Right Co
A market intelligence pipeline that turns scattered customer signals into structured business decisions.
The Problem
Right Co is a company building products in a competitive market. Like most early-stage companies, they had data — customer interactions, market signals, competitor activity — but it lived in disconnected places: CRMs, spreadsheets, ad platforms, and support tools.
The result was that strategic decisions happened on gut feel and periodic manual research. By the time a report was assembled, the data was already stale. The team needed continuous intelligence, not monthly summaries.
My Approach
The solution was a data pipeline designed around one principle: intelligence should arrive automatically, not be requested manually.
Key architectural decisions:
n8n for pipeline orchestration. Each data source gets its own workflow branch. n8n handles scheduling, retries, and error reporting without requiring a custom daemon or cron infrastructure.
Supabase as the central data store. All incoming data — normalized to a common schema — lands in Supabase. This gives a single queryable source of truth with row-level security and real-time capabilities built in.
Schema-first normalization. Every data source is different. The pipeline transforms raw inputs into a shared schema before writing to the database. This means downstream consumers (dashboards, reports) never need to know where data came from.
Trigger-based alerting. Rather than requiring someone to check dashboards, Supabase database functions trigger alerts when defined thresholds are crossed — unusual activity patterns, significant competitor changes, or volume anomalies.
Architecture
Data Sources
├── Customer behavior events (product analytics)
├── Market signal feeds (external APIs)
└── Manual data inputs (forms, uploads)
n8n Pipeline
├── Ingestion workflows (per source, scheduled)
├── Normalization transforms
├── Deduplication logic
└── Write to Supabase
Supabase
├── Normalized data tables
├── Database functions (triggers, alerts)
└── REST + real-time API for consumers
Outputs
├── Dashboard views (live queries)
└── Automated alerts (email / webhook)Impact
The pipeline eliminated the manual data assembly process entirely. Market intelligence went from a monthly effort to a continuous, automatic system. The team gained the ability to detect behavioral patterns and market shifts in near-real-time, enabling faster and more confident product decisions.
The design is also maintainable: adding a new data source means adding one new n8n workflow branch and a normalization step — the rest of the pipeline picks it up automatically.