building the first case studies, and making small, high-impact, customer-driven improvements.

Roomful must move from “what else can we add?” to “who uses this, how much do they trust it, how do they scale it, and do they pay for it?”