Almost every website-builder comparison you will read was written once and never checked again. The prices in it were true on some unstated day, possibly years ago, and the page gives you no way to tell. That is the single most common reason the advice you find is wrong, and it is invisible.
So this page does the unglamorous opposite. Every price on this site carries the date it was verified against the vendor's own pricing page, and this is the log of what has moved since I started checking. It includes the things I got wrong, labelled as mine, because a change log that only ever blames vendors is just marketing with timestamps.
Last full sweep
October 11, 2026
All five builders re-verified against their own pricing pages.
Vendors that moved
0 of 5
Every published figure held for three weeks.
Next sweep due
Early January
Quarterly is the minimum that keeps a dated price honest.
How to read the labels
Four kinds of entry, and the difference between them is the whole point of the page.
The vendor moved the number, and I hold dated readings on both sides of the move. The strongest claim here.
It changed between my readings and the direction is not in doubt, but the earlier figure came from a less rigorous check than the one that replaced it.
My error, not theirs. Named, dated, and with the damage described.
Nobody here had checked before, so there is no before. Logged so a future reader can tell a stable vendor from an unchecked one.
October 11, 2026
The free plan allows 1,000 pages and 10 CMS collections, not 30 and 2
This site said the free tier was capped at 30 pages from 2026-09-22 until today. It is not. Framer's own FAQ reads: projects on the Free plan include access to 10 CMS collections, 1,000 pages, 5 MB file uploads and one free locale. The 30-page and 2-collection figures belong to the $10 Basic plan, which I read off the plan comparison table and wrongly applied to both. The plan cards state no free-tier limit at all, which is how the mistake survived a check. It reached nine pages and it reversed the advice on one of them, the fan-sites guide, where I told people a large fan site would outgrow the free tier. It would not.
A content-only editor seat is $10 a month, half the full seat
Framer lists two kinds of extra seat: an additional editor at $20 a month and a content editor at $10, on both Basic and Pro. This site had only ever quoted the $20 figure. Not a change, just something nobody here had looked at, and it halves the cost of the partner test for anyone who only needs someone editing copy.
Four of five vendors held every price for three weeks
Re-verified against each vendor's own pricing page today: Wix $17 / $29 / $39 / $159, Squarespace $19 / $29 / $49 / $99, Hostinger Premium $2.99 on 48 months renewing at $10.99, Durable Launch $22 yearly and Grow $41 yearly. All unchanged since 2026-09-22. Recorded because a log that only appears when something moves cannot tell you whether a vendor is stable or whether nobody checked.
September 28, 2026
The free tier is a 300-credit monthly budget, not an unmetered allowance
Netlify's free plan runs on credits: 300 a month, where a production deploy costs 15 and bandwidth costs 20 per gigabyte. Whether this replaced an older unmetered free tier or had been in place for some time is not established, because this was the first reading taken here. Flagged as a first observation rather than a change for that reason.
September 22, 2026
Premium teaser rose 50 percent, from $1.99 to $2.99 a month
The four-year prepay went from $95.52 to $143.52. The renewal rate held at $10.99, so the gap between teaser and renewal narrowed from about 5.5x to about 3.7x. This is the clearest vendor move in the log: dated readings on both sides, same plan, same term. It also broke a sentence this site had published on three pages, that four years of hosting, domain and email cost under $100. At $143.52 it no longer does, and those pages were rewritten rather than quietly patched.
The second tier was renamed and repriced, Business $2.99 to Unlimited $3.99
Renewal on that tier held at $16.99. Worth noting next to the row above: when a host moves its promotional pricing it tends to move the whole ladder, so a quote for one plan ages the others too.
Grow fell from $85 to $41 a month on yearly billing
Month-to-month went from $99 to $49. A large cut, and the direction is not in doubt, but the earlier $85 figure came from a less rigorous reading than the one that replaced it, so this is logged as likely rather than proven.
The Launch plan gained a free custom domain
Worth a line because it changes the arithmetic rather than the sticker: Durable's true-cost figure here dropped by a domain renewal a year. The free plan still cannot publish to a custom domain.
Extra editor seats became a flat $20, from a $20 to $40 range
The old range depended on plan. It is now one number on both Basic and Pro, which makes the partner test cheaper and easier to quote.
Basic allows 2 CMS collections, not 1
A smaller error than today's but the same kind: a limit read once and never re-checked.
What three sweeps actually show
Across three sweeps the vendor that has actually moved is Hostinger, twice, both times on promotional pricing rather than renewal pricing. Durable moved once and in the customer's favour. Wix and Squarespace have not moved a published figure since this site started checking. Framer moved its seat pricing once; everything else attributed to Framer in this log was my error, not theirs.
The pattern worth taking away is not about any one vendor. It is that promotional pricing moves and renewal pricing mostly does not. Hostinger's teaser went up 50 percent while its renewal rate sat still. That is the opposite of how these are usually quoted: every comparison table leads with the promo, which is the number most likely to be stale, and buries the renewal, which is the number you will actually live on. Price the renewal.
The uncomfortable half of this page
Three of the ten entries below are my own errors. That is a high proportion and I am not going to dress it up. The October entry is the bad one: I read a limit off a plan comparison table, applied it to a tier that table does not describe, published it across nine pages, and let it reverse the recommendation on the fan-sites guide for three weeks.
What I would say in my defence is only this: it was caught by a scheduled re-check rather than by a reader, and the fix is on the page rather than in a quiet edit. The failure mode I am trying to avoid is not being wrong. It is being wrong and stale and unfalsifiable at the same time, which is the normal condition of pricing content on the web. A dated log makes the first survivable and the other two impossible.
Method, so you can judge the entries
Prices come from the vendor's own pricing page, read directly, not from a comparison site. Where a vendor's page states the same limit differently in two places, as Framer's does, both are read and the discrepancy is the finding. Where a page renders in a non-US currency or with tax included, which Wix's did in September, the figure is not used until a second independent source agrees on the US number. Nothing in the tables on this site is carried forward from a previous sweep without being re-read.
Everything above is checkable against the vendor in about five minutes, which is the point. If you find something here that is wrong, the contact form reaches me and the correction goes in this log with your finding credited.
The current prices, with the catch stated on every entry.
Open the stack directoryKeep reading
The stack directory: current prices, catch stated on eachWhat a website really costs in 2026Every builder review