Dev & Engineering · Engineering, IT & AI
Should you build or buy GraphQL Federation & Gateway Platform?
GraphQL Federation & Gateway Platform software lets organizations combine multiple GraphQL services (subgraphs) into a single unified API (supergraph) that clients query through a central gateway, enabling independent teams to own their own service schemas while presenting a coherent API surface to consumers. It handles schema composition, query planning, and routing.
The build-vs-buy decision for GraphQL Federation & Gateway Platform turns on how much the self-hosted OSS router and schema registry cover your daily workflow versus where hosted tooling adds composition checks and team management that require meaningful integration work to replicate, and how the post-Apollo cloud routing landscape has shifted the OSS maturity curve; the specifics of your schema change velocity and team count decide it.
Build it, buy it, or bridge?
When building makes sense
The OSS landscape for GraphQL federation shifted substantially after Apollo discontinued managed cloud routing in March 2026. Apollo Router is Apache-2.0 licensed and production-ready. GraphQL Hive is MIT-licensed, self-hostable, and provides a schema registry with composition checks. WunderGraph Cosmo offers another Apache-2.0 router option. Multiple teams now run complete federation stacks without commercial tooling. The supergraph schema itself — subgraph boundaries, federation directives, and schema evolution rules — is architecturally proprietary to your team regardless of which router runs it. Owning that schema governance layer is strategically meaningful for GraphQL-first orgs, and the OSS tooling has matured to the point where the daily workflow (router deployment, schema registration, can-compose checks) is covered without a $2,500/mo commercial subscription.
When buying makes sense
Apollo GraphOS earns its keep when the team needs hosted composition validation at CI scale across many subgraphs, usage analytics tied to schema fields, and team management features that OSS registries don't provide without additional integration work. The deployment-tracking and field-level usage analytics that tell you which parts of your schema are actually queried before you deprecate them are genuinely useful governance features that require real work to replicate from raw router logs. At high service counts with frequent schema changes, the hosted governance layer reduces coordination overhead in ways that matter. The cost gap against free OSS is real, so the question is whether the governance features are load-bearing for your team's schema change workflow.
The desk read
A supergraph schema encodes your entire service topology. Subgraph boundaries, federation directives, and schema evolution rules reflect specific architectural decisions. A competitor seeing your schema design would understand how your services are structured and where your domain boundaries land. Apollo's March 2026 discontinuation of managed cloud routing pushed significant adoption toward self-hosted OSS alternatives, and the timing accelerated maturity of tools like Hive and the WunderGraph Cosmo router that were already viable.
Apollo Router is Apache-2.0 licensed. GraphQL Hive is MIT-licensed and self-hostable. Multiple teams now run production federation without commercial tooling. Buying earns its keep when the team needs hosted schema registry management, composition validation at CI scale, and team collaboration features that OSS doesn't provide without additional integration work. Apollo GraphOS at $2,500/mo is a harder sell for teams whose router and schema registry needs are covered by self-hosted OSS. The build case gets serious when your schema change velocity is high enough that you'd invest in governance tooling anyway, and when owning that layer is faster than waiting on a vendor's release cycle.
Frequently asked
What is GraphQL Federation & Gateway Platform software?
GraphQL Federation & Gateway Platform software lets organizations combine multiple GraphQL services into a single unified API that clients query through a central gateway, enabling independent teams to own their service schemas while presenting a coherent API surface. It handles schema composition, query planning, and routing.
When does building GraphQL Federation & Gateway Platform make sense?
Building with OSS tooling like Apollo Router, GraphQL Hive, and WunderGraph Cosmo makes sense for GraphQL-first orgs with engineering capacity — all three are Apache-2.0 or MIT-licensed and cover the daily router and schema registry workflow at near-zero cost.
When does buying GraphQL Federation & Gateway Platform make sense?
Buying earns its keep at high service counts where hosted composition validation at CI scale, field-level usage analytics, and team management features reduce coordination overhead in ways that are hard to replicate from raw OSS components.
What are the main GraphQL Federation & Gateway Platform vendors?
Representative vendors include Apollo GraphOS, Inigo, The Guild GraphQL Hive, WunderGraph Cosmo. B4 Pro scores the full set.