Wednesday, September 9, 2026 · Week 37 DE · EN · FR · ES Dark
Expert Opinions

Platform Engineering: Why Developer Experience Is the New Competitive Advantage

TL;DR Platform Engineering builds internal developer platforms that abstract away complexity. Internal Developer Portals (Backstage, Port) offer self-service access to infrastructure and services. Golden Paths define the recommended way to complete standard …

By Alec Chizhik August 14, 2025 4 min read
Platform Engineering: Why Developer Experience Is the New Competitive Advantage

TL;DR

  • Platform Engineering builds internal developer platforms that abstract away complexity.
  • Internal Developer Portals (Backstage, Port) offer self-service access to infrastructure and services.
  • Golden Paths define the recommended way to complete standard tasks – without restricting freedom.
  • Teams with strong Developer Experience deploy 4x more frequently and experience 3x fewer incidents.
  • This trend resolves the “You Build It, You Run It” overload created by the DevOps movement.

DevOps created a problem it can’t solve: developers are expected to write code, manage infrastructure, build pipelines, configure security scans, and participate in on-call rotations. Cognitive load is exploding. Platform Engineering is the answer – a dedicated discipline that builds internal platforms enabling developers to be productive without needing deep expertise in cloud infrastructure.

The DevOps Paradox: More Responsibility, Less Productivity

“You Build It, You Run It” was the right idea at the wrong time. When Amazon introduced the principle, their teams had dedicated tooling teams providing deployment pipelines and monitoring infrastructure. In most companies, “You Run It” was implemented without that support – the result: developers spend 30-40% of their time on infrastructure instead of features.

Platform Engineering corrects this imbalance. A dedicated platform team builds and operates the internal developer platform, enabling developers to be productive without having to write Kubernetes YAML, Terraform modules, or CI/CD pipelines from scratch.

40%
of their time on infrastructure instead of features. Platform
50%
faster. The productivity gain justifies the investment.

The Internal Developer Platform (IDP)

An IDP consists of four layers: Infrastructure Orchestration (Terraform, Crossplane) provisions cloud resources. Container Orchestration (Kubernetes, ECS) runs workloads. CI/CD (GitHub Actions, ArgoCD) automates deployment. Developer Portal (Backstage, Port) provides the self-service interface.

The portal is the visible layer: developers create new services via templates, view the status of their deployments, access documentation, and request infrastructure resources with a single click. What happens behind the scenes – Terraform apply, Kubernetes deployment, DNS configuration – is abstracted away.

Golden Paths: Opinionated, Not Restrictive

Golden Paths are the recommended way to complete standard tasks: “This is how you build a new microservice,” “This is how you deploy to production,” “This is how you set up monitoring.” They are opinionated – there’s a clear recommendation – but not restrictive. Teams can deviate from the Golden Path if they have a valid reason.

In practice, this means: a service template in Backstage generates a repository with a CI/CD pipeline, Dockerfile, Kubernetes manifests, monitoring configuration, and documentation. The developer only needs to modify the business logic – everything else is preconfigured. From code commit to production deployment in 15 minutes instead of 3 days.

Backstage as the De-Facto Standard

Spotify open-sourced Backstage in 2020, and it has since become the de-facto standard for Internal Developer Portals. Its plugin ecosystem includes over 150 plugins: Kubernetes monitoring, CI/CD integration, API documentation, cost management, security scans.

Backstage’s strength lies in its Software Catalog: a central registry of all services, APIs, libraries, and teams, including ownership information. Who maintains this service? Where is the documentation? Which APIs does it expose? The catalog answers these questions – eliminating the need for Slack threads and Confluence searches.

ROI of Platform Engineering

The data is consistent: teams with strong Developer Experience deploy 4x more frequently, have 3x fewer change failure rates, and onboard new developers 50% faster. This productivity boost justifies investing in a platform team (typically 3-5 engineers for 50-200 developers).

The hidden ROI lies in retention: developers who work with modern tools and platforms tend to stay longer. In times of talent shortage, this is a critical factor. Companies that force developers to work with outdated tools and manual processes will lose them to organizations investing in Developer Experience.

Frequently Asked Questions

What’s the difference between Platform Engineering and DevOps?

DevOps is a culture and methodology – collaboration between development and operations. Platform Engineering is a discipline – a team that builds an internal platform. Platform Engineering implements DevOps principles, but in a way that doesn’t overwhelm developers. It’s the productive evolution of DevOps.

How large should a platform team be?

Rule of thumb: 1 platform engineer per 15-25 developers. A startup with 20 developers doesn’t need a dedicated platform team – shared responsibilities and good templates suffice. For 50+ developers, a dedicated team (3-5 people) becomes sensible; for 200+ developers, it’s essential.

Do you need Backstage, or are CI/CD tools enough?

CI/CD is one component of the platform, not the platform itself. Backstage or alternatives (Port, Cortex) provide the self-service layer that integrates CI/CD, infrastructure, monitoring, and documentation. For small teams, a well-configured GitHub with Actions and templates may be sufficient.

How do you measure Developer Experience?

The SPACE metrics (Satisfaction, Performance, Activity, Communication, Efficiency) offer a framework. More pragmatically: developer surveys (quarterly), time-to-first-commit (onboarding), deployment frequency, lead time for changes, and the number of self-service requests vs. tickets to the platform team.

Can Platform Engineering be introduced incrementally?

Yes – and that’s the recommended approach. Start with the biggest pain point – often CI/CD or Kubernetes deployment. Build a Golden Path for the most common use case, gather feedback, and iterate. Backstage can start minimal (Software Catalog) and gradually expand with plugins.

Header Image Source: Pexels / Markus Spiske

Also available in

FrançaisEspañolDeutsch
MBF Media Newsletter

The monthly briefing for decision-makers

Once a month, the MBF Media Newsletter gathers what matters from cloudmagazin, MyBusinessFuture, Digital Chiefs and SecurityToday, curated by the editorial team.

25,000 IT and business decision-makers read this newsletter. Read along.

Subscribe for free
MBF Media Newsletter, aktuelle Ausgabe auf dem iPhone
A magazine by Evernine Media GmbH