How Kohler Modernized Global Content Delivery With AWS EKS And A Composable Headless Architecture

Introduction

Kohler, a global leader in lifestyle and home design, needed to modernize a sprawling network of regional websites that had evolved independently over time. Each site came with its own workflows, infrastructure, and performance patterns. During high-traffic campaigns, particularly for new product launches, the legacy platform often struggled to maintain uptime and responsiveness.

Beyond technical debt, the business challenge was clear: Kohler needed a unified, composable, and cloud-native platform capable of scaling globally while maintaining editorial agility and operational control.

When Axelerant’s engineering team engaged, the goal was to build once, scale everywhere, delivering an architecture that balanced flexibility, cost efficiency, and reliability across multiple geographies.

The Engineering Challenge

Kohler’s existing infrastructure was monolithic and regionally fragmented. Each regional team managed its own stack, leading to inconsistent publishing workflows, uneven performance, and unpredictable infrastructure costs. During campaigns, traffic surges could overwhelm servers, and manual intervention was often required to restore stability.

The modernization plan called for:

• Consolidation of regional sites into a single global headless architecture.
• Seamless scalability to handle variable workloads.
• Improved cost efficiency through dynamic resource allocation.
• Automated provisioning to eliminate operational overhead.
• Stronger observability and governance to improve reliability and compliance.

Building the Foundation: EKS + EC2 Synergy

At the heart of Kohler’s new digital platform is a containerized headless architecture running Drupal and Next.js on Amazon Elastic Kubernetes Service (EKS).

Instead of adopting a fully managed or serverless approach, Axelerant opted for EKS with EC2-backed nodes, a strategic choice to preserve fine-grained control over compute allocation, cost, and scaling behavior. This hybrid strategy allowed for predictable performance while still leveraging Kubernetes’ elasticity.

“For Kohler, the balance between scalability and operational control was critical. EKS with EC2-backed nodes gave us the best of both worlds, control where it mattered, and automation where it counted,” said Bassam Ismail, Director, Engineering at Axelerant.

Key architectural patterns included:

• Multi-AZ redundancy: Ensuring resilience and high availability across regions.
• Dynamic autoscaling: Leveraging Kubernetes’ Horizontal Pod Autoscaler (HPA) and Karpenter to provision and terminate nodes dynamically.
• Right-sizing & cost-awareness: Autoscaling policies driven by Prometheus telemetry, with spot and schedule-aware capacity for burst workloads.

This combination resulted in a 70% reduction in AWS costs after optimization, without sacrificing performance.

Observability-First Engineering

One of the early design decisions was to instrument first, optimize later.

The engineering team implemented Prometheus and Grafana alongside AWS CloudWatch and CloudTrail, establishing a unified observability stack.

This instrumentation revealed “hot paths” and CPU utilization trends during load tests, allowing the team to fine-tune autoscaling thresholds, node lifecycles, and caching strategies.

OpenTelemetry-style traces were extended across Drupal and Next.js components to correlate backend API latency with frontend rendering times.

The result was a data-driven approach to performance tuning, Kohler’s platform didn’t just scale reactively, it scaled intelligently.

Infrastructure As Code And GitOps Automation

To ensure consistent and repeatable rollouts, Axelerant implemented Terraform and Helm, combined with GitOps pipelines using ArgoCD.

This setup automated:

• Environment provisioning (Dev → Stage → Prod)
• Zero-downtime deployments with canary rollouts
• Version-controlled infrastructure changes

Every environment was declaratively defined, tested, and deployed through Git commits, enabling full traceability of infrastructure evolution.

“GitOps allowed Kohler’s delivery team to deploy confidently, knowing every change was observable, versioned, and reversible,” Bassam added.

This automation reduced provisioning time by more than 30%, eliminated human configuration errors, and accelerated release velocity by 40%.

Edge Delivery, Caching, And Security

Kohler’s new platform was architected for global reach with regional precision.

Edge delivery was achieved through Amazon CloudFront, with latency-based routing managed by Amazon Route 53. 

The combination of Application Load Balancer (ALB) and autoscaling EC2 nodes ensured high availability across availability zones.

Caching and search layers were optimized using:

• ElastiCache (Redis) for low-latency data retrieval
• Amazon OpenSearch for high-speed search indexing and content discovery
• Full-page caching and CloudFront edge caching for static delivery acceleration

To protect the platform, AWS WAF and bot mitigation were integrated to reduce malicious traffic, leading to a 90% drop in attack surface incidents post-launch.

Operational Governance And Lifecycle Management

Axelerant’s engineering delivery extended beyond build execution to operational governance.
Runbooks, standard operating procedures, and upgrade schedules were built into the delivery cadence to manage the platform’s lifecycle predictably.

EKS cluster upgrades were treated as planned release activities, tested and validated in staging before promotion to production.

This approach turned potential maintenance disruptions into routine, reliable iterations.

“We don’t see operations as separate from engineering. Governance is part of the build. It ensures long-term stability and predictable evolution,” Bassam emphazied.

Business And Engineering Impact

The measurable outcomes were substantial:

These outcomes reflect not just technical achievement, but delivery maturity—the ability to execute at enterprise scale with precision and repeatability.

Lessons For Engineering Leaders

Kohler’s replatform offers valuable insights for teams architecting global-scale digital ecosystems:

1. Instrument Early – Build observability before you build scale.
2. Treat Infrastructure As Product – Manage clusters and lifecycles like application releases.
3. Automate For Consistency – Use IaC + GitOps to replace manual governance.
4. Stay Edge-Aware – Global delivery is not one-size-fits-all; adapt caching and WAF policies regionally.
5. Optimize Continuously – Telemetry and cost metrics should guide ongoing scaling strategies.

These principles transformed Kohler’s architecture from reactive to proactive, driven by data, not assumptions.

Building For The Future

Kohler’s global replatform is more than a migration story, it’s a model for modern digital engineering.
By combining Kubernetes scalability, observability-driven automation, and composable architecture, Axelerant delivered a platform that not only performs at global scale but evolves predictably with business growth.

The engagement underscores what defines Axelerant’s approach: deep engineering excellence, operational discipline, and a commitment to measurable outcomes.

“Our mission is to make composable platforms sustainable. The real success isn’t just in going headless, it’s in engineering systems that stay resilient, observable, and cost-efficient long after launch,” concluded Bassam Ismail, Director, Engineering at Axelerant.

About Axelerant

Axelerant’s Digital Engineering practice helps global brands design and build composable digital platforms that scale. 

From DevOps automation and cloud-native modernization to enterprise CMS and headless integration, Axelerant engineers deliver measurable performance improvements backed by operational excellence.

Learn more about our Digital Engineering services.