MaaS — Mindshare as a Service

A SaaS platform for mindshare analytics: audit your niche, discover KOLs that convert, and run multi-channel campaigns (X, LinkedIn, Shorts).

Product and visual direction

• Landing: One strong CTA (“See what I’m missing”), no visible “login” button; user is invited to discover, not to sign in.
• Onboarding: 7 questions in 4 steps; answers feed dashboard data; dashboard UI stays the same.
• Dashboard: Data (KOLs, intelligence, ROI) shown in defined blocks; intuitive display that does not look like classic analytics tools (cards, tables, charts). Distinct, readable.
• Auth: Mock by default; Supabase for storage (survey results, API extractions) and optional real auth later. No GitHub OAuth in product direction.
• Data: Stored in Supabase (or similar), not on Vercel. See docs/DIRECTION_ARTISTIQUE_ET_TECHNIQUE.md for the full direction (artistic and technical).

Stack

• Frontend: React 18, Vite, Tailwind CSS, React Router
• Backend: Node.js, Express
• Auth: Supabase (optional; mock auth when not configured)
• Payments: Stripe Checkout (optional)

Run locally

From the project root:

npm run start

Then open your browser at http://localhost:5173 (the command does not open it automatically). The landing page leads to onboarding (7 questions in 4 steps) then the dashboard.

MaaS — Mindshare as a Service

MaaS is a prototype SaaS that identifies high-impact content and KOLs (Key Opinion Leaders) to help early-stage SaaS teams choose channels, angles and creators that convert. The repo contains a React frontend (landing, onboarding, dashboard) and a Node.js backend with extraction/orchestration endpoints.

Deployment domain

The intended production domain for the landing and app is: maa-s.vercel.app

Quick overview

• Landing: single-CTA focused on conversion. Onboarding: short survey that feeds personalized dashboard suggestions. Dashboard: prioritized actionable insights (what content worked, which KOLs convert, ROI signals). Missing data is shown as “Coming soon”.

Tech stack (strict minimal)

• Frontend: React + Vite + Tailwind
• Backend: Node.js + Express
• Storage/analytics: Supabase (optional)
• Hosting: Vercel (frontend) + optional server on Node (backend)

What to know before running

• Copy .env.example to .env and populate required keys when you need real integrations. The repository intentionally ignores .env — do NOT commit secrets. If secrets were ever committed, rotate them immediately.

Minimal environment variables (examples in .env.example):

• PORT — backend port
• VITE_SUPABASE_URL, VITE_SUPABASE_ANON_KEY — Supabase (optional)
• STRIPE_SECRET_KEY, STRIPE_WEBHOOK_SECRET — Stripe (optional)

Run locally

Install and run both frontend and backend from the repository root:

npm install
npm run start

Open http://localhost:5173 to view the frontend. The app falls back to mocked/sample data when integrations are not configured.

Notes on APIs & next steps

• The backend contains connectors for X (Twitter), YouTube, LinkedIn, Meta/TikTok and others under backend/services/. Each connector expects API keys in environment variables. See the docs folder for integration links.
• Missing or rate-limited data is shown as “Coming soon” on the dashboard; the frontend already contains placeholders for these states.
• Required later: hourly availability checker to remove content that disappeared from source platforms, KOL country-based grouping and influence-based ranking (implemented in the frontend as filters/sorting but requires backend data).

Vision (short)

Automate campaign discovery and content generation via AI agents that can propose, test and optimize creator-driven campaigns with minimal human oversight. Keep the product focused on actionable copy/angles and creator matching that actually drives conversions.

If you want, I can now:

• Add an infra diagram and a short deploy checklist for Vercel + Supabase.
• Implement the backend endpoint skeletons required by the dashboard filters and hourly checker.

— Small, actionable README to help contributors get started.