Cloudflare Pages vs Vercel
Edge JAMstack.
Overview
Cloudflare Pages and Vercel both deploy front-end apps to a global edge, but the centre of gravity is different. Vercel is Next.js-first with deep server component and ISR support; Cloudflare Pages bundles into the Workers runtime with first-class Durable Objects, KV, and R2. Pick on framework gravity and the platform primitives you actually plan to use.
- Vercel. Best-in-class Next.js support, ISR/streaming SSR, mature analytics and image optimisation, deep GitHub integration, premium pricing once you outgrow the free tier.
- Cloudflare Pages. Workers runtime under every route, KV/D1/R2/Queues at the edge, generous free limits, framework-agnostic bias.
- Operational fit. Vercel wins where the app is Next.js and DX is the priority; Cloudflare wins where the edge primitives (storage, queues, durable state) are part of the architecture.
- Per-app decision and exit cost. Both accept standard frameworks; lock-in shows up in adapters (next-on-pages, vercel/og) and platform APIs.
The approach
Match the host to the framework, the data model, and the team's existing cloud gravity. Edge primitives matter once you go beyond static pages.
- Framework gravity check. Heavy Next.js usage with ISR and Server Actions leans Vercel; framework-agnostic apps lean Cloudflare.
- Edge-primitive inventory. If the app needs KV, Durable Objects, queues, or R2 at the edge, Cloudflare keeps everything on one platform.
- Cost modelling at scale. Vercel function invocations and bandwidth get expensive past hobby-tier; Cloudflare's request pricing scales differently.
- Document the choice and the trigger to revisit. Capture rationale and the metric (traffic shape, framework move, edge-primitive need) that would flip it.
Why this compounds
The right host keeps paying back: deploys ship in seconds, the runtime serves the framework's strengths, and the bill scales with traffic instead of with framework friction.
- Operational consolidation. One host per app type means one deploy pipeline, one set of env-var conventions, one monitoring surface.
- Cost predictability. A documented choice keeps surprise bills out of the next quarter.
- Engineering velocity. A host aligned to the framework removes the workarounds that drain Friday afternoons.
- Decision trail for the next app. Each documented choice teaches the next team which questions to ask, not which host to default to.