Short version. If one or two people look after a marketing site of a few dozen pages and a blog, and it does not need to plug into anything unusual, Webflow is normally the better buy. If the site is the business — memberships, bookings, a CRM sync, thousands of pages, or five people editing at once — WordPress is normally the better buy.
Neither is a bad tool. The expensive mistake is building in the one that suits year one and having to rebuild in the other by year three. So the question is not "which is better", it is "which of these constraints will I hit, and when".
Design control: Webflow wins, and it is not close
Webflow gives a designer direct access to the box model. Flexbox, grid, breakpoints, states, custom classes, scroll-linked interactions — all of it visual, all of it producing reasonably clean HTML and CSS. If the Figma file has an asymmetric hero and a sticky sidebar that changes behaviour at 1024px, a Webflow designer builds it without opening a ticket with a developer.
WordPress can match any of that, but only by writing it. A block theme with custom blocks, or a headless front end, will do anything Webflow does. What you cannot do is get that control out of a page builder without paying for it in weight. Elementor and Divi sites routinely ship several hundred kilobytes of CSS and JavaScript to render three sections, and that shows up in Core Web Vitals. Put what you have through our website speed test before believing anyone about which platform is faster: what was built matters more than what it was built in.
One catch with Webflow's design freedom that nobody raises at the demo: styling is class-based and global. A client who edits a class called "heading-2" because it looked wrong on one page has just changed it on ninety. That is training, not a bug, but it is a real support cost.
The CMS is where the decision actually gets made
Webflow's CMS is genuinely pleasant for structured content: collections, typed fields, reference and multi-reference fields, collection pages generated from one template. For a blog, a team directory or a case study library it is better designed than most WordPress setups you will inherit.
Then you meet the ceilings. Check the current figures on Webflow's plan page; the shape has been stable for years:
- Item caps per site. On the order of 2,000 CMS items on the CMS plan and 10,000 on Business. Plenty — until you build 300 city pages across 8 services and every combination is an item.
- Reference depth. You can bind a field from a referenced item. You generally cannot walk a reference of a reference. Any content model that assumes you can needs flattening, which means duplicated data and a sync job to keep it honest.
- No server-side logic. No querying, no business rules at request time. Filtering and faceted search happen client-side or through an add-on such as Finsweet Attributes or Jetboost, and those degrade as item counts rise.
- Redirect limits. Migrating an old site with thousands of legacy URLs can exceed what Webflow will store, and you end up handling 301s at a proxy layer.
WordPress has none of those caps in any practical sense. Custom post types, taxonomies, ACF field groups, WP_Query, a REST or GraphQL layer if you want one. A 40,000-post site on decent hosting is an ordinary Tuesday. The price is that somebody has to model it properly; a badly modelled WordPress install is worse to live with than a tidy Webflow one.
If the plan involves programmatic pages — service by city, product by application — that is a WordPress or custom development conversation, not a Webflow one. It is also the moment to involve whoever handles your local SEO, because the content model and the ranking strategy are one decision made twice.
Hosting: one of these you can move, one you cannot
Webflow hosting is not optional. You can export a Webflow site's static HTML, CSS and JavaScript, but the export does not include the CMS. Anything bound to a collection stops being dynamic the moment it leaves Webflow, and on Enterprise plans code export is often switched off entirely.
The honest description: you can leave Webflow, but you leave the CMS behind and rebuild it. For a 12-page brochure site that is an annoyance. For a 900-item content library it is a project with its own budget.
WordPress is a folder of files and a MySQL database. Move it with a migration plugin, or with rsync and mysqldump. You are locked into nothing but your own past choices — and a page builder that stores content as shortcodes is a lock-in of its own.
The flip side is real, and Webflow earns the credit for it: nothing to patch, no plugin conflicts, no PHP version bump breaking a form, no compromised install quietly serving spam links. The maintenance bill for a Webflow site is that there mostly is not one.
Plugins: an ecosystem versus building it yourself
The WordPress plugin ecosystem is its biggest advantage and its biggest liability in the same breath. Whatever you need — Gravity Forms for conditional forms, FacetWP for faceted search, WooCommerce for a store, WPML for translation, WP Rocket for caching — exists, is mature, and costs roughly $50 to $300 a year each.
The liability is that every plugin is third-party code with database access, and every one is an update that can break a Tuesday. A 35-plugin site is not a website, it is a small integration project you now own. Our rule on any WordPress build is that a plugin has to earn its place: if it does one small thing, we write the forty lines instead.
Webflow's app ecosystem is much smaller: native forms, the API, and a Make or Zapier scenario for the rest. For most marketing sites that is enough. For transactional depth it is not — Webflow Ecommerce handles a simple catalogue and simple tax, then stops being sensible around variant matrices, subscriptions, trade pricing or ERP sync. That is a WooCommerce or Shopify conversation, which is what our e-commerce work usually turns out to be.
What a handover actually looks like
WordPress handover is a database dump, the files or a Git repository, hosting credentials, an admin account, and the bit everyone forgets — plugin licence keys reassigned to the client's own email. If those licences stay on the agency's account, the client's plugins stop receiving security updates the day the relationship ends.
Webflow handover is a site transfer into the client's own Workspace, after which their card pays the site plan. But editing anything structural needs a paid Workspace seat, roughly $19 to $49 per seat per month depending on plan, and on most plans content-only editors need a seat too. Five people who need access at $49 a seat is $245 a month — $2,940 a year — before the site plan itself.
When a Webflow site starts costing real money
Webflow is cheap to run, right up until a specific point where it is not. The tipping points, in the order we tend to see them:
- Editor count. Past three or four people who need access, per-seat pricing overtakes what a managed WordPress host charges for unlimited users.
- Item count. Near the CMS cap you either upgrade the plan or start deleting content. Programmatic location and product pages get you there sooner than a blog will.
- Integrations. The first "can it talk to our booking system" is fine. By the fourth you have a stack of Make scenarios on per-operation billing, and that stack is now load-bearing.
- Anything needing logic. Gated content, member areas, quote calculators pricing against live data. Each workaround is a script pasted into a page's custom code, and custom code in Webflow has no tests and no staging worth the name.
When two or more of those are true, the annual cost of staying usually beats the one-off cost of moving — and you pay for the migration anyway, because the CMS does not travel with you.
What each one costs
Bands, not quotes. What moves them: page count, bespoke design versus a template, how many CMS templates are needed, integrations, volume of content migration, and whether copy and photography are in scope.
| Item | Webflow | WordPress |
|---|---|---|
| Build — brochure site, 5–15 pages | $2,500–$9,000 | $2,500–$12,000 |
| Build — CMS-driven site | $8,000–$30,000 | $10,000–$60,000+ |
| Platform and hosting per year | $170–$470, plus seats | $60–$1,200 |
| Licences per year | $0–$400 (apps) | $150–$900 (plugins) |
| Maintenance per month | Low — nothing to patch | $75–$400 care plan |
Our own work starts from $200 as a minimum and moves with scope; our pricing page sets out how that is put together. Anyone quoting a fixed number before seeing your content inventory is guessing.
The cheap option, and when it is right
If the site is five pages, one editor, a contact form and no roadmap, both of these are overkill. Squarespace or Wix will handle it for the price of a coffee a week, and the money saved is better spent on the ads or the SEO that make the phone ring.
Rough rules that hold up in practice:
- Marketing site, small team, design-led, no unusual integrations — Webflow.
- Content at scale, membership, e-commerce beyond the basics, or an in-house team who already know WordPress — WordPress.
- The site is a product rather than a brochure — neither; a framework front end with a headless CMS.
- What you have is dated or slow but not structurally wrong — rebuild the front end, keep the platform.
If you are weighing this up with a real content inventory in front of you, it is a twenty-minute conversation rather than a proposal. Book a call, or read how we handle web design and development, and bring the list of what the site must do in three years. That list decides the platform, not the demo.
