Skip to content
All case studies

02 · Expondo, global e-commerce · PurpleTalk

VAT Validation & Correction Engine

Rule-driven replacement for a manual monthly VAT correction process, with approval workflows and NetSuite-ready exports.

Context

Expondo's finance team corrected VAT data once a month by hand. The process was manual and error-prone. The engine replaces it with rule-driven validation of VAT IDs, approval workflows and exports that NetSuite can take directly.

My role

Lead Front-End Engineer.

Architecture

Architecture, left to right. Users: Finance team → Frontend: React · 100% TypeScript · MUI → Backend and output: REST APIs, NetSuite-ready export.

Frontend scope. The REST backend sits outside my scope. Dashed boxes are outside my scope.

Key decisions

  1. 01

    100% TypeScript

    Strict types and shared interfaces give the screens predictable data contracts and fewer runtime defects.

  2. 02

    Centralised state, memoisation, lazy loading

    Data-heavy workflows stay fast and free of prop drilling. Routes and components load on demand, which keeps the initial bundle small.

  3. 03

    Rules in the workflow, not in people's heads

    Validation, corrections and approvals follow explicit rules. [WHY: ask Saif — where the rules run (client or API) and the trade-off]

Results

  • Replaced the manual monthly correction process with a rule-driven system.
  • Sustained 80+ Lighthouse scores.

Stack

  • React
  • TypeScript
  • MUI
  • REST