Roadmap · Prototype → Market
From clickable prototype to live in your first city.
A local card & sealed-product marketplace lives or dies on liquidity. These six phases
prove that real collectors will buy, sell and trade near each other — by hand — before the
backend gets built. Validate first, engineer second.
Overall progress
15%4 of 27 steps
Est. cash to public launch · Phases 0–4
~$19,900excl. founder time · Phase 5 scales separately
🎯
This month: finish the prototype's onboarding + trade-offer flow, lock a real
name (drop "Poké" — trademark risk), pick one dense launch metro, and start recruiting 15–25 local
collectors using the app as your demo. No new backend yet.
0
Foundations
Now · before more build
- ✓Clickable prototype built — cards + sealed, themes, AI grade, matching$0
- ✓Category scope set — singles and sealed product$0
- ✓Pricing-integrity rule drafted — median + 40% outlier cut$0
- ·Pick launch metro (density beats breadth)$0
- ·Name + trademark — drop "Poké" (working title)$1,500
- ·Incorporate + domain + logo$1,300
Est. phase cost · ~$2,800
Next
Prototype + scope + pricing rule are done. Remaining: name, city, incorporation.
1
Collector validation
Week 2–6 · still no new code
- ·Recruit 15–25 local collectors by hand (LGS, events, Discord)$500
- ·Waitlist / "collectors near you" page (domain + host)$100
- ·Broker 20–40 trades/sales manually; seed liquidity$800
- ·Instrument match rate, repeat, meetup completion$0
Est. phase cost · ~$1,400
Next
Use the interactive app as your pitch/demo to line up the first collectors.
🚦 Gate: if you can't seed local trades by hand, an app won't fix it.
2
Real product
Week 6–14 · now you build
- ✓UI / interactive prototype built$0
- ·Backend + DB + deploy (Supabase / Postgres + PostGIS, Expo build)$665
- ·Auth + profiles + privacy-safe approximate location$0
- ·Card + sealed catalog & pricing data (TCGplayer / Pokémon TCG API)$300
- ·Listings + photo uploads + map discovery$200
- ·Clean-pricing v0 live (median + 40% rule)$0
- ·AI grading — third-party / human-in-loop pilot$1,000
- ·Trust & safety — ID verify, ToS, escrow (Stripe Connect)$4,000
Est. phase cost · ~$6,200 + payment fees
Next
PostGIS geo + Stripe Connect + the catalog/pricing pipeline are the real work — the screens are done.
- ·Invite-only launch in one metro$0
- ·Run every trade/sale end-to-end, fix top friction$1,000
- ·Hit liquidity — ≥3 collectors with matches per search (incentives)$2,000
Est. phase cost · ~$3,000
🚦 Gate: median "want → found nearby" under a few days before going public.
4
Public launch — one city
Month 4–6
- ·Open publicly — LGS partnerships, local events, card Discords/Reddit$5,000
- ·Turn on Pro + promoted listings + grading credits$500
- ·Support, ratings quality, payouts tooling$1,000
Est. phase cost · ~$6,500
🚦 Gate: Pro conversion + take-rate must cover CAC before scaling.
5
Scale + the moats
Month 6+
- ·Replicate city → city, one dense metro at a time$10,000+
- ·Real pricing-integrity engine (eBay Insights) + AI grading model$8,000
- ·Raise — pilot metrics + the data moat are the pitch (legal)$5,000
Est. phase cost · ~$23,000 scales with growth
Costs are rough cash-out estimates for a solo founder building with AI assistance — they exclude your time and salaries.
"~3% / txn" is Stripe's per-transaction cut (passed through, not a fixed cost). Trust & safety bundles ID verification,
platform-liability insurance, and a lawyer-drafted ToS / marketplace terms — get a real attorney quote before relying on this number.
Card & pricing data (TCGplayer, eBay Marketplace Insights) requires API approval; AI grading starts as a third-party / human-in-loop
service, not an in-house computer-vision model. "Poké" is a placeholder — see the name + trademark step.