Every major conference in 2025 and into 2026 has pushed the same narrative: Build an Internal Developer Platform, hire a platform team, pave the golden paths, and watch your developers become 10x more productive. The pitch is compelling. The execution is failing at most organizations, and the reason has nothing to do with technology. It is a staffing problem that nobody wants to address honestly.
Building a good internal platform requires a rare combination of skills. You need people who understand infrastructure deeply, who can think in abstractions and APIs, who have genuine empathy for developer experience, and who can resist the urge to build what is technically interesting in favour of what is actually useful. That profile does not exist in abundance. It barely exists at all. Most organizations are staffing platform teams with either senior infrastructure engineers who have never built a product or product engineers who have never operated infrastructure at scale. Both fail, just in different ways.
The infrastructure engineers build platforms that are technically excellent and operationally invisible to developers because the abstractions do not match how developers think. The product engineers build beautiful developer portals backed by fragile automation that breaks the first time someone needs to do something the golden path did not anticipate. And in both cases, the platform team is understaffed because leadership allocated headcount based on building the platform, not maintaining, evolving, and supporting it indefinitely.
Here is the uncomfortable truth: A platform team is not a project team. It is a product team, and it needs to be funded and staffed like one. That means dedicated product management, user research with your own developers, a support rotation, a roadmap driven by internal adoption metrics, and a long-term commitment that survives quarterly re-orgs. Most companies allocate three to five engineers for six months, declare the platform “done,” and wonder why adoption stalls at 20%.
If you are serious about platform engineering, start with the staffing model. Define the skill profile honestly. Accept that you will need to grow these people internally because you will not find them on the market.
Budget for a team that is at least 50% larger than you think you need. And commit to a multi-year roadmap. Anything less and you are building a graveyard of good intentions with a Backstage frontend.