The term Internal Developer Platform (IDP) gained traction during 2023. The idea: reduce developer cognitive load by giving them a unified portal to discover services, provision environments, run common workflows, and see their systems’ state. Three options dominate the conversation: Backstage (Spotify, open source), Port (commercial, low-code), and Cortex (commercial, scorecards). This article is an honest comparison with criteria to choose.
The Problem an IDP Solves
Developers in mid-to-large organisations spend time on:
- Discovering: “what service does X? where is the code? who maintains it?”
- Provisioning: “how do I request a DB? how do I spin up staging?”
- Repetitive workflows: “how do I rollback? how do I scale this?”
- Observability: “is my service OK? are there errors?”
An IDP centralises this: a portal that’s the door to the internal ecosystem. The alternative is fragmentation — stale wikis, Slack channels, tribal knowledge.
Backstage: The Open-Source Standard
Backstage was born at Spotify and open-sourced in 2020. Today it’s a CNCF project with a large community.
Strengths:
- Pure open source (Apache 2.0). No vendor lock-in.
- Extensible via plugins. Catalogs, docs, costs, cloud resources, Kubernetes — dozens.
- Large community. Meetups, conferences, contributing companies.
- YAML-in-Git catalog. Services are described next to the code.
- TechDocs: docs-as-code rendered in-portal.
Weaknesses:
- High adoption curve. Going from “npm create app” to production takes months.
- Continuous maintenance. It’s a React app you operate. Plugin updates break.
- Customisation requires code. Adding a new tab is a PR against portal code.
- Discovery takes work. Without a rich entity catalog, empty Backstage disappoints.
Backstage is for: companies with a dedicated platform team (3-10 people minimum) and multi-year commitment.
Port: Low-Code and Faster
Port (Israeli, founded 2022) takes a different approach. Commercial SaaS with visual configuration.
Strengths:
- Setup in hours, not months. Model is pre-designed; you adapt it.
- Dynamic catalog. Ingestion from AWS, GitHub, Kubernetes, Terraform state, JIRA — automatic.
- Visual blueprints and self-service actions. Developer self-service without code.
- Integrated scorecards: “production-ready”, “observability level”, etc.
- More polished UI than Backstage out of the box.
Weaknesses:
- SaaS. Data outside your network (though they offer private cloud).
- Commercial. Pricing per service or per developer.
- Less extensibility. If you need something very custom, Port may fall short.
- Vendor lock-in. Migrating to another IDP would be rebuilding the portal.
Port is for: companies wanting a productive IDP quickly, without investing in an internal platform.
Cortex: Scorecards Focus
Cortex (founded 2019, rebranded 2020) positions strongly in service catalog + engineering standards.
Strengths:
- Sophisticated scorecards: define levels (Bronze, Silver, Gold) and measure services.
- Automatic catalog from GitHub, Jira, Datadog, etc.
- Reliability workflows: on-call, post-mortems, runbooks.
- Powerful queries: “show services without owner with >100 errors/day”.
- Engineering excellence: focus on measurably raising technical level.
Weaknesses:
- High price. Cortex is among the most expensive.
- Less self-service than Port. Focus on visibility and measurement, not provisioning.
- Dense UI. Curve for non-technical users.
- Less open ecosystem than Backstage.
Cortex is for: companies wanting to raise technical level measurably, with emphasis on maturity observation.
Comparison Table
| Aspect | Backstage | Port | Cortex |
|---|---|---|---|
| Model | OSS | SaaS | SaaS |
| Time-to-value | 3-6 months | weeks | 1-2 months |
| Extensibility | React code | Low-code | Configurable |
| Self-service | With work | Native | Less |
| Scorecards | Plugin | Native | Core |
| Cost | Infra + team | SaaS license | SaaS license |
| Lock-in | Low | High | High |
Decision by Size
My recommendation by context:
Startup / team <30 engineers:
- A sophisticated IDP is overkill. A well-maintained README + tidy Slack channel + Kubernetes dashboards cover 80%.
- If you need something: Port free or basic tier.
Mid-market / 30-150 engineers:
- Port is the sweet spot: productive quickly, no dedicated platform team needed.
- Cortex if focus is raising technical level (e.g. after a major incident).
Enterprise / 150+ engineers:
- Backstage starts justifying itself through extensibility and customisation.
- Dedicated team (3-10 people) operates and evolves it.
- Alternatively Port or Cortex if you prefer commercial management.
The Common Error: Portal Fetish
Many teams install Backstage and are then surprised developers don’t use it. Typical cause: the portal has only a static catalog — no real self-service, no executable workflows, no actions.
A useful IDP offers:
- Catalog with rich information (not just name but dependencies, SLIs, on-call).
- Self-service actions (create repo, provision DB, rollback).
- Integrated automation (not “click this link, then copy-paste”).
- In-context observability (service metrics directly visible).
Without that, it’s just a disguised wiki.
Crawl, Walk, Run
Pragmatic approach to building an IDP:
- Phase 1: service catalog (who owns what, links to repos and dashboards).
- Phase 2: docs as code (TechDocs equivalent).
- Phase 3: self-service actions (create service, provision).
- Phase 4: scorecards and measurable engineering standards.
- Phase 5: complex workflows (blue-green, orchestrated rollbacks).
Trying everything at once is the recipe to abandon.
Lighter Alternatives
If the decision is not a full IDP, intermediate points:
- README + GitHub Topics: curate repos as minimal catalog.
- OpsLevel Service Catalog: similar to Cortex, more maturity-focused.
- Structured wiki in Notion/Confluence: low-tech but works for teams <50.
- Slack + bots for self-service (create repo via
/create-service).
Not everything must be a giant React portal.
Conclusion
IDPs are useful for teams that grew past the point where “ask on Slack” works. Choice between Backstage, Port, and Cortex depends on resources, culture, and ambition — Backstage for who can invest in platform and wants control, Port for speed and pragmatism, Cortex for measurable technical discipline. Avoiding the trap of a pretty portal without real content is the most important advice. Value is in content and workflows, not in UI.
Follow us on jacar.es for more on platform engineering, developer experience, and IDPs.
