Back to Lab
Business Operations2025 - 2026

Bespoke Stock Manager

A high-integrity logistics tool designed to eliminate human error. We architected a database-centric system with an immutable ledger that automates the complex relationship between physical components, digital sales bundles, and end-to-end financial tracking.

Stack
Next.js 16 / TypeScript
PostgreSQL / Supabase
UI/UX
Radix UI / Tailwind CSS
Paradigm
Database-Centric / RPC
Client
Private Enterprise

The Entropy Problem

Before intervention, the client operated a high-volume logistics business relying solely on manual synchronization between warehouse spreadsheets and their e-commerce frontend.

This fractured state resulted in:

  • Phantom Inventory: Sales made on items physically out of stock due to sync delays.
  • Replenishment Guesswork: Ordering static amounts rather than calculating dynamic deficits.
  • Data Silos: Financial liabilities regarding suppliers were calculated manually, leading to drift.

System Fragility

Previous reliance on "avoiding human error" was unsustainable at scale. A single missed spreadsheet update caused cascading failures in fulfillment.

Architecture

Component Normalization

We decoupled the concept of "what we sell" from "what we have". By treating inventory items as atomic units and sales products as molecular recipes, we achieved true data integrity.

Logic Engine

LIVE
Component A
150stock
Component B
Limiting
80stock
fn(min)
Digital Bundle
Composite
80units

Available for immediate fulfillment

The Source

Physical entities (components) are tracked via an immutable ledger. Every damage report, arrival, or manual correction is a signed transaction.

The Recipe

Logic layers that define relationships. A single digital product may require multiple physical components. The system calculates availability based on the scarcest component.

The Sync

When component stock changes, a background worker recalculates all dependent recipes and pushes the new "Buildable Quantity" to the sales channel API.

Engineering Highlights

Immutable Audit Ledger

Stock levels are never simply "overwritten". Every change is recorded via a transactional RPC that creates an entry in an immutable log table, complete with a reason code, user ID, and a before/after snapshot. This creates a forensic audit trail for every item.

Transactional RPCsData Integrity

Algorithmic Replenishment

A core database function analyzes target stock levels against real-time component deficits to automatically generate draft procurement requests. The system manages the entire multi-stage procurement lifecycle, updating stock and financial ledgers only upon final receipt validation.

Automated Financial Ledgers

The system links physical operations to financial data. Upon receiving goods, it automatically calculates and accrues the corresponding financial liabilities into a dedicated ledger. This provides a real-time, auditable view of outstanding debts, which can be settled through a secure, transactional reconciliation process.

PL/pgSQL FunctionsFinancial Reporting

Database-Level Security (RLS)

Leveraging PostgreSQL's native Row Level Security, access control is enforced at the data layer itself, not just in the API. A custom permissions system allows granular control, ensuring warehouse staff can update stock counts but are physically prevented by the database from viewing sensitive financial data like cost prices or liability ledgers.

PostgreSQL RLSCustom Permissions