Cortex vs Backstage: What's the best internal developer portal?

Cortex: Is It Worth the Cost?

Pros of Cortex

Cortex markets itself as a robust internal developer portal solution for teams looking to manage microservices and improve software ownership. It includes some useful features right out of the box, such as integrations for service cataloging and scorecards for tracking software standards. Cortex’s developer-centric features, like targeted alerts and service health metrics, aim to provide engineering teams with greater insight and control.

But while these capabilities sound appealing, the reality often falls short.

Cons of Cortex

Limited Flexibility

Cortex’s semi-rigid data model means that many entities in your catalog—like services and teams—are locked into specific types. This lack of flexibility can become a significant pain point as your architecture evolves. For example, Cortex doesn’t easily support non-service entities like machine learning models, Kubernetes clusters, or infrastructure components, which forces users to shoehorn those items into predefined categories. This constraint limits your ability to model your ecosystem in a way that reflects the true structure of your business.

Beyond this, Cortex also struggles to provide the flexibility needed for self-service actions. There’s limited support for asynchronous actions or ephemeral environments, and manual approvals require custom workflows. Cortex relies on webhooks rather than offering full integration support, which adds more complexity when connecting tools like GitLab or Jenkins .

High Manual Effort

One of Cortex’s biggest drawbacks is the manual effort required to maintain its service catalog. Unlike OpsLevel, which automates catalog updates based on actual data pulled from repositories, Cortex demands a lot of hands-on management. This significantly increases operational overhead, making it more difficult to ensure your catalog remains accurate and up to date.

Even routine tasks, such as updating ownership information, must be done manually in Cortex, which introduces the risk of stale data. For large teams with frequent changes in service ownership, this becomes a major bottleneck.

Weak Scorecard Customization

Another downside of Cortex is that their scorecards lack the depth and customization needed for complex engineering environments. Scorecard rules aren’t differentiated based on the maturity of the service, and you’re forced to apply the same global standards to all services, which may not reflect the true needs of each team or project.

Slow Implementation and High Cost

Cortex’s implementation time is another significant disadvantage. While OpsLevel customers typically get up and running in 30 to 45 days, teams using Cortex often report needing 6+ months to reach full deployment. This delay directly impacts developer productivity and the time-to-value of the platform.

Cortex is priced at $65/user/month which becomes prohibitive for many companies with large engineering organizations.

In addition to the licensing costs, the cost to implement and onboard is astronomical. In Cortex’s commissioned TEI report, it states a conservative estimate of 4 hours per engineer in implementation time. This means for a team of 240 engineers and 300 services, the cost of implementing Cortex would result in an additional $150,000 per year in people costs. That’s not even counting the additional 8 hours of training that each engineer would need to operate the portal.

The combination of high manual effort and slow implementation drives up the total cost of ownership (TCO), making it harder for teams to justify the investment.

Backstage: The Open-source Dream with Hidden Costs

Pros of Backstage

Backstage, the open-source solution to build your own internal developer portal from Spotify, offers something unique: a completely customizable framework that lets you build exactly what you need. It’s backed by a large and enthusiastic developer community, and for organizations with the resources to dedicate to building out a platform, Backstage can provide unparalleled flexibility.

With its plugin-based architecture, Backstage gives teams the power to create a tailored developer experience. You can build out your own integrations, customize workflows, and even create templates for common developer tasks.

Cons of Backstage

Steep Learning Curve and High Resource Requirement

However, the flexibility of Backstage comes at a steep price—one that many engineering teams don’t realize until they’re already deep into the implementation process. Setting up Backstage is not trivial. Unlike OpsLevel, which is ready to use within weeks, Backstage requires significant developer effort, often necessitating the full-time attention of multiple engineers.

To fully customize Backstage, you’ll need developers with specialized knowledge in React, TypeScript, and authentication protocols like SAML and OpenID. Maintaining the system post-deployment is equally taxing, as new plugins and integrations need constant updates. Many mid-sized organizations report that they need 3–15 full-time engineers just to maintain Backstage, making it a costly solution.

Low Adoption Rates

Another critical issue with Backstage is low developer adoption. While it’s true that Spotify’s internal teams have a high adoption rate for the platform (99%), external organizations report an average adoption rate of just 10%. This is primarily due to the platform’s complexity and the lack of a clear prioritization mechanism for developers. Without proper alerting and task management, developers are often overwhelmed by the sheer volume of data presented in Backstage, leading to poor adoption and reduced productivity.

Inflexible Data Models and Plugin Limitations

Despite its extensive plugin marketplace, Backstage still struggles with the same fixed data model issues as Cortex. Backstage requires manual data ingestion via YAML files, making it difficult to scale for large organizations with complex service architectures. Additionally, plugins don’t solve the fundamental issues with Backstage’s data model and manual maintenance burden.

The lack of out-of-the-box integrations also forces teams to either build custom plugins or rely on third-party agencies for support, further driving up costs and complicating maintenance.

Why OpsLevel Stands Above the Rest

Ease of Use & Quick Implementation

OpsLevel is designed for fast, easy implementation. Unlike Cortex and Backstage, which require months (or longer) to fully deploy, OpsLevel can be set up in just 30 to 45 days. The platform’s automated catalog creation eliminates the need for manual service updates, pulling real-time data from your repositories to keep your catalog accurate without the hassle.

For teams that need to move quickly and maintain a dynamic service architecture, OpsLevel’s ability to reduce operational overhead is a game-changer.

Customizability Without the Hassle

While Cortex and Backstage both offer customization, OpsLevel strikes the perfect balance between flexibility and ease of use. You can easily set organization-wide standards, but also create scoped standards for specific services or teams with Scorecards. Whether you’re tracking service maturity, security compliance, or operational readiness, OpsLevel’s customizable Scorecards allow you to apply different rules based on the context.

This flexibility ensures that engineering teams can enforce the right standards without being locked into rigid or global rules, as is often the case with Cortex.

Comprehensive Campaigns and Standards Tracking

OpsLevel also excels in its approach to tracking standards and managing change. With OpsLevel, you can create non-scorecard campaigns that provide engineers with the context, instructions, and reminders they need to implement changes. Notifications can be delivered through email, Slack, or in-app alerts, helping developers stay on top of what’s important.

OpsLevel’s reporting capabilities extend beyond simple pass/fail checks. You can generate detailed reports on the health and compliance of your services, ensuring you have a complete view of your operational landscape.

Cost-effective and Scalable

OpsLevel delivers all these capabilities at a fraction of the cost of Cortex and Backstage. By reducing the manual effort required to maintain your catalog and automating routine checks and updates, OpsLevel significantly lowers the total cost of ownership (TCO). This makes it an ideal choice for organizations that want powerful functionality without the excessive cost and resource burden.

When it comes to choosing the right IDP, the drawbacks of Cortex and Backstage are hard to ignore. Cortex’s rigid data model, high manual effort, and steep costs make it difficult for many teams to justify. Meanwhile, Backstage’s open-source flexibility is overshadowed by its complexity, high resource requirements, and low adoption rates.

OpsLevel offers the best of both worlds: an intuitive, out-of-the-box solution that scales with your team, automates critical tasks, and provides the flexibility to customize your experience—all without the overhead and resource drain of Cortex or Backstage. If you’re ready to experience a better way to manage your developer platform, request a demo today and see what OpsLevel can do for your team.

‍