How OpexGenie Delivered 15–20% Telecom Cost Savings for Enterprises by Automating Invoice Validation Across 30,000+ Bills

Client

OpexGenie is a UAE-based telecom cost-optimisation company that helps enterprises identify and recover overpayments on their telecom spend. In markets like the UAE, Saudi Arabia, and the wider GCC, large organisations routinely manage multi-operator telecom contracts spanning fixed line, mobile, and data services — often across thousands of SIM cards, circuit IDs, and billing codes. The gap between what a company negotiates in a contract and what it is actually charged can be significant, and without automated tooling, it goes undetected. OpexGenie engaged Ajackus to build the technical platform that would make systematic telecom cost validation possible at enterprise scale.

Challenge

Enterprise telecom billing is one of the most opaque categories of business spend. A single large enterprise in the UAE might receive monthly invoices from DU, Etisalat, and STC simultaneously — each formatted differently, each covering hundreds of circuits or SIM subscriptions, and each potentially hundreds of pages long. The expectation that a finance or procurement team could manually cross-reference these invoices against contract rate cards is unrealistic at any meaningful scale.

The result is a category of systematic overpayment that is well-known in telecom procurement circles but structurally difficult to address without purpose-built tooling. OpexGenie’s premise was that this problem was solvable — but solving it required engineering a platform capable of handling the volume, variety, and precision that enterprise billing demands.

Unvalidated Spend at Scale

Enterprise telecom invoices were processed and paid without line-item validation against contracted rates. With invoices spanning thousands of pages and multiple operators, manual verification was not operationally feasible, and discrepancies between billed and contracted amounts accumulated undetected month after month.

Multi-Operator Format Fragmentation

DU, Etisalat, and STC each produce invoices in different formats, structures, and data schemas. A platform that could only process one operator’s format had limited commercial value. The engineering challenge was building a multi-format ingestion and parsing layer that could normalise disparate invoice structures into a single, comparable data model.

No Automated Contract-Rate Comparison

Even where invoice data could be extracted, there was no system to hold the contracted rate card and compare it against actual charges at line-item level. This gap between invoice data and contract data was the core business problem — and required building both the data extraction and the comparison logic as integrated components.

Missing Analytics Visibility

Finance teams and procurement managers had no consolidated view of telecom spend trends, operator-level cost comparisons, or discrepancy patterns over time. Without visibility, cost recovery was reactive and incomplete — limited to whatever an analyst happened to notice in a spot check.

Slow Time-to-Insight for Cost Recovery

Telecom billing disputes typically have time-limited recovery windows. If discrepancies are identified too slowly — months after the billing period — commercial recovery options are reduced or eliminated. The platform needed to process invoices quickly enough that actionable discrepancies surfaced within the same billing cycle.

Goals

The project focused on building a cloud-native telecom invoice validation platform capable of processing enterprise-scale billing data from multiple GCC operators and surfacing quantifiable cost recovery for clients.

Journey

The Ajackus team took end-to-end ownership of the OpexGenie platform — from architecture design through to cloud deployment — operating as the core engineering function for the full product build. The Ajackus team deliberately prioritised the multi-operator parsing architecture first, recognising that the platform’s commercial value depended on supporting DU, Etisalat, and STC simultaneously from day one.

Cloud Infrastructure Architecture on AWS

The Ajackus team designed the platform on Amazon AWS with Docker containerisation, selecting this architecture specifically for its ability to scale horizontally as invoice volumes grow. Rather than building for OpexGenie’s current invoice volume, the architecture was scoped for the client’s projected growth, ensuring that processing a multiple of the current invoice count would not require infrastructure rearchitecting. Spring Boot was chosen for the backend services for its proven performance characteristics under concurrent data processing workloads — critical when processing hundreds of pages of invoice data per enterprise client per month.

Multi-Operator Invoice Ingestion and Dissection Engine

The core engineering challenge of the platform was building an invoice ingestion layer that could accept invoices from multiple operators — each with distinct billing formats, data structures, and field conventions — and normalise them into a unified data model. The Ajackus team built operator-specific parsing modules for DU, Etisalat, and STC, each handling the nuances of that operator’s invoice format before passing normalised data to the shared validation and analytics layer. PostgreSQL was selected as the data store for its reliability with structured tabular data at scale, and for the query performance characteristics needed when comparing thousands of line items against contract rate-card entries in real time.

Contract Validation Logic

The commercial engine of the platform is the contract validation layer, which compares each extracted invoice line item against the applicable contracted rate for that service, circuit, or SIM. The Ajackus team built this as a configurable rules engine, allowing OpexGenie to upload client-specific rate cards and define the comparison logic appropriate to each contract structure. When a discrepancy is identified — a charge above contracted rate, an unauthorised service, or a billing code not present in the contract — the system flags it with the information needed to initiate a dispute or recovery claim. This design, deliberately separating the contract configuration from the validation engine, allows new clients to be onboarded without changes to the underlying code.

Analytics and Reporting Dashboard

The Ajackus team built the client-facing analytics interface using Angular and ChartJS, giving enterprise finance teams and procurement managers a consolidated view of their telecom spend. The dashboard surfaces operator-level cost breakdowns, month-on-month trend analysis, discrepancy summaries by category and severity, and total recoverable-cost calculations. The choice of ChartJS within the Angular framework was driven by the requirement to render complex, multi-series financial data clearly within the same session as invoice review.

Testing and Quality Assurance with JUnit

Given the financial consequences of incorrect validation — a false positive discrepancy triggers a commercial dispute; a false negative means a missed recovery — the Ajackus team invested heavily in automated testing using JUnit. The validation logic was built with comprehensive unit and integration test coverage, ensuring that the comparison engine produces consistent, accurate results across the full range of invoice formats, contract structures, and edge cases encountered in real enterprise billing data.

Results

The OpexGenie platform has processed over 1 million bill pages across 30,000+ enterprise invoices, identifying 15–20% in recoverable or avoidable telecom costs for clients — savings that were entirely invisible before automated validation existed.

What went well:

Operational Improvements

• Over 30,000 enterprise telecom invoices ingested, parsed, and validated — a volume that would require a full-time analyst team to process manually, now handled automatically within each billing cycle
• Invoice processing across three major GCC operators (DU, Etisalat, STC) handled by a single platform, eliminating operator-specific manual workflows
• Enterprise clients receive actionable discrepancy reports within their billing cycle, maximising the commercial window for cost recovery
• Configurable rate-card upload allows OpexGenie to onboard new enterprise clients without platform code changes

Technical Achievements

• 1 million-plus bill pages processed through the multi-operator ingestion and normalisation engine without data loss or processing failures
• Spring Boot backend and Docker containerisation on AWS deliver horizontal scalability validated to support significant volume increases without rearchitecting
• JUnit-tested validation logic with comprehensive unit and integration test coverage ensuring mathematical accuracy across all supported invoice formats
• Operator-specific parsing modules for DU, Etisalat, and STC normalise disparate invoice formats into a unified PostgreSQL data model
• Angular and ChartJS analytics dashboard renders complex multi-operator, multi-period spend data in a consolidated interface

Business Impact

• Enterprise customers identify an average of 15–20% in recoverable or avoidable telecom costs — savings systematically invisible before automated validation
• OpexGenie can now serve enterprise clients with multi-operator telecom environments as a single integrated engagement
• The platform’s configurable architecture positions OpexGenie to expand coverage to additional GCC operators without rebuilding core infrastructure