Data Engineering · 4 min read

Makro consolidates 10+ data platforms onto two Tencent Cloud services

Makro moved more than 10 data products onto Tencent Cloud's DLC and WeData, reporting zero discrepancies, a 25% efficiency gain and table validation cut from two hours to five minutes.

By Tom Whitfield828 words

Key takeaways

  1. More than 10 data products from multiple vendors consolidated onto two Tencent Cloud services: Data Lake Compute (DLC) and WeData
  2. Migration completed with zero data discrepancies and uninterrupted business operations, per both parties
  3. Downstream operational efficiency improved by more than 25%, with over 20 business units modernized
  4. Nearly 10,000 production tasks now execute daily on Tencent Cloud's big data platform
  5. CodeBuddy-assisted validation cut single-table checks from one to two hours to approximately five minutes
Tencent Cloud Completes Enterprise Data Platform Consolidation for Makro - The Fast Mode
Tencent Cloud Completes Enterprise Data Platform Consolidation for Makro - The Fast Mode — wilhei / pixabay

Makro has collapsed a data estate spanning more than 10 products onto two Tencent Cloud services, reporting zero data discrepancies and uninterrupted operations throughout the cutover.

Both parties make that claim, and no independent audit accompanies it. Makro operates the wholesale business under CP AXTRA Public Company Limited, part of CP Group, one of Thailand's largest wholesale and retail brands. Its Makro PRO application holds recognition as Thailand's leading Thai-owned first-party (1P) e-commerce platform, meaning the retailer owns the seller-customer transaction and the resulting customer data.

The trigger was data volume, not a stated platform failure. Continued growth in digital operations and expansion into ASEAN markets drove sharp increases in transaction, product and membership data. Environments for data development, analytics, scheduling and governance stretched across more than 10 products from multiple vendors. Each environment carried its own tooling, and the sprawl did not scale.

What did the executives say?

Tanit Chearavanont, Group Chief Wholesale Business Officer at CP AXTRA, said: "Tencent Cloud has demonstrated strong technical expertise and a deep understanding of our business needs throughout this transformation journey." He added: "We look forward to further strengthening our collaboration to unlock greater value from our data assets and accelerate future AI-driven innovation."

Poshu Yeung, Senior Vice President of Tencent Cloud and Head of Tencent Cloud International, framed the constraint: "Large-scale data platform modernization presents significant challenges around workload migration, business continuity, and data consistency." He said: "We are proud to support Makro in completing this landmark transformation, which demonstrates how enterprises can consolidate complex data environments while maintaining operational stability."

What did the migration actually consolidate?

Two services now carry the load. Tencent Cloud Data Lake Compute (DLC) is a serverless data lake analytics service, meaning compute starts on demand rather than on a provisioned cluster. WeData is Tencent Cloud's platform for data integration, development, orchestration (sequencing dependent jobs) and governance (defining data quality, ownership and access rules).

By unifying previously fragmented workflows under one architecture, Makro says it streamlined data operations and created a scalable foundation. The same data assets now support both operational decision-making and AI model development, reducing cross-system complexity. Future workloads such as warehouse forecasting and inventory optimization remain groundwork, not delivered capability.

The difficulty is structural. Large-scale data environments support hundreds of business processes, applications and workflows. Maintaining operational continuity and data consistency through a migration of that size is the part both sides describe as the hard requirement.

How much did the numbers move?

Reported figures, all supplied by the parties:

  • Downstream operational efficiency improved by more than 25%.
  • Over 20 business units, including supply chain and finance, completed data and analytics modernization.
  • Nearly 10,000 production tasks now execute daily on Tencent Cloud's big data platform.
  • Data tables, fields and relationships are centrally managed, enabling traceability from data origin through transformation to downstream use.
  • Enterprise-wide data assets became visible and manageable through a single platform for the first time.

Governance is the quieter result. Central management of tables, fields and relationships gives Makro line-of-sight on where a number came from, how it was transformed and where it was consumed. That is a prerequisite for the AI work the company intends to run on the same assets.

How did AI shorten validation?

Tencent Cloud applied CodeBuddy, its own AI-assisted coding solution, to the migration. The tool automatically analyzed business SQL scripts, performed code conversion and validated data consistency at scale. On detected discrepancies it generated detailed reports for engineers to triage. Validating one data table previously took one to two hours; it now takes approximately five minutes.

Those gains are narrow and specific, which makes them more credible as tooling claims. They are not independently benchmarked, and the release does not say whether CodeBuddy stays available to other customers on comparable terms.

What covers the rest of the stack?

Beyond the data platforms, Tencent Cloud provides containers, databases and Tencent Cloud EdgeOne, a content delivery and edge acceleration service. The collaboration now spans from foundational infrastructure to mission-critical data platforms, according to the release.

What should a data team take from this?

  • Treat the 25% efficiency gain as customer-reported. The release gives no baseline, no measurement window and no third-party verification.
  • The framing as a benchmark for large-scale consolidation in Southeast Asia comes from the vendor's own release. One named customer in one country is a case study, not a benchmark set.
  • Missing from the disclosure: migration duration, contract value, project headcount, and the names of the more than 10 retired products. Procurement teams should request all four before drawing conclusions.
  • The reusable pattern is retirement of overlapping tools, one governance layer, and AI-assisted validation during cutover. The 25% figure suggests the fragmentation itself carried measurable operational cost.
  • Caveat the AI angle. Warehouse forecasting and inventory optimization remain future intent, with no model, accuracy metric or deployment scale disclosed.

Google News: Data Governance (Source)

Price tags

Share this article:

More from Tom Whitfield

Tom Whitfield

Show full bio

Staff writer covering regulation, policy and their impact on Analytics Digest is a professional daily at Analytics Digest.

16 articles

Read also

« Previous articleNext article »