Skip to content
Deployment platformPartner

Netlify

The platform that invented deploy previews, and still the most even-handed home for a multi-framework estate. Here's where Netlify earns the shortlist in 2026, and where a Next.js estate belongs on Vercel instead.

Need a partner to implement or migrate?

This page gives you an honest read on where Netlify fits. If you’re rethinking the hosting layer of a whole web estate, start at Hosting & delivery. If you want a team to plan and run the move, see Website migration.

What is Netlify?

Netlify is a front-end deployment platform. You connect a Git repository and every push becomes an immutable, atomically deployed version of your site, served from a global CDN, with its own Deploy Preview URL and one-click rollback to any previous build. Netlify invented that workflow, and effectively every competitor now copies it.

Around the delivery layer sits a platform: serverless Functions, Deno-based Edge Functions, background and scheduled functions, Blobs for object storage, an Image CDN, Forms, and Netlify Database, which is managed serverless Postgres (built on Neon) that spins up an isolated branch database for each Deploy Preview. The deliberate difference from Vercel is neutrality: Netlify treats Astro, Nuxt, SvelteKit, Eleventy, Hugo and Next.js as equal citizens rather than optimising for one framework it also owns.

What changed in 2026

Two things, and both matter to how you’d budget for it.

Pricing moved from seats to credits. As of April 2026, the Pro plan is a flat team price with unlimited members instead of a per-seat charge. That covers Owners, Developers, Publishers, Git contributors, and Reviewers. Consumption (production deploys, compute, bandwidth, web requests, AI inference) is metered against a monthly credit allowance, with credits topped up as you use them. For an organisation with dozens of occasional contributors who mostly need to open a preview and approve it, this is a real reduction versus per-seat platforms. For a high-traffic estate, it moves more of the bill into a meter you now have to forecast.

The workflow was re-pitched around AI agents. Netlify’s current positioning is “agent experience”: coding agents run against the real project, and every change, human or agent, lands as a Deploy Preview that a person reviews before production. Strip the framing and the substance is the old Netlify argument, which is still the right one: the preview URL is the review gate, and everything else is built around protecting it.

Where Netlify fits

  • Multi-framework estates. Large organisations rarely have one front end. Ten years of brand sites, campaign microsites, and documentation arrive on different frameworks, from different teams and different agencies. Netlify’s framework neutrality means you consolidate them onto one platform, one deploy model, and one set of access controls without rewriting them first.
  • Organisations with many occasional contributors. Brand managers, regional marketers, legal reviewers, external agencies. Under flat-price Pro they cost nothing extra to add, and they get the one thing they actually need: a real URL to look at before anything ships.
  • Teams already on it. Netlify is the incumbent on a lot of marketing estates. If it works and the bill is predictable, migrating away is rarely the highest-value thing to do with a rebuild budget. The CMS layer usually is.

Where it doesn’t

Next.js estates that live at the front of the release curve. Netlify runs Next.js through its own runtime rather than as a first-party product. It has improved substantially, but adapter-based support structurally trails the framework, and the gap shows up in exactly the features large content sites depend on: ISR behaviour, image optimisation, middleware. If the estate is Next.js and you want each release working as designed on day one, that’s Vercel.

Bandwidth-heavy delivery. At sustained high traffic, metered bandwidth is the wrong shape. Cloudflare’s no-egress-metering model or your own CDN contract will win on cost. See Cloudflare, or bring your own cloud and CDN from the hosting layer.

Deep platform coupling. Forms, Identity, Blobs and Netlify Database are genuinely convenient, and they are the parts of the platform that do not port. Framework code moves between hosts in an afternoon; a five-year-old estate with authentication and submitted records living inside its host does not. We use Netlify’s delivery and compute freely, and keep the stateful surface small and replaceable on purpose.

When teams choose it, and when they don’t

Choose Netlify whenLook elsewhere when
The estate spans several frameworks and you want one platform that treats them all as first-classIt's a Next.js estate and framework features need to work on release day, which points to Vercel
Many people need to review previews and few of them write code, so flat-price Pro removes the seat taxSustained high bandwidth makes metered delivery the wrong pricing shape
You're already on Netlify, it's stable, and the rebuild budget is better spent on the CMS layerData residency or existing cloud commitments push you into your own tenancy
You want deploy previews, atomic deploys, and instant rollback as platform defaults, not a build projectYou need long-running services, workers, or heavy backend jobs, so pair a platform like Render behind it

Netlify pricing: the credit model, in plain terms

Netlify no longer prices by seat. It prices by credits, a single unit that every kind of usage is charged against, and that one change is the thing to understand before you budget for it.

Four tiers, each with a monthly credit allowance: a free plan for individuals, a low-cost Personal plan, a flat-priced Pro plan for teams with unlimited members, and a custom Enterprise plan that carries the items procurement asks about (SSO and SCIM, log drains, organisation-level management, high-performance builds, 24/7 support and a contractual uptime SLA). When the allowance runs out you buy top-up credit packs; the plan price is a floor, not a cap.

Credits are consumed by everything the platform does, at published rates. As of August 2026 that’s roughly: 15 credits per production deploy, 20 credits per GB of bandwidth, 10 credits per GB-hour of function compute, and 2 credits per 10,000 web requests, with AI inference priced per model. Top-up packs put a credit at a fraction of a cent, so a production deploy lands around ten cents and a gigabyte of bandwidth around thirteen. Verify the current rates on Netlify’s pricing page. This is the layer that moves.

The consequence: small usage is genuinely cheap and forecasting is harder, because one number now hides four different growth curves. The drivers to model before you commit are bandwidth and request volume (the largest line on a content-heavy estate), function compute, and deploy frequency across every branch and preview environment you run. Take a real month of traffic and deploy activity, convert it to credits at the rates above, and compare that to the plan allowance, rather than reasoning from the plan price alone.

The Bejamas take

We’re a Netlify partner, and we still recommend by estate rather than by badge. Netlify is the most even-handed platform in this category, and the flat-price team plan can make it noticeably cheaper than per-seat alternatives for organisations where lots of people review and few people commit. Where we push back is on the stateful extras: the moment Forms and Identity hold something the business needs, you’ve bought a dependency, not a feature. If you’re weighing it directly against the alternatives, read Vercel vs Netlify, or self-hosting vs both. In an audit we model your traffic, contributor count, and framework mix across all three paths and recommend the one the numbers support.