The website builders worth paying for in 2026

Every round-up of website builders is a ranked list, and every ranked list is a small lie. The winner is always the one that does the most things, which means the recommendation at the top is almost never the right answer for the person reading it. A consultant who needs one page that looks considered gets pointed at a platform built for marketing teams with a content calendar, spends a weekend inside it, and never publishes.

So this one is not ranked. We built the same site on each platform — a one-person consultancy: an intro, a short services list, four pieces of work, a handful of writing, and a way to get in touch — and then noted the exact step where each build turned annoying. That step is more informative than any feature table, because it is the thing you will hit in week two rather than in the demo.

What came out of it is that these seven products are not competing. They sit in about five different categories, and once you know which category you are shopping in, most of the list disappears before you have compared a single feature.

Two questions remove most of the list

The first is how many pages the site genuinely needs. Not how many you can imagine wanting — how many a visitor needs in order to decide about you. For a large share of personal and freelance sites the honest answer is one, and if it is one, half of this comparison is irrelevant and expensive. We have written separately about what a one-page site can and cannot carry; the short version is that one page has the useful property of being finished, and multi-page sites mostly die of the pages nobody filled in.

The second is whether anything on the site has to change on a schedule after launch. A services page that gets edited twice a year is static content and needs no CMS. A blog you intend to actually write, a case-study library, a job board, a shop — that is a content model, and a content model is the single most expensive line item in this entire field. On Webflow it is the difference between the $15 plan and the $25 one. On Ghost-style publishing tools it is the whole product. On the lightweight tools it does not exist.

Answer those two and you have usually eliminated four or five of the seven below. Then the feature comparison is between two products instead of seven, and it takes ten minutes instead of a weekend.

Wix is the freest canvas, and the hardest one to change your mind on

Wix was the fastest build of the six general builders, and the least fussy about placement. You drag an element and it stays where you put it, which sounds trivial until you have spent an hour fighting a system that keeps re-flowing your layout into its idea of correct. For anyone who has a picture in their head and wants it on screen without learning a layout model first, this is the least frustrating tool in the comparison.

The painful step came later, and it is documented in Wix's own support material rather than in any review: "It's not possible to switch to a different template for a site you already created." We picked a template, built out five sections, decided the template was wrong for the tone we wanted, and discovered that changing it means starting again. Every other builder here lets you regret a design decision. Wix makes you commit to one before you have seen your own content in it, which is exactly the moment you know least.

Pricing: Free $0, Light $17, Core $29, Business $39, Business Elite $159 — all annual, paid in full for the year. Wix does not publish monthly rates outside the logged-in checkout, so if you want to pay month to month you will not know the price until you are partway in. Light includes a free domain for the first year, 2 GB of storage, unmetered bandwidth, two collaborators and no Wix banner. The trap in that sentence is the domain, and Wix states it plainly: monthly plans do not include the vouchers, so the free domain only exists if you pay a year up front.

The free tier is permanent, needs no card, and gives you siteprefix.wixsite.com/siteaddress with 500 MB of storage and 1 GB of bandwidth. The Wix banner and the Wix favicon are compulsory on it. As a place to try the editor it is genuinely useful. As a place to send a client, it is not.

Wix is the answer if you have no design training, want to place things yourself, and want business tools — bookings, invoices, a shop — in the same account. It stops being the answer the moment you suspect your first design instinct might be wrong.

Squarespace protects you from yourself, then charges for the escape hatch

The Squarespace build looked better than the Wix build, and we did less to make that happen. That is the entire product. The system constrains type scale, spacing and colour to combinations that hold together, and the result is that a person with no design background produces something that does not look like a person with no design background produced it. Nothing else here does that as reliably. We go through the platform in detail in our full Squarespace review.

Two things hurt. The first is that there is no free plan — Squarespace says so outright. You get a 14-day trial with no card, the site stays private, and after the trial the content is, in their words, marked for permanent deletion. Building carefully in that window and then getting pulled onto something else for three weeks means starting over. Blueprint AI, their generator, is free to use, but publishing anything requires a paid account.

The second is the one that actually caught us. We wanted a small piece of custom CSS to fix the spacing on one section, and code injection starts at Core, at $29 per month billed annually. On Basic, at $19, there is no route to it. That is a $120 difference over the year for one CSS rule, and it is the kind of thing you discover after you have committed.

Basic gets you a free domain for the first year, unmetered bandwidth, up to 1,000 pages, 30 minutes of video storage, two contributors and no Squarespace branding. Squarespace does not publish general file storage in gigabytes. Monthly billing is available on every tier and costs meaningfully more: $25 against $19 on Basic, $39 against $29 on Core.

Squarespace is right for design-conscious people who want a consistently good result more than they want control. It is wrong if you already know exactly what you want, because the same constraints that save an amateur will bill an expert $29 a month for permission to override them.

Webflow's real price is on four axes at once

Webflow is the most capable tool here and the hardest to budget for, and those two facts are related. The build was the slowest of the six, which is not a criticism — you are working with the actual box model, and the reason designers use it is that it does not invent a simplified layout language and then trap you inside it.

The painful step was administrative. Webflow bills on two separate axes that people routinely confuse: a site plan per site, and a workspace plan per workspace, with seats on top of that and add-ons on top of those. Site plans run Starter $0, Basic $15 annual or $25 monthly, Premium $25 or $39 — Premium is new and replaces the old CMS and Business tiers — then Team at $2,500 per month on an annual contract. E-commerce is a third scale again. Workspace plans run Starter $0, Core $19 or $28, Growth $49 or $60, with Freelancer at $16/$24 and Agency at $35/$42. Seats are Full $39/$45 and Limited $15/$19. Add-ons include Optimize from $299/$379, Analyze from $9/$12, Localize from $9/$12 and AI credits at $20/$25.

Then the detail that matters most for the site we were building: Basic at $15 gives you a custom domain, a removable badge, 300 static pages and 10 GB of bandwidth — and no CMS. The CMS starts at Premium. Our four project pages were fine as static pages. The writing section was not, and that single requirement moved the site from $180 to $300 for the year.

The free Starter plan is a preview rather than a plan: mysite.webflow.io, two static pages, 1 GB of bandwidth, 50 form submissions, badge permanently attached. Webflow's AI Site Builder generates up to five pages in the creation flow as of the 5 February 2026 update, with animations and eight editable theme categories, but it does not change the billing picture at all.

Trustpilot shows 1.4 out of 5 across 232 reviews. That is a small, self-selecting sample and we would not build a verdict on it, but the complaints cluster around billing rather than the product, which is consistent with what the pricing structure looks like from the outside.

Webflow earns its cost for professional design and marketing teams with real CMS, governance and localisation needs. For one person and one site it is an expensive way to buy capabilities that will sit unused. If you are weighing it against the other design-led option, we put them head to head in Webflow versus Framer.

Framer is priced per site, which is only fine until there are three

Framer produced the best-looking build in the least time of any of the design-led tools, and the editing model — where the thing you are designing is the thing that ships — is the most pleasant part of the whole comparison. The AI layer, Framer Agents, builds pages on the canvas from a description and leaves the output editable rather than locked, which is more than most generators manage.

The pricing is where it turns. Free $0, Basic $10 annual or $15 monthly, Pro $30 or $45, Enterprise custom — and every one of those is per site. Extra editors are $20 per month each. For our one site that is $120 a year and entirely reasonable. For the freelancer who ends up hosting a side project and a client's landing page in the same account, it is three subscriptions, and nothing in the interface warns you that this is how it works.

Basic includes a custom domain, a free .com but only on annual billing, 30 pages, two CMS collections holding 1,000 items, 50 GB of bandwidth per month and 1,000 AI credits. The badge disappears on any paid plan or as soon as a custom domain is attached. Storage as a separate figure is not published. The free tier gives you 500 one-off credits, 1 GB of bandwidth, a Framer subdomain and a visible badge.

The published ceilings are high but they exist: 700 pages, 40,000 CMS items, 2 TB, 10 seats. Trustpilot sits at 1.6 out of 5 from 129 reviews, again mostly about support, refunds and unexpected charges — small sample, treat as colour rather than verdict, but the pattern of complaint again points at billing.

Framer is for solo designers and small teams who care how the site moves and are building one thing at a time. It stops being the cheap option the moment you are building several.

Carrd is the cheapest honest answer, and the tier it advertises is not the one you need

Carrd took under an hour, and the result was a perfectly respectable one-page site. There is no CMS, no page tree, no shop, no ambition beyond the one thing, and the price reflects that: annual billing only, Free $0 for three sites, Pro Lite $9 per year for three, Pro Standard $19 per year for ten, Pro Plus $49 per year for twenty-five.

The painful step was the checkout. Pro Lite is the number everyone quotes, and Pro Lite has no custom domain — Carrd's own documentation describes it as being for users who do not use custom domains. If you want the site on your own name, you are on Pro Standard at $19. That is still the lowest price in this comparison by a wide margin, but the advertised entry price is not the price of the thing most people are trying to buy.

Branding is removed from Pro Lite upward. Uploads cap at 16 MB for images and GIFs and 64 MB for video. Total storage and visitor limits are not published. The free tier gives you three sites with a "Made with Carrd" footer, an element limit, and no forms, widgets, analytics or custom domain. There is no AI feature anywhere in the product or the documentation, which in 2026 reads almost as a stance.

Carrd is right when the site is one page, the budget is small, and you already know what the page should say. It is wrong when you need the thing to grow, and it is wrong when the blank editor is the obstacle, because Carrd hands you exactly that.

WordPress is the only one you can still be running in ten years

Self-hosted WordPress was the longest build and the only one that produced something we could pick up and move somewhere else. Nothing here is close on that point. There is no vendor badge and no vendor at all — the software is free under the GPL, and the cost is hosting plus a domain. If the site has to outlive a company's product strategy, this is the list of one.

The painful step is the renewal cliff, and it is the most consistently under-reported number in this entire market:

Host	Intro	Renewal
Hostinger Premium (48 months up front)	$2.99/mo	$10.99/mo
IONOS Grow (1 year)	$1/mo	$12/mo
SiteGround StartUp (12 months up front)	$2.99/mo	$17.99/mo
Bluehost Starter (36 months)	$3.99/mo	$9.99/mo
WP Engine Startup (annual)	$30/mo, 25,000 visits	not published
Kinsta Single 35k	$35 monthly, $30 annual	same — no intro pricing

IONOS Grow is the teaching example: one dollar a month for the first year, twelve times that afterwards. Kinsta is the only host in the set without the mechanic at all, and it is priced accordingly. Entry tiers at the four budget and mid-range hosts give 10–25 GB of storage and a free domain for the first year. We unpick this pattern across the whole market in how website builder pricing actually works.

The second cost is maintenance, and it has a hard number attached. Patchstack's State of WordPress Security in 2026 counted 11,334 new vulnerabilities across the ecosystem during 2025, up 42% year on year: 91% in plugins, 9% in themes, and six in core, all low risk. New holes are typically exploited within about five hours. Read that split carefully, because it is the actual lesson — the core is not the problem, what you bolt onto it is. A WordPress site with four plugins and an update habit is not a security project. A WordPress site with thirty plugins and no update habit is somebody else's server.

Installation is not the cost anyone claims it is; every host has a one-click installer. The cost is theme choice, plugin restraint and update discipline, month after month, for as long as the site exists.

reach removes the step people actually fail at

Every builder above hands you an empty canvas and a template gallery, and that is where most personal sites die — not at the design stage, but at the "I'll do it this weekend" stage, repeated for a year. On a general builder the first real decision is which of forty templates to commit to, and the second is what to write in a hero section. Those two decisions are the entire difficulty of a personal website.

reach is built for one job and removes both. You upload a CV and a photo, answer a short form, choose a colour and a layout family, and it composes a finished one-page site: hero, CV timeline, projects, values, contact and a few others, assembled from vetted layout fragments rather than generated as markup, which is why the output does not break. The page is generated in about twenty seconds and you can be live in under two minutes on a free yourname.joinreach.app subdomain. The first decision you make is whether the draft in front of you is right, which is a different and much easier decision than starting from nothing. The editor is a real one — inline text, drag, colour, fonts, image replacement, section add and rebuild — and it stores styles per viewport, so a font change on desktop does not leak onto mobile.

The limits are hard, and they are not softened by any of that. It makes exactly one page — one index.html, no sub-pages, no navigation, no site tree. There is no CMS, so nothing grows. There is no e-commerce of any kind. The contact section is a mail link and profile links, not a form that posts anywhere. There are no visitor analytics, no team accounts, no HTML export and no custom code field, so you cannot take the markup elsewhere and you cannot paste your own CSS in. There is no version history; saving overwrites the page. The editor needs a screen wider than 820 pixels — below that it is a read-only preview. And you cannot connect a domain you already own: the only route to a custom domain is buying one through reach, which is an outright blocker if your name is already sitting at a registrar.

Free gives you 40 credits; Premium is $4.99 per month or $49 per year and is required for a custom domain. Regenerating a section costs a flat 5 credits, while adding, rebuilding and editing sections cost nothing. reach sells domains as a name.com reseller with a $3 margin on the registrar price, registered in your name, one year at a time, auto-renew off by default. The subdomain publishes in seconds; a custom domain takes minutes to hours while DNS and the certificate settle.

The year-one number, with the domain in it

Headline monthly prices are not comparable across this field, because some include a domain for the first year and some do not, and because several tiers only exist if you pay twelve months up front. Here is the same site, on a custom domain, for the first twelve months:

Platform	Plan used	Year one	Domain in year one
Carrd	Pro Standard	$19	bought separately
reach	Premium	$49	bought through reach, registrar price plus $3
Framer	Basic, per site	$120	free .com, annual billing only
Webflow	Basic (no CMS)	$180	no free-domain offer advertised
Webflow	Premium (with CMS)	$300	no free-domain offer advertised
Wix	Light	$204	included, annual payment only
Squarespace	Basic	$228	included first year
WordPress	Hostinger Premium, 48 months up front	$36	included first year
WordPress	Kinsta Single 35k, annual	$360	bought separately

Prices checked August 2026. Two things fall out of that table. The cheapest WordPress row is only cheap because you paid for four years in one go, and year five costs $131.88 at the renewal rate — the intro price is a loan against your future self. And the spread between the cheapest and the most expensive way to put one person on one domain is roughly nineteen-fold, for outputs that a visitor would not be able to tell apart.

What we would actually pick

For a company site with a services tree, a blog somebody maintains, a shop, or a lead form wired into a CRM: Webflow if there is a designer in the building and a budget that can absorb four billing axes, WordPress if the site has to outlive whoever built it and someone will own the updates. Those are the only two answers we would give for that shape of site, and the choice between them is really a choice about who does maintenance.

For a small business run by one or two people who want bookings and invoices in the same place as the website: Wix, on Light or Core, with the caveat that you should look at the templates for longer than feels necessary, because that decision is final. If the result mattering more than the control is the priority, Squarespace Basic, and budget for Core if you ever expect to need a line of CSS.

For a designer or a small studio building one site at a time and caring how it moves: Framer. Watch the per-site billing before you host anything else in the account.

For one person who needs one page: it depends on which problem you actually have. If the obstacle is money and you know what the page should say, Carrd Pro Standard at $19 a year is the cheapest defensible answer on the market. If the obstacle is momentum — if the page has been on the list for a year because starting is the hard part — reach is the tool that removes the starting step, and you should check first that you are not blocked by needing a domain you already own.

The failure mode we watched most often in this exercise was not choosing badly. It was choosing a platform sized for the site someone imagined building, then never building the much smaller site they actually needed.