Customer story
How Solin serves 2.5M database queries a day with Prisma
Solin, a fitness-creator marketplace, ran a serverless backend that kept exhausting its database connections during traffic spikes, causing failed requests and a poor experience. With Prisma ORM and Prisma Accelerate, it now serves 2.5 million database queries a day with zero connection issues.
Accelerate's connection pooling ended the failures, and per-query caching made landing pages lightning fast, which fed directly into better conversion.
Customer
About Solin
Solin is a fitness marketplace that connects fitness creators with consumers. Creators sell workout programs, challenges, memberships, and cookbooks, while consumers get community features and transformation-tracking tools. The platform runs a fullstack serverless app in the fitness-technology space.
Industry
Fitness technology / creator marketplace
Stack
Remix, Vercel serverless, PostgreSQL on Heroku
Using
Prisma ORM · Prisma Accelerate
The challenge
Solin runs a fullstack Remix app on Vercel with serverless functions, and that setup hit a familiar serverless wall: too many database connections. As the user base and query volume grew, its PostgreSQL database on Heroku couldn't hold enough connections, and traffic spikes exhausted the limit.
When the connections ran out, requests failed and the user experience suffered:
Serverless exhausts connections. Many concurrent functions opened more connections than the database could sustain.
Spikes caused failed requests. When the limit was hit, requests dropped and users felt it.
Growth made it worse. A growing audience and rising query volume kept pushing against the ceiling.
Why Solin chose Prisma
Solin already used Prisma ORM to query its database. When connection limits became the bottleneck, Prisma Accelerate added the connection pooling and caching the serverless backend needed to scale, without changing how the team wrote queries.
Connection pooling at scale
Accelerate's pool manages database connections so serverless functions stop exhausting the limit.
Per-query caching
ttl and swr options set per query, ideal for landing pages whose content rarely changes.
No change to the query layer
Accelerate layered onto the Prisma ORM queries Solin was already writing.
“Accelerate is a perfect fit for landing pages. We are able to take advantage of caching to speed up queries and reduce latency, making them lightning fast. This obviously means we have a faster landing page, leading to better conversion.”
Blake Carroll
CTO, Solin
How they use Prisma
Solin's API layer uses Prisma ORM to query the database, with Prisma Accelerate's connection pool sitting in front to manage connections at scale across its serverless functions. On top of pooling, the team uses Accelerate's caching on a per-query basis, applying ttl (time-to-live) and swr (stale-while-revalidate) where it helps most, particularly on landing pages with content that rarely changes.
That combination keeps the serverless backend from running out of connections during spikes, while caching keeps high-traffic pages fast.
generator client {
provider =
}
model User {
id String @id @default(cuid())
email String @unique
DateTime
posts Post[]
}
model Post {
String @id
author User @relation
}
Schema valid
The results
Prisma took Solin from connection failures to serving millions of queries a day, reliably.
2.5M queries / day
Database queries served daily through Prisma.
Zero connection issues
No connection problems since adopting Accelerate.
Faster landing pages
Cached pages load lightning fast, improving conversion, while reduced server load lowers costs.
Build on the same stack as Solin
If your serverless app is straining its database connections, Prisma can pool and cache its way past the limit.
- Free to start, no credit card required
- One TypeScript stack your agent can operate end to end
- Trusted by 500K+ developers