🧩 Superstar Sale · Simple Sites

The simple-sites track: small, boring websites built from verified Starter Story case studies — a daily-habit quiz game, a hard-paywall niche job board, and a B2B data directory. Same method-factory pattern as every other business here: only case studies with receipts on camera make the shortlist, and every headline number is treated as a peak, not a median.

Overview The rail The pick Ground rules Repos & infra Verification 🧪 Pair test plan

Overview

First site live: Hire After 503 case studies graded on receiptsQuiz game + B2B directory = unbuilt candidates

Three Starter Story founder interviews, all with revenue dashboards shown on camera — the strongest evidence tier in that whole research set — point at the same shape: a simple website, one engine, built non-technically with Claude from a written spec, monetized by programmatic ads, a hard paywall, or a data subscription. The headline numbers ($40K/month, $25K/month, $32K MRR) are real dashboards but survivorship-framed; the page below carries the verdicts and the honest math.

Discord channel#sites (Super Star Sale guild)
Source 1 — job boardCareer Hound — "I make $25K/month from one simple website" (2026-09-24 video 2, Worth Trying 8/10)
Source 2 — data directoryAngelMatch — "I make $32K/month from 3 simple, boring websites" (2026-09-24 video 3, Worth Trying 8/10)
Source 3 — quiz gameGeo Sports — "I built this website in 8 hours with Claude. Now it makes $40K/month" (2026-09-27 video 12, Mixed 7/10)
RailVerify receipts → pick ONE model + niche → spec → Claude build block → Mike steps (domain, Stripe, accounts) → distribution rail per model → spin off niches on the same engine
Owner agentssuperstarsaledev + superstarsalemarketing
All three sources are Starter Story interviews: the channel's core business is selling entrepreneurship content, so every episode is also a funnel. Numbers are internally consistent and dashboard-backed, but self-reported — no third-party audit exists for any of them.

The rail (class procedure)

  1. Verify the case study before building. Shortlist only dashboard-on-camera sources (Stripe/ads dashboards pulled up live). Headline revenue = best month, never median. The founder's starting position (audience, prior launches, years of iteration) is explicitly non-copyable — copy the tactics, never the starting point.
  2. Pick ONE model + niche per site. Quiz game, paywall job board, and B2B data directory are three different engines. One site each; never blended.
  3. Spec first, then one Claude block. Written spec → single 8-hour build block → screenshot-driven troubleshooting when stuck (the Geo Sports playbook, safe for non-technical builders).
  4. Mike steps (agent cannot do these): domain purchase, Stripe/payee setup, app accounts, ad approvals.
  5. Distribution is designed, not hoped for: quiz game = design the share artifact first (53% organic share rate in the source); job board = content thrifting + timeless formats repeated forever; data directory = programmatic SEO (half of AngelMatch's revenue).
  6. Ship, then spin off. One proven engine clones into niche spin-offs (Geo Sports → Geo History → Geo Footy pattern) — never two models on one site.

The pick

SlotModelSourceWhy / status
Building nowFree-data + hard-paywall job boardCareer Hound (2026-09-24 v2)Built as Hire After 50 — free verified-employer board (12 employers carrying AARP Employer Pledge / Age-Friendly Certified / federal NPS markers, re-check dates on every row) + Weekly Kit $7/mo. Live; checkout wiring is the open Mike step.
Next candidateDaily-habit quiz gameGeo Sports (2026-09-27 v12)Web-first, zero-friction viral loop (score screenshot = the share artifact), one engine → niche spin-offs. Verdict Mixed 7/10: the mechanics are copyable, the 4M-follower launch pad is not.
Dark horseB2B data directoryAngelMatch (2026-09-24 v3)Picks-and-shovels for urgent, expensive B2B pain (founders MUST raise); programmatic SEO compounds; $62/month stack. Catch: source hand-waves the data-collection legality/effort — scoped before any build.
Career Hound went first because the research set graded it Worth Trying with the playbook concrete enough to run the same week — and because a job board's data layer (verified employers) is checkable fact, not scraped gray area.

Ground rules (from the verdicts, enforced on every build)

Repos & infra

ThingValue
Hire After 50 (built)microspaces/hireafter50 (public) — live at hireafter50.vercel.app; employer data in assets/js/jobs-data.js; Weekly Kit via launch-list modal (checkout not wired yet)
Build method (verified on this machine)Written spec → Claude in one block → screenshot-everything debugging; static deploy on Vercel
Research docsdocs/research/ in microspaces/superstarsale — playlist reports 2026-09-24 and 2026-09-27
Money railsStripe (Mike-only account/payee), programmatic ad network, per-model
Issuesmicrospaces/superstarsale, label pair-test until the track gets its own repo

Verification

Case-study grading (2026-09-24 + 2026-09-27 research reports): Career Hound — revenue dashboards pulled up on camera, numbers internally consistent (779 subs × $7/wk ≈ stated MRR with plan mix), 31% churn admitted honestly. AngelMatch — dashboards for all three sites shown live, portfolio math checks (29K + 2.8K + 0.26K ≈ 32K), the dying $260/mo site shown instead of hidden. Geo Sports — strongest receipts in the set (~$1,200/day ad dashboard + 4-week Stripe figures on screen) but MAU/share-rate self-reported and the headline run-rate extrapolates a sub-30-old dashboard.

Builds to date: Hire After 50 live (HTTP 200, title + 12-employer data file verified 2026-10-16). Quiz game and B2B directory: not started — candidates only, per the pick table.

🗂️ Roadmap

Live roadmap, managed as GitHub issues (superstarsale issues). Add or close items on GitHub — ☑ = done, close as “not planned” to remove.