Webflow versus Next.js and Astro for a content site in 2026
A Webflow site plan with the CMS costs $25 a month. An Astro site on a static host costs roughly nothing to serve and about a day of engineering to stand up. Neither figure tells you which to pick, because the price of a content site was never the hosting bill. What decides it is who can change the site, how fast it loads for real visitors, and what happens the first time you need something the builder does not do.
What $25 a month actually buys
Webflow's site plans start at a free Starter tier with a webflow.io domain, 2 static pages and 1 GB of bandwidth. Basic is $15 a month billed yearly and lifts that to a custom domain, 300 static pages and 10 GB. Premium, at $25 a month billed yearly, is the first plan with the Webflow CMS, and it adds site search, code components and bandwidth tiers running from 50 GB to 2.5 TB. For a marketing site with a blog attached, Premium is the plan, and $25 is a fair price for what it does.
The bill tends to grow sideways rather than upward. Webflow splits site plans from Workspace plans, and a few things an engineer treats as table stakes sit on the Workspace side of that line. Code export needs Workspace Core at $19 a month. 301 redirects need Workspace Growth at $49. The Optimize add-on for A/B testing starts at $299 a month for 25,000 page views. None of these are unreasonable on their own. They just arrive one at a time, and each arrives as a subscription.
Then the ceiling. Above Premium, the next platform tier Webflow sells is Team at $2,500 a month on an annual contract, which is where publishing workflows, the site activity log and the site API live. Between $25 and $2,500 there is nothing at all. If you outgrow Premium because a second editor needs staged publishing or scoped permissions, the smallest step available to you is a hundredfold increase. That gap is the most consequential number on the pricing page, and it says nothing about the quality of the Designer, which remains excellent.
What you get for owning the code instead
Astro publishes a field comparison drawn from HTTP Archive and the Chrome UX Report: 71% of real Astro sites pass Core Web Vitals, against 48% of WordPress sites and 35% of Next.js sites. Those are measurements from actual visitors on actual devices, not lab scores from a clean laptop. Astro sits at the top because its default is zero client JavaScript and you opt into hydration one component at a time. Next.js gives you more, and ships more, which is the honest reason its field number is lower. For a content site that is a real argument for Astro over Next.js, and a weak argument for either over Webflow until you look at what you can fix when a number goes bad.
That last part is where it matters. When we moved Stay World Class off Webflow and Xano onto Next.js, NestJS and Supabase, the Lighthouse performance score went from 55.91 to 91 and largest contentful paint came down 77%. None of that came from a clever trick. It came from being allowed to delete scripts, change how images were served and control the render path, which on a hosted builder is not yours to change. The data moved through a four stage ETL that mapped every legacy record ID onto a PostgreSQL UUID, so nothing lost its history on the way across.
The editing story is the objection worth taking seriously, and it has a real answer now. Astro content collections give you typed Markdown with schema validation, or a loader pointed at whatever CMS your writers already like. Your marketer keeps a familiar editor and stops filing tickets to change a headline. Setting that up is genuine work, somewhere between a day and a week depending on how much content you are carrying, and anyone who tells you it is an afternoon is selling something.
When staying on Webflow is the right answer
If your site is a brochure with a blog, has no custom logic behind it, and the person who changes it most often is a designer rather than an engineer, Webflow wins and it is not close. Visual iteration in the Designer is faster than any code workflow I have used, and publishing without a deploy pipeline is a real feature, not a compromise. A site like that on Premium at $25 a month is a good deal, and moving it would buy you a Lighthouse score nobody asked for.
The same holds if nobody on your team writes code and there is no budget to hire someone who does. Migrating then swaps a $25 monthly bill for a standing dependency on one contractor, which is a worse position than the one you were in. Platform risk is real, but so is bus factor of one.
Here is the test I would actually apply. Go back through the last six months and count how many requests were answered with some version of "Webflow can't do that", or were solved with an embed, a custom code block and a Zapier step holding it together. If the answer is none, you are on the right platform and you should enjoy it. If you can name three, you are already paying for the migration in workarounds and getting none of the benefit.
