Strategy overview · Sep 28, 2026 · Chris Machetto
Residential Services Platform — Product & Technology Strategy
Build one client-owned platform — Next.js web + a native offline-first provider app on Supabase Postgres — and launch it as a managed cleaning service in a few pilot buildings in about 20 weeks, for roughly 3,000–4,400 hours of work and about $500–$600 a month in technology. Grow it into a vendor marketplace, sales system and AI-assisted operation over the following 6–7 months.
CTO recommendation
| Question | Answer |
|---|---|
| What exactly should we build? | A residential services operating platform: instant quote and booking for residents, a native provider app that works without signal, and a command center for dispatch, payments and support — multi-tenant from day one. |
| What first? | The loop quote → book → clean → pay → rate → rebook, for residents in 1–3 pilot buildings, with recurring plans. |
| What not to build initially? | Open marketplace (vendor profiles, bidding, vendor-set prices), AI voice, custom CRM, native customer app, property-manager portal, PMS integrations, AI dispatch or pricing. |
| Which architecture? | TypeScript monorepo; Next.js on Vercel for marketing, customer app and command center; React Native (Expo) provider app; Supabase Postgres with Row Level Security; PowerSync for offline; Stripe; Twilio; HubSpot; PostHog; Sentry. Details in the Architecture tab. |
| React / Next.js / Webflow? | React and Next.js: yes. Webflow: not at launch — the landing page lives in Next.js; Webflow only if marketing must self-edit visually. |
| What does the client own? | Everything: code, database, cloud, domains, Stripe, Twilio, AI, analytics and app-store accounts, credentials and documentation — opened under the LLC from day one. |
| Outsource or integrate? | Outsource the build through Phase 3 while the client owns every account; integrate payments, messaging, voice, CRM, sync, analytics and monitoring rather than building them. |
| Realistic MVP timeline | 20 weeks to pilot launch (16 optimistic, 26 conservative); stable pilot by week 26. |
| Realistic full-platform timeline | About 50 weeks, roughly 11–12 months (9 to 16 months range). |
| Development cost | MVP 3,000–4,400 hours; full platform 5,700–8,650 hours — $300k–$440k and $570k–$865k at a $100/h blended rate. |
| Monthly technology cost | ~$520 at pilot, ~$570 at 100 customers, ~$1.6k at 1,000, ~$6.4k at 10,000, ~$27k at enterprise scale — plus ~3.1% payment processing. |
| Biggest technical risks | Offline sync in real buildings; cross-tenant data exposure; an undecided provider model; scope creep into the MVP. |
| In-house or outsourced? | Outsourced senior team now; client hires a technical product owner in Phase 0 and a first engineer around Phase 4 to take ownership. |
Decisions before development starts
These change scope, cost or architecture; each should be closed by the end of Phase 0.
| # | Decision | Options | Why it matters |
|---|---|---|---|
| 1 | Provider model | W-2 employees · independent contractors · vendor companies · mix | Stripe Connect in or out of the MVP (~3 weeks); payroll vs payouts; legal exposure |
| 2 | Who pays | Residents per visit · buildings on contract · both | Invoicing, pricing and the tenant model |
| 3 | Pilot buildings and market | 1–3 named buildings; launch city and state | Connectivity testing, zones, legal requirements, SMS registration |
| 4 | Pricing model | Flat by home size · hourly · square footage; add-ons; frequency discounts | The price engine and quote UX |
| 5 | White-label ambition | Single operator · license the software to other cleaning companies within 24 months | Strictness of isolation, tenant billing, branding |
| 6 | Marketing site editing | Next.js (recommended) · Webflow | Stack, cost, who edits |
| 7 | CRM | HubSpot (recommended) · another CRM · build | Sales workflow and data sync |
| 8 | Provider app form | Native (recommended) · PWA for a small Android-heavy pilot | Offline reliability and cost |
| 9 | Budget and launch date | Range and hard date, if any | Team size and scope |
| 10 | Legal entity and account setup | LLC + EIN, admin mailbox, 1Password vault | Every account is opened on it; SMS registration and Stripe need it |
Immediate admin actions: send the NDA; confirm LLC formation; draft the MSA/SOW with IP assignment.
Missing from the discovery call
Ask these on the next call; each answer replaces an assumption in the Product tab.
- Company name, current stage (operating today or pre-launch), and current customer and provider counts
- Which buildings “go down” — names, floors, and whether it's cell signal, building Wi-Fi, or both
- Services beyond standard cleaning at launch (deep, move-out, laundry, other home services) and the 12-month service roadmap
- Current pricing sheet and average ticket
- How cleaners are engaged and paid today; insurance and background-check requirements
- Who the sales team sells to (residents, property managers, both); expected number of reps in year one
- Current tools (booking software, CRM, spreadsheets, phone system) and data to migrate
- Relationship with buildings: contracts, revenue share, amenity programs, access arrangements
- 12-month targets: customers, jobs per month, buildings, markets
- Budget range, funding timing, and any fixed launch date
- Who on the client side decides, who runs ops during the pilot, and who owns the platform after handoff
- Brand status: existing name, domain and visual identity, or new brand (a proposed name and logo are in the AI-builder SOP)
- What “build remote” meant: remote delivery team, or remote work features in the product