Reconciling messy financial data across disconnected systems

*This article is adapted from * Tech in Asia

This article is a part of Startup Spotlight, a series that features young, up-and-coming startups.

While interviewing 74 candidates for a finance role, Tedo Ziraga discovered a shared frustration. Candidates described manual data reconciliation as a severe bottleneck that ruined month-end close cycles.

He launched Matchmade to provide the missing infrastructure for modern finance operations.

😟 Problem

Enterprise finance teams in Southeast Asia manually reconcile data across dozens of disconnected systems. Large retail chains process millions of transactions monthly across Shopee, Stripe, and Grab.

These formatting differences force staff to waste hours in spreadsheets.

Errors in manual reconciliation lead to revenue leakage and delayed closing periods for high-volume enterprises. Finance teams struggle to trust their own numbers.

💡 Solution

Matchmade built a rule-based engine to automate complex financial reconciliation workflows at scale. The software deploys in under two weeks and features several core capabilities:

• Ingests raw data from over 10,000 pre-built sources covering banks, payment gateways, and marketplaces.
• Transforms and consolidates data using customizable rules to completely eliminate AI hallucination risks.
• Matches data across disconnected systems and flags specific exceptions for non-technical users to resolve.
• Automates downstream processes using application programming interfaces and agentic layers with traceable storage.

📊 Market size

The global enterprise finance operations market represents a US$98 billion opportunity. Highly complex reconciliation workflows across Southeast Asia, Japan, and South Korea account for US$1.6 billion of this total. Industries like retail, financial services, and logistics drive this regional demand.

🤝 Team

• Tedo Ziraga. CEO and co-founder. He spent over a decade scaling tech companies like Gojek and Tinvio across the Asia-Pacific region.
• Kornelius Samuel. CTO and co-founder. He previously led multi-country product engineering teams at IBM and Gojek.
• Gilang Gibranthama. CCO and co-founder. He brings a decade of sales and marketing experience from Gojek and Binar.

🚀 Traction

• Reached US$1 million in annual recurring revenue in under two years with an 88% gross margin.
• Grew revenue 3.3x year over year with 99% cohort retention and 120% net revenue retention.
• Secured enterprise clients including Pizza Hut and Wingstop across Indonesia, Singapore, and Malaysia.

🏆 Competition

Matchmade competes across four fragmented software categories. Traditional ERP systems lack native data operations, while reconciliation tools like BlackLine require clean data and expensive consultants. Data pipelines like Fivetran demand dedicated technical teams.

Matchmade differentiates itself by combining raw data ingestion with customizable rule-based operations and long-term storage. The platform provides a user-friendly interface designed specifically for non-technical finance professionals.

💰 Financials

The company operates on a tiered enterprise software-as-a-service model. Customers pay annual subscriptions priced by transaction volume, data sources, and use case complexity.

Matchmade raised US$1.7 million in an October 2023 seed round led by Vertex Ventures Southeast Asia and India. Antler also participated in the financing.

The startup is currently raising a US$3.6 million series A round. The team targets an early Q4 2026 close to fund expansion into Japan and Korea.

🚩 Risks

• Matchmade’s margin story depends on converting messy reconciliation work into reusable rules across banks, gateways, marketplaces, and enterprise systems. That is a hard scaling problem because fragmented, non-standard finance workflows often resist clean automation. 
• Matchmade’s exception-led model only works if finance teams trust the system enough to move month-end close out of spreadsheets and let APIs trigger downstream actions. Finance automation research highlights why trust in data quality and close-process control remain central concerns. If exception volume stays high, approvals can become the new bottleneck.