Framer to a custom site
A design tool needs a designer
Framer is very good software and this page is not going to pretend otherwise. It describes itself as a design agent native to the canvas, and the logos on its homepage are Miro, Perplexity, Mixpanel and Zapier3— software companies with design teams on staff. It is built for them, and it shows.
Most small businesses hire a designer exactly once, for the build. Afterwards they are holding a professional canvas with nobody to drive it, on a plan that charges per editor, and Framer’s own documentation says the site cannot be exported1,2.
The fit problem
Three mismatches, none of them faults
Everything below is something Framer publishes about itself. The question is not whether the product is good. It is whether it was designed around a business like yours.
01
It is a design tool, and it says so
Framer describes itself as a design agent native to the canvas, and the logos on its homepage are Miro, Perplexity, Mixpanel and Zapier. Those are venture-backed software companies with design teams on staff. Nothing about that is a criticism; it is a description of who the product is aimed at, written by the people who aim it.
Framer's own positioning: "the design agent for every step from idea to launch", "native to the canvas". Flagship customers are software companies rather than local businesses.3
02
It is priced per person, like software for a team
Twenty dollars a month for each additional editor, on top of the site plan. That is completely reasonable for a marketing team of six who are all in the canvas every week. For a business where two people occasionally need to change a phone number, it is a subscription shaped like somebody else's org chart.
Additional editors are $20/month each, charged per seat on top of the site plan; content-only editors are $10/month on Pro.2
03
You cannot take the site with you
This is the one that matters most and the one nobody checks before signing up. There is no export. If you ever want to leave — different agency, different budget, different direction — you do not move the site, you rebuild it from scratch somewhere else.
"Framer does not offer HTML export for self-hosting." Published sites "are not fully static and cannot be exported as standalone HTML files" because optimisation features depend on Framer's infrastructure.1
What it costs to keep
Metered pages, metered collections, and a price per person
Worth reading as a description of the intended customer rather than as a complaint about cost. Every line here makes sense for a marketing team and rather less for a plumber.
| Plan | Price | Pages | Content and extras |
|---|---|---|---|
| Free | $0 | Framer subdomain only | 10 collections, 1 GB bandwidth |
| Basic | $10/mo | 30 pages | 2 collections, 1,000 items |
| Pro | $30/mo | 150 pages, then $20 per 100 | 10 collections, then $40 per 10 |
| Each extra editor | $20/mo | Charged per seat, on top of the plan | Content-only editors $10/mo on Pro |
Figures from Framer’s published pricing2. The row to look at is the last one. A service business with twelve services, a few locations, a blog and some case studies is already past Basic’s thirty pages and two collections, and if three people need to touch the site you are paying for two extra seats on top of that. None of which is unfair — it is simply priced for somebody else.
The alternative
A site you actually own
The point of moving is not that the new site is faster, although it usually is. It is that the thing you paid for becomes an asset you hold rather than a subscription you rent.
- Built in Next.js and deployed from a repository in your name, not mine
- If you leave, you take the code — there is nothing to rebuild and nobody to ask
- Hosted on Vercel, with scaling, security and the build pipeline handled here
- No per-editor pricing, because there are no seats to buy
- Changes go through me at a pace that suits how often you actually change things
If your site is a WordPress install rather than a Framer project, the argument is a different one and it is on the WordPress page.
And if you are comparing Framer against the other visual canvas rather than against a CMS, Webflow is on the Webflow page. Showit sells to the same instinct from the other end, a canvas for people whose work is visual, and it is on the Showit page.
Against this argument
Four reasons to keep what you have
You have a designer, or you are one. Then everything above stops applying. Framer in the hands of somebody who can use it is excellent, fast, and produces work that is hard to match on a budget. The mismatch is about the driver, not the car.
The site is small and you never touch it. If it is eight pages, it looks good, and nobody has needed to change anything in a year, the lock-in is theoretical and the per-seat pricing is one seat. Paying five figures to solve a problem that has not happened is a bad trade.
It is genuinely well engineered. The hosting is good, the sites are fast, and the reason there is no export is that real infrastructure sits behind the published page. That is a legitimate technical position rather than a trap, even though the effect on you is the same.
Rebuilding costs real money. Because there is no export, moving means building the site again. That is a genuine cost with no partial credit, and it should be justified by something concrete rather than by a principle about ownership.
The honest trigger is narrow: nobody on your side can confidently change the site, the seat and page costs have crept past what it is worth, or you need to own the thing for a reason you can name. Short of one of those, stay.
If you do move
Four things about rebuilding rather than transferring
It is a rebuild, and the design survives
Because there is no export, the site is reconstructed rather than transferred. That sounds worse than it is: the design, the copy and the structure all carry across, and in practice the reconstruction is the fastest part of any migration on this site because the decisions have already been made. What you are paying for is the build, not the thinking.
The CMS content comes out
Blog posts, case studies, team members, whatever collections exist can be exported as data even though the site cannot be exported as a site. That content moves into the new build with its structure intact.
The URLs are mapped from a crawl
Every address that exists today gets a one-to-one redirect, built from a crawl of the live site rather than from a list somebody remembers. Framer sites tend to have clean, shallow URL structures, which makes this one of the easier redirect maps to build.
The animation question, answered honestly
Framer is very good at motion, and some of it is genuinely hard to reproduce. Most of it is not — scroll effects, reveals, hovers all have direct equivalents. Anything elaborate gets looked at individually before anything is quoted, and occasionally the answer is that a particular effect is not worth rebuilding.
The method is the crawl, redirect map and rebuild set out on the migration page, and what the new site is like to own is on the web design page.
Pushback
Including the one about no-code
Often you would not, and this page is not arguing that Framer builds bad sites — it frequently builds better-looking ones than what replaces them. The question is narrower: can you change it, and can you keep it? If updating a page means opening a professional design canvas and making design decisions, and if there is no designer on your side of the table any more, then you own a site you cannot maintain. Add that there is no export, and the decision to leave gets more expensive the longer you wait rather than less.
No-code means you do not write code. It does not mean you do not need design skill, and on a canvas-based tool those are different things. Changing text is genuinely easy. Adding a section, adjusting a layout so it still works on a phone, or keeping a new page consistent with the rest of the site are design tasks, and a tool that gives you full creative control also gives you full responsibility for the result. That is the right trade for a designer and the wrong one for somebody who wanted to update their prices.
Not through Framer. Their help documentation states plainly that Framer does not offer HTML export for self-hosting, and explains why: published sites are not fully static and rely on Framer's own infrastructure for pre-rendering, image handling, font subsetting and caching. Third-party scrapers exist and produce something that looks like your site, but what they give you is captured output rather than a maintainable codebase, and you would not want to build the next three years on it.
A site built in Next.js, hosted on Vercel, deployed from a repository that belongs to you. That last part is the point of the exercise. If you fall out with me, change direction, or want to hand it to somebody in-house, you take the code and go — no rebuild, no exit fee, no conversation about whether you are allowed to. Scaling, security and the build pipeline are handled on my side, so what you own is the asset rather than the maintenance.
It should look the same or better, because the design already exists and is being rebuilt rather than reinvented. What genuinely differs is elaborate motion: Framer's canvas makes certain animations trivial that take real work elsewhere. Most effects have direct equivalents and get rebuilt without drama. If something is both elaborate and load-bearing, you will hear that before you are quoted, not after.
Usually two to five weeks, at the faster end of anything on this site, because the design decisions are already made and the content is already written. Cost sits in the $2,500 to $15,000 web design range depending on page count and how much needs rethinking rather than rebuilding. Hosting afterwards is a fraction of a Framer plan with several editor seats on it.
Sources
All three are Framer’s own pages
Unusually for these pages, there is no third-party research here and none is needed. Every claim on this page is something Framer publishes about its own product, its own prices, or its own customers.
- 1Can I export my website to HTML and self-host it?
Framer · Checked September 2026 · Framer's own help documentation
- 2Pricing
Framer · Checked September 2026 · Published plan limits and per-editor seat pricing
- 3Framer homepage
Framer · Checked September 2026 · The company's own positioning and customer logos
The company describing itself
One claim is deliberately absent. Several third-party articles state that a Framer site is taken offline when a subscription lapses. Framer’s own cancellation page says only that plan benefits remain active until the end of the billing cycle and does not document what happens afterwards, so it is not asserted here. Prices and limits are current to September 2026.
Work out whether you can drive what you own
Send the address, your plan, and how many people need to change things. You will get back what it costs you a year, what rebuilding would cost, and whether the honest answer is to leave it alone.