How Experian, a leading credit bureau serving 8 million financial consumers, modernized a critical data platform — migrating 162 collections and 150+ business logic functions to AWS managed services with zero downtime.
Experian, one of Colombia's leading credit bureaus serving approximately 8 million financial consumers, needed to modernize a critical data platform running on self-managed MongoDB 3.6 on Amazon EC2. Triggered by MongoDB end-of-life and internal audit requirements mandating the separation of business logic from the database layer, Experian partnered with MacondoTek to execute a full cloud-native modernization. MacondoTek secured AWS Partner Network funding for a Proof of Concept, then executed a full migration of 6 databases, 162 collections, and 314+ GB of data to Amazon DocumentDB — while refactoring 150+ stored database functions into independent AWS Lambda microservices — all with zero downtime and zero incidents.
Experian is part of a global conglomerate present in 37 countries with over 16,000 employees. Its local platform allows approximately 8 million financial consumers to query, correct, and improve their credit history, access personalized financial offers, and manage payment agreements. The platform supports multiple credit bureau products and processes critical financial data daily.
MacondoTek is a cloud consulting and professional services firm headquartered in Atlanta, USA, with nearshore delivery teams across Latin America, specializing in cloud architecture, data platforms, and application modernization on AWS. As an AWS Advanced Tier Partner with Service Delivery Program designations for Amazon RDS and Amazon DynamoDB, MacondoTek supports organizations across the full cloud lifecycle, from migration through cloud-native development to 24/7 managed operations. Its proprietary MTK CloudOpps platform powers FinOps and cost governance engagements with enterprise-wide visibility across large multi-account AWS environments, including deep experience serving financial services organizations that must balance reliability, security, and compliance with cost optimization.
Experian’s legacy architecture relied on MongoDB 3.6 hosted on Amazon EC2, where both data and business logic were tightly coupled. Core application logic was implemented as JavaScript functions stored directly inside the database, invoked by Java services running in Kubernetes (Amazon EKS). As the platform evolved, this architecture created compounding limitations.
MongoDB 3.6 was approaching end-of-life, requiring a platform upgrade to maintain security and support. Internal audit requirements mandated the separation of business logic from the database layer to improve governance and maintainability. With business logic embedded as stored JavaScript functions, independent testing and deployment were impossible. The database team was burdened by the operational overhead of self-managing MongoDB on EC2, with no elastic scaling capability to handle demand spikes — and no automated credential management or centralized observability.
Experian needed a modernization strategy that would address compliance requirements, reduce operational risk, and align the platform with cloud-native AWS architecture — all without disrupting service to millions of financial consumers.
MacondoTek structured the modernization as a two-phase engagement — beginning with a low-risk, AWS-funded Proof of Concept, then executing a full platform migration using a proven zero-downtime methodology.
The architecture introduced a multi-account design separating the application and data layers, with communication routed via AWS Transit Gateway and private VPC Endpoints with mandatory TLS. AWS Secrets Manager automated credential management with scheduled rotation, combined with IAM/IRSA for containerized workloads in Amazon EKS. Monitoring was unified in Amazon CloudWatch across all services. A data anonymization process — 30 obfuscation functions across 27 collections — was also modernized as part of the engagement, rebuilt on AWS Step Functions and Amazon EventBridge for automated, auditable execution.
MacondoTek structured the delivery as two distinct phases — beginning with an APN-funded PoC to validate the approach at low risk, then executing the full migration with a proven parallel validation methodology that ensured zero downtime for 8 million financial consumers.
Leveraging its APN membership, MacondoTek collaborated with AWS to secure funding for a PoC designed to validate the migration strategy before any production commitment. The PoC migrated a representative set of MongoDB databases and collections, refactored 10 stored database functions into AWS Lambda, and validated compatibility between MongoDB 3.6 and Amazon DocumentDB 5.0 through performance benchmarking and response-time testing. The successful PoC confirmed architectural viability and enabled the full migration.
MacondoTek designed and implemented a comprehensive migration to Amazon DocumentDB and a microservices architecture on AWS Lambda — covering 6 databases, 162 collections, 314+ GB of data, and 426 indexes. Between 151 and 214 business logic functions were migrated from stored MongoDB procedures to AWS Lambda (Node.js), exposed through Amazon API Gateway with private endpoints. A factory-model approach delivered 50 Lambda functions per sprint. A parallel validation (marcha blanca) methodology ran every transaction against both legacy and new systems for two weeks prior to cut-over, with discrepancies logged to Amazon DynamoDB and daily control metrics reviewed. The cut-over was executed progressively (0% → 80% → 100%) within an 8-hour window with a validated rollback path ready throughout.
The migration completed with zero downtime and zero incidents for end users, validating MacondoTek’s progressive cut-over methodology and automated reconciliation framework. The platform now handles demand spikes elastically through Amazon DocumentDB’s auto-scaling, with meaningful improvements in response times for business-facing teams.
As an APN member, MacondoTek secured AWS funding support for the Proof of Concept — reducing project risk and accelerating the path to production before any full commitment was required.
Deep experience in financial services platforms where reliability, compliance, and governance are non-negotiable requirements for every architectural decision.
MacondoTek's parallel validation (marcha blanca) approach and progressive cut-over design ensured that service to 8 million financial consumers was never at risk during the transition.
MacondoTek engineers worked side-by-side with Experian's internal teams throughout, ensuring knowledge transfer and full alignment with internal governance and security standards.