DeepVisor

DeepVisor is an ad account intelligence system for ecommerce brands and lead-service businesses.

It connects to an ad account, preserves useful history, explains what is working and what is not, and turns recommendations into reviewable work across a dashboard, calendar, reports, and campaign views.

DeepVisor is not meant to be another analytics wall. The product direction is simple: help operators and owners understand performance quickly, decide what to do next, and keep that work organized in one place.

What DeepVisor Does

• Dashboard: a daily account read with spend, outcomes, strongest campaigns, weak spots, and the next thing to review
• Calendar: queued work that can be approved, modified, scheduled, or ignored
• Reports: owner-ready summaries of what changed, what worked, what failed, and what should happen next
• Campaigns: a fuller table view for scanning campaign health without losing context
• Integrations: connect platforms, choose the primary account, and control sync state

Who It Is For

• ecommerce businesses that want a clearer view of revenue-driving campaigns, creatives, and trends
• lead-service businesses such as dental, roofing, tiling, home services, and other consult or quote-driven operators
• founder-led or lean marketing teams that need one operating surface instead of jumping between ad platforms, spreadsheets, and reports

Current Product Scope

• Meta is the primary live integration today
• one primary ad account is selected per connected platform
• the first sync is recent-first, then full history is backfilled in the background
• recommendations are explainable and review-first
• the product is being shaped around ecommerce and lead-service operators rather than agencies

Stack

• Next.js + React + TypeScript
• Supabase for auth, database, and server-side data access
• Mantine and shared marketing/UI primitives for the product and public site
• internal sync, repository, and assessment layers under src/lib/server

Local Development

npm install
npm run dev

Useful scripts:

• npm run build: production build check
• npm run types:supabase: regenerate typed Supabase schema bindings

Repo Pointers

• Root marketing and app routes: src/app
• Meta integration and sync flow: src/lib/server/integrations/README.md
• Sync orchestration: src/lib/server/sync
• Assessments and account intelligence: src/lib/server/intelligence

Product Direction

DeepVisor should become the place where an operator can answer:

• What changed in the account?
• Which campaigns or ad sets are strongest right now?
• What needs attention?
• What should be approved or queued next?

The long-term goal is durable account memory, better recommendations over time, and a cleaner operating workflow for businesses that rely on paid acquisition.