Local Tutor Marketplace App in 2026: Market Size, Revenue Precedents, Cost to Build
Last updated: 29 April 2026Category: MarketplaceData source: MyAppTemplates analysis of 2026 public SOW benchmarks and shipped-app case studies.
Executive Summary
What it is. A two-sided mobile app that connects parents (or adult learners) with vetted local tutors for in-person or video lessons. Tutors list subjects, hourly rates, and availability; parents search by subject, location, level, and price; the app handles booking, payment, and reviews and takes a transaction fee on each lesson.
Who pays. Parents of school-age children paying $30–$120 per hour for tutoring, and adult learners paying for languages, music, or test prep. Tutors are the supply side and rarely pay listing fees in 2026 — the platform clips a 15–25% transaction fee on each completed lesson. That single fee structure is what makes this category work: every other monetisation path has been tried and lost to it.
Why now. Post-pandemic learning loss is still being priced into family budgets in 2026; private-tutor demand in the US and UK is up sharply versus 2019, and incumbents (Wyzant, Superprof, Care.com's tutor vertical) are national, web-first, and weak on local geo-search and mobile UX. A regional, mobile-first, neighbourhood-tight marketplace is a defensible wedge — that's the entry door, not 'beat Wyzant globally'. The boilerplate's $199 foundation removes the week-one infrastructure work; Claude Code wires the marketplace-specific features against it.
Build cost by scope
Tutor marketplace: scope variants from MVP to scale
Five honest scope tiers — pick one, ship it, then climb.
Every DIY build starts with the same flat boilerplate fee:$199 one-time — column below shows marginal Claude Code API spend on top
#
Scope variant
What's in it
Agency Quote
+ AI Spend
Savings
Build Time
1
Lean MVPOne city, manual matching
Auth, tutor profiles, search by subject, manual booking via in-app message, Stripe payments outside Connect (you reconcile manually), reviews
$45k–$70k
$120
99.7%
5–7 days
2
Solo launchPublic app, automated booking
Lean MVP + calendar availability, automated booking flow, push notifications for lesson reminders, basic geo-search by postcode
$70k–$100k
$185
99.7%
8–10 days
3
Solo at 1k usersConnect payouts, ID checks
Solo launch + Stripe Connect for tutor payouts, Persona ID verification, in-app chat (Workers Durable Objects), refund/cancellation flow
Four shipped tutor marketplaces with real revenue. Ranges are estimated from public App Store rank and Sensor Tower / AppFigures benchmarks, 2026, plus disclosed company numbers where available — they are bands, not exact figures.
Spotlight Build
Wyzant — US national, web-first
ModelHourly transaction fee (~25%)
Estimated MRR$8M–$15M MRRDisclosed historical GMV plus 2026 category growth — wide band intentionally
Why it mattersProves transaction-fee model scales. Weak on mobile UX and local geo-search — that's the wedge.
Spotlight Build
Superprof — global, France-led
ModelTutor subscription + connection fees
Estimated MRR$5M+ MRRPublic revenue disclosures from parent company filings, 2024–2025
Why it mattersTutor-pays model works at scale but creates supply-quality issues — many ghost listings. A demand-side-paid (transaction-fee) model wins on trust.
Why it mattersVertical wedges (one subject, one geography) get to product-market fit faster than horizontal-everything launches.
2. Market size and demand signal
The category has visible, sustained demand — not a TikTok-trend spike that disappears in six months.Head-keyword volume (US + UK, 2026): 'tutor near me' ~165k/mo, 'maths tutor' ~90k/mo, 'find a tutor app' ~22k/mo. Category growth in private tutoring spend is estimated at 8–12% YoY in English-speaking markets through 2026, with the strongest growth in primary-school maths and exam prep (SAT, GCSE, 11+).
Demand signal
Unmet-need evidence
App Store reviewsWyzant and Superprof both sit at 3.5–4.0 stars with persistent complaints about ghost tutors, no-shows, and weak local search.
Reddit / parent forumsr/Parenting and Mumsnet have weekly threads asking 'best app to find a local maths tutor' — the answer is usually a Facebook group, which is the gap.
TAM proxyUS private tutoring market estimated at $10B+ in 2026; even a 0.1% city-level slice with 20% take rate is a $2M ARR business.
3. What to ship in week one
Solo founders fail this category by building all of Wyzant on day one. Don't. Ship the smallest thing that lets one parent in one city book one tutor for one lesson and pay through the app.
Week-one scope
Lean MVP feature list
AuthPhone OTP — already in the boilerplate at app/(auth)/phone-register.tsx and verify-code.tsx. Use the two-variant pattern for parent vs tutor.
Tutor profileName, subjects, hourly rate, postcode, 200-word bio, one photo. Schema-first via Drizzle.
SearchPostcode + subject + price filter. SQL LIKE is fine at MVP — Algolia comes at 1k users.
PaymentStripe in single-account mode; you reconcile and pay tutors via bank transfer for the first 50 lessons. Connect comes at the Solo-at-1k tier.
ReviewsFive-star + 100-char comment after lesson completion. No photo uploads.
Differentiation angles
Wedges that still work in 2026
HyperlocalOne city, walking-distance search, in-person bias. Wyzant and Superprof are national and weak here.
Vertical subjectJust 11+ exam prep, or just GCSE maths, or just piano. Vertical wedges hit PMF faster than horizontal launches.
Tutor-quality biasCurated supply with required ID + qualification verification. Premium 30% take rate, lower volume, higher trust.
Parent group integrationBuilt for school WhatsApp / parent-group sharing. Referral mechanics that match how parents actually find tutors.
How to build this against the boilerplate
Five steps from clone to first paying lesson. Assumes you've already bought the $199 boilerplate.
1
Day 1 — Clone and run
Clone the repo, run the mobile app on a simulator, deploy the Workers backend. Auth, billing adapter, and CI are already working — that's the week-one infrastructure you skip.
2
Day 2 — Schema and routes
Use the @backend-dev subagent and /new-feature tutor-profile to scaffold the tutor and booking models in db/schema.ts and routes/. Domain-isolated by the boilerplate's modular pattern.
3
Day 3–4 — Mobile screens
Use @mobile-dev to build the tutor list, profile detail, and booking-request screens against the existing tab navigation and theme system. Phone OTP screens are already there.
4
Day 5 — Payments wedge
Wire Stripe in single-account mode against the boilerplate's billing adapter. Stripe Connect is a 1–2 day add at the Solo-at-1k tier — the adapter pattern accepts it cleanly, you just implement the Connect integration.
5
Day 6–7 — Polish, deploy, recruit 10 tutors
Sentry is scaffolded — turn it on. Deploy via the GitHub Actions workflow already in the repo. Then go offline and recruit tutors in one neighbourhood by hand. The product is ready before the supply is.
Frequently Asked Questions
Is this idea saturated?
No, but it's competitive. Wyzant, Superprof, and Preply own the global / national web layer; none of them own a single local mobile experience well. The wedge is a tight geography (one city, one subject vertical) with a mobile-first booking flow. 'Saturated' would be 2018 — in 2026 the incumbents have visible product-quality complaints and weak local geo-search.
Why a transaction fee instead of a subscription?
Because tutors will not pay a subscription until they're already getting bookings, and parents will not subscribe to a marketplace they haven't tried. Transaction fees align everyone: the platform earns when a lesson actually happens. Superprof's tutor-subscription model creates ghost listings and is a known weakness.
Do I need Stripe Connect on day one?
No. Run Stripe in single-account mode for your first 50–100 lessons and pay tutors via manual bank transfer. The boilerplate's billing abstraction accepts Connect as an adapter when you're ready — typically a 1–2 day build with the @backend-dev subagent.
What about background checks and safeguarding?
Required for any product touching minors. Persona or Onfido against the boilerplate's auth flow handles ID verification — that's a 1-day integration. Enhanced DBS / FBI-grade background checks are the 100k-user tier and have real per-check costs ($15–$50 per tutor). Don't skip; this is the trust moat.
How fast can I realistically get to first revenue?
Solo founder, full-time, with the boilerplate: a Lean MVP in 5–7 days, your first ten tutors recruited offline in week two, first paid lesson within three weeks. Most of weeks three through eight is supply-side hustle — the code is not the bottleneck.
What kills this category most often?
Going horizontal too early (every subject, every city), under-investing in tutor verification, and treating it as a software product instead of a two-sided marketplace where you must do unscalable supply recruitment yourself for the first 100 tutors.
What's NOT in the boilerplate that I'll need to build?
The marketplace-specific layer: Stripe Connect integration, in-app chat (build on Workers Durable Objects), geo-search beyond simple postcode matching, dispute and refund workflows, an admin moderation panel, and any background-check provider integration. The boilerplate gives you auth, billing abstraction, edge runtime, CI, and the modular architecture to add these cleanly. Claude Code does the wiring.
Tutor marketplaces are won locally, not globally.
A solo founder with the $199 boilerplate, Claude Code, and 10–14 days of focused work can ship a Lean MVP that handles auth, profiles, search, booking, payment, and reviews against working scaffolding. The harder problem is supply: recruiting your first 50 tutors in one neighbourhood by hand. The boilerplate compresses the software work so the founder time goes where it actually matters.