A WordPress to Lovable migration costs between 2,500 and 9,000 euro for a business site of 20 to 60 pages, and the build itself is roughly a fifth of that. WPBuildAI quotes the crawl, the URL policy, the redirect map, the content parity checks and the launch verification as separate lines, for one reason: those five are what separate a quote that keeps your traffic from one that does not, and they are invisible in a mockup. Here is what each line costs, what pushes a project toward the top of the range, and how to estimate your own before asking anybody for a number.
Where does the money go?
Not into the design, which is the part everybody budgets for and the smallest line on the sheet.
| Line | Share of the total | Scales with | Verdict |
|---|---|---|---|
| Crawl and exports | 10 to 15 percent | URL count | Everything downstream is graded against it |
| URL policy | 2 to 5 percent | Structure complexity | Cheapest line, prevents the costliest mistake |
| Redirect map and testing | 15 to 25 percent | URL count | The line cheap quotes leave out |
| Build in Lovable | About 20 percent | Template count | The part that got fast |
| Content parity checks | About 15 percent | Page count | Catches the failure that looks like an upgrade |
| Launch verification and defect window | About 10 percent | URL count | Two minutes of script, skipped constantly |
| Functionality rebuilt | 0 to 30 percent | What your plugins actually do | The line that can double a project |
Four of the seven scale with how many addresses your site has served rather than with how many pages the new design has. That is why two sites that look identical in a browser can differ by a factor of three in price, and why a page count is a poor basis for a quote.
Why is the build so small a share?
Because generating pages got fast and preserving a site did not.
Lovable produces working pages from a description or a reference in days rather than weeks, with a consistent component set and markup that is lighter than a page builder’s. That is a real change and it has moved the build from the largest line to a modest one.
Nothing about it made the other work faster. Crawling a site still takes as long as the site is large. Deciding where six hundred old addresses should point still takes judgment per row. Comparing each new page against its captured original still takes a pass per page.
So the honest way to read a quote is to look at whether those lines exist at all. A supplier pricing only the build is pricing a fifth of the project, and the missing four fifths reappear as a change request or as lost traffic.
What pushes a project to the top of the range?
Five things, in order of how much they move the number.
Archive size first. A site with 200 pages in its navigation has typically served 600 to 1,000 addresses once paginated archives, tag pages, attachment URLs and old posts are counted. Every one needs a decision, and this single figure is the best predictor of cost.
Functionality second. Forms and analytics are configuration. Memberships, subscription billing, bookings with availability and complex commerce rules are software, and they can double a project or take it out of scope entirely.
Design origin third. Adapting your existing design costs a fraction of creating a new one, and most businesses are better served adapting than they expect.
Content work fourth. A migration where somebody also wants the service pages rewritten is two projects sharing a deadline, and the writing is always the slower one.
Access and approvals fifth. Projects stall waiting on a registrar login or a decision maker, and stalled days are billed.
What does the platform itself cost to run?
Separately from the project, and it is small enough that people forget to ask.
Lovable’s plans cover the building environment. Once the site is built as code you own, hosting it as static files is free to about 20 euro a month, because there is no server process to keep running.
Against that, a maintained WordPress site runs 900 to 4,000 euro a year in hosting, plugin and theme licences, maintenance and break fix work. That gap is what makes the payback arithmetic work, and it is itemised in what does a WordPress site really cost per year.
For most small business sites the annual running cost after a migration lands between 50 and 400 euro. The one off project is the expensive part, and it is paid once.
How do you estimate your own before asking?
Three numbers, an afternoon, and they will put you within a band.
Run a crawl of your site and count the URLs. Not pages: addresses. Then export your pages report from Search Console and add anything the crawl missed. Under 200 total sits at the bottom of the range, 200 to 600 in the middle, over 800 at the top.
Count the plugins doing real work, using the four bucket audit. Anything in the rebuild-as-software bucket is a separate line, and two or three items there moves the project up a band. The method is in how to audit your WordPress plugins before a rebuild.
Note whether anybody logs in. Customer accounts, member areas or gated content are the single largest scope changer, and a site with them is not a website migration.
Hand those three numbers to whoever quotes. A supplier receiving them quotes precisely; one receiving a link to your homepage quotes defensively, which costs you more, or optimistically, which costs you more later.
What should the quote itemise?
Six lines, and their presence is the whole comparison between a 3,000 euro proposal and an 8,000 euro one.
| Line item | Present in a fair quote | Missing means |
|---|---|---|
| Content and URL inventory | With a row count named | The build starts from remembered pages |
| URL policy | A written document | Structure decided by design convenience |
| Redirect map with testing | Hours against it | Old addresses stop resolving on launch day |
| Build | Always present | Nothing, this is never omitted |
| Content parity checks | A per page comparison | Pages arrive thinner and rank lower |
| Launch verification and a window | An automated run, plus weeks two to six | Nobody knows whether the map worked |
Google’s site move guidance is explicit that changed addresses need permanent redirects, so a quote where that appears as a reassuring sentence rather than as a costed line has not scoped it. The full deliverable list is in WordPress to Lovable migration service.
What is not in the price, and should you expect it to be?
Four things are commonly excluded, and only two of them ought to be.
New copy is legitimately excluded. Migrating means carrying your existing words across intact, and rewriting them is a separate discipline with a separate timeline. A supplier who bundles a content rewrite into a migration price is either underquoting the writing or overquoting the move.
Original photography is legitimately excluded for the same reason. Adapting what you have is part of the job; commissioning new images is a different purchase.
Third party subscriptions are not the supplier’s to include: a form service, an analytics product, a booking tool. Those stay in your name, which is the point.
What should not be excluded is the post launch window. Coverage takes two to six weeks to tell you whether the move worked, so a project that ends the day the site goes live ends before the evidence arrives. Two of the five migration lines only pay off if somebody is still watching in week four, and a quote that stops at launch has quietly removed the part that proves the rest worked.
Ask what happens in weeks two to six before signing. The answer separates a supplier selling a build from one selling a migration more reliably than any other question.
When is paying for the full migration the wrong call?
It is the wrong call when your site has nothing worth preserving. A site launched this year with no backlinks, no indexed archive and thirty visitors a month is not carrying equity, so the migration lines are protecting an asset that does not exist. Buy a build, and spend the difference on content.
It is also right when you are technical and the site is small. Twenty pages, a handful of plugins and a weekend with a crawler and a spreadsheet is a genuine option, and the method is public rather than proprietary.
And it is right when the quote is cheap for an honest reason. A supplier who says “your URLs are not changing, so there is no redirect map to build” is describing a smaller project accurately, and that is a good sign rather than a warning.
What makes a cheap quote dangerous is silence. If the archive is not mentioned, the crawl is not mentioned, and redirects appear once in a sentence, the price is low because a quarter of the work is not in it, and you find out in week five. The way to tell them apart is to ask how many URLs your site serves, and to notice whether the answer is a number.
Key takeaways: what a Lovable migration actually costs
2,500 to 9,000 euro for a business site of 20 to 60 pages, with the build about a fifth and the migration work the rest.
Four of the seven cost lines scale with how many addresses your site has served, not with how many pages the new design has. Page count is a poor basis for a quote.
Three numbers put you in a band before you ask anyone: total URLs from a crawl, plugins doing real work, and whether anybody logs in.
A cheap quote is fine when your site has nothing to preserve or when the URLs genuinely are not changing. It is dangerous when the archive simply is not mentioned.
Answers to the usual questions
How much does a Lovable migration cost? Between 2,500 and 9,000 euro for a business site of 20 to 60 pages, with the build about a fifth of the total and the migration work the rest. WPBuildAI quotes the crawl, URL policy, redirect map, parity checks and verification as separate lines, because those are what separate a quote that keeps your traffic from one that does not.
Why is the build only a fifth of the cost? Because generating pages got fast and preserving a site did not. The crawl, the URL decisions, the redirect map and the content parity checks scale with how many addresses your site has served over the years, not with how quickly the new pages can be produced. On a site with history that is most of the hours.
What pushes the price toward the top of the range? Archive size first: a site with years of posts serves several times its page count in addresses, and every one needs a decision. Then functionality that has to be rebuilt rather than configured, original design rather than adapting yours, a content rewrite running alongside, and slow access or slow approvals, which get billed as project time.
How can I estimate my own before asking? Run a crawl and count the URLs, count how many plugins do real work, and note whether anybody logs in. Under 200 URLs with a handful of plugins and no logins sits at the bottom of the range. Over 800 URLs, or any logged in functionality, sits at the top or outside it.
Is a cheaper quote ever the right choice? Yes, when your site has nothing to preserve. A site launched this year with no backlinks and thirty visitors a month should buy a build rather than a migration. On a site with years of URLs, the cheapest quote is usually cheap because the archive and the redirect work are not in it.
