What is Webiny in 2026?
Webiny is an open source, self-hosted platform that runs on AWS serverless services: Lambda, DynamoDB and S3. Its Community Edition is MIT licensed and includes a Headless CMS, a Website Builder and a File Manager. The vendor’s homepage now positions it as an AI-programmable CMS, with an MCP server so developers can use AI tools to extend the platform. It lists Publishing Workflows and Tenant Manager among its other products, so enterprise features exist beyond the free edition.
The key constraint is stated in the project’s own README: Webiny runs only on AWS. There is no GCP, Azure or on-premise option.
Who Webiny suits
Webiny suits a company that already runs on AWS, has TypeScript engineers, and wants a CMS inside its own cloud account for data residency, security review or cost control. It also suits SaaS builders who need multi-tenancy, where each customer gets its own content space.
It is a poor fit if you want a CMS that runs on a laptop and a small VPS, if nobody on the team has AWS experience, or if you want a vendor to carry all operations. You own the infrastructure, even if the code is deployed for you.
What does it cost to own?
There is no per-seat licence in the Community Edition. The cost is AWS usage, deployment pipelines and engineers who understand the stack. The vendor links a hosting cost calculator from its site. We do not quote AWS figures, because they depend on traffic and content volume. We scope the build after a short audit.
When we would pick something else
- No AWS commitment, and a self-hosted Node.js backend: Strapi or Directus.
- A managed SaaS where someone else operates the CMS: Contentful or Sanity.
- A GraphQL-first hosted option: Hygraph.
Our GraphQL-ready headless CMS list puts these side by side.
What we build on Webiny
- Models in the Headless CMS app with field level permissions and localization.
- Infrastructure as code with Pulumi, separate dev, staging and production environments, and a CI pipeline.
- Plugins for the admin app and API, written in TypeScript against Webiny’s extension points.
- Next.js or Astro reading the GraphQL API, with preview and cache invalidation.
- Tenant isolation and per-tenant content for SaaS products.
- Content loaded by script from the old CMS, with redirects. Our migration checklist covers the steps.
How a Webiny project runs
Before we write anything, we agree on the AWS account structure, regions, and who holds deployment access. A sandbox environment goes up first, so the team sees the real cost and deploy time of the stack, not an estimate.
After that we model content, build extensions and the frontend, and set up monitoring for Lambda errors, API latency and spend alerts. Load tests run against a staging environment with realistic content volumes. We hand over infrastructure code and runbooks, so your cloud team can operate the system without us.
For a developer who has shipped on this stack, see how to hire a Webiny developer. The broader comparison is on our headless CMS hub.
