About

As one of the new generation of Developer Experience-first companies, Prisma builds on the latest web technologies to create compelling, productive, environments for data-driven application development. The founders initially built a backend-as-a-service platform called Graph.cool, based on GraphQL, but are now pivoting to a broader story, where GraphQL is used as a data access layer for back end data stores, abstracting away complexity for front-end developers and making it easier to build apps with persistent storage.

A few related trends are coming together here. Relational databases, notably PostgreSQL and MySQL, are having a resurgence. GraphQL is going bonkers, as REST begins to look a bit long in the tooth, and application development patterns change. Type-safety is back, as evidenced by the programming languages gaining attention in the industry right now – see for example Rust. JavaScript continues to dominate front-end development, with TypeScript in particular emerging as a force in its own right. Prisma is a modern, open source Object Relational Mapping (ORM) play, tapping into the intersection of all of these trends.

The Prisma platform allows back end engineers to build database-enabled back end APIs with TypeScript and Go, for front-end developers to take advantage of.

Prisma is incorporated in the U.S., but half of its employees are based in Berlin, with the rest distributed in the U.S., Canada and Europe. The team is led by Søren Bramer Schmidt (CEO) and Hervé Labas (VP of Product).

Prisma’s vision (according to the company):

Prisma helps data owners and application developers to effectively collaborate by introducing an Application Data Platform to standardize workflows involving the two groups. This significantly speeds up application development while giving data owners operational insight and full governance control.

Prisma’s seed investors are pretty impressive-including Augusto Marietti (CEO Kong), Guillermo Rauch (CEO Zeit), Spencer Kimball (CEO CockroachDB), and Nicholas Dessaigne (CEO Algolia). These are all Developer Experience-first founders.

Size

• 35 employees
• Prisma has raised a total of $17.1m in three funding rounds
• 2016 Pre-Seed $610k
• 2018 Seed Preferred $4.5m
• 2020 Series A $12m
• 10,000 active developers
• 100k weekly downloads

Products

The platform is built around the Prisma Schema, which includes a description of the data model, data sources connected to, and generators specifying what clients should be generated based on that model. The schema can be introspected from existing databases. Other Prisma components include:

• Prisma Client: a type-safe query builder for TypeScript (with an early access version for Go now available). Designed for Visual Studio Code, it offers auto-completion, linting, formatting and so on.
• Prisma Migrate: a schema migration tool (auto-generated, imperative), currently in beta.
• Prisma Studio: a visual data browser to simplify managing data while developing apps.

Prisma’s roadmap includes support for MongoDB, and change data capture for GraphQL subscriptions. The product however that should really help Prisma to make a broader impact is a cloud-based service aimed at developer teams.

Competitive Landscape

In the Node.js space Prisma competes with ORMs including TypeORM, Sequelize, and MikroORM.

On the GraphQL side folks like Apollo, Firebase, Hasura, StepZen, and Strapi are in the mix.

Prisma is betting on developer experience as both a key adoption vector and moat. It won’t be alone there. There are a host of plays out there seeking to win the hearts and minds of developers that grew up using modern APIs. AWS changed the game with Lambda functions and then AppSync, helping GraphQL to become the industry’s de facto standard for API creation and management. The serverless model is finding new niches. We’re seeing a return to a more two-tier model, with rich front ends, often written in React, orchestrating back end services and functions. FaunaDB calls this model client/serverless, with a nod to the earlier client-server era.

Managed data layers based on GraphQL for data access and introspection, with API management functionality for access management, throttling, etc., are going to be a key enabling trend here. We’re not going to see a renaissance of the DBA, but platform owners are going to need new tools and platforms to enable developer productivity with control.

Go to Market

Prisma is looking to establish a bottom-up approach, with developers adopting the open source product, allowing it to penetrate enterprises, then proposing a commercial offering around operational data access and management. SaaS is likely to be critical to success here. OpenSource + SaaS is the new industry model for successful startups.

Prisma also plans to partner with next generation database companies such as PlanetScale, Vectorized, and CockroachDB to jointly win customers.

PlanetScale is interesting for context here. Another Developer Experience-first startup looking to win MySQL workloads with a highly scalable open source auto-sharding architecture originally built at YouTube called Vitess. As mentioned above MySQL is seeing renewed attention, and PlanetScale’s timing is good. I have had an early look at the platform and the developer experience is slick, with a quick mean time to dopamine.

We don’t yet have a generally agreed industry term for the managed platforms being built for serving data via APIs, but a market is emerging for them. Prisma describes them as Application Data Platforms, particularly for internal use in enterprises. The existing category of API Gateways, defined by companies such as Apigee in the 2005+ timeframe, doesn’t capture it. In a recent keynote at Prisma’s first Enterprise event I described the rise of “enterprise data fabrics”. Whatever we end up calling the category, it’s likely to be a substantial play, as enterprises hire programmers that grew up with modern APIs.

Dislosure: This is an independent piece of research. Clients mentioned include AWS, Fauna, Google Cloud (Firebase), Hasura, Prisma, and StepZen.