Short answer: Webflow is a good product for a design-led team whose site is mostly pages and who want to change layout themselves without waiting on a developer. It gets expensive and constraining as the team grows: seats, CMS items, localisation and bandwidth are all metered, the code export leaves your CMS content behind, and anything beyond embeds and a form handler becomes a workaround. WordPress, built properly on a custom block framework rather than a theme, wins for content-heavy sites, deep MarTech integration (Marketo, HubSpot, 6sense, GTM, OneTrust, Greenhouse), ownership and portability, and the size of the talent pool; the price is a developer and a care plan. There is a third option almost no comparison mentions: a flat-file build, which for a brochure or campaign site that has to be fast, cheap to run and readable by AI agents is often the right answer. We build the second and third and not the first, so read the Webflow column knowing that. We have tried to be fair to it.

What are you actually buying?

Webflow is a hosted visual builder with a CMS attached. You design in the browser, publish to Webflow's hosting, and the platform handles servers, CDN, SSL and updates. Nothing to patch. The 2026 version has moved hard toward marketers and AI: Webflow AI is on by default across all Workspaces (Webflow updates, January 2026), Webflow AEO (answer-engine analytics plus agents that ship approved changes) is available to Enterprise customers (Webflow updates), the CMS was re-architected and every site migrated by April 2026 (CMSWire), and Webflow Conf in September 2026 previewed Source, an "agentic workspace" for marketers and AI agents, plus Campaigns with HubSpot and Salesforce integration (CMSWire).

WordPress is open-source software you host wherever you like. It runs 39.4 percent of all websites and 58.1 percent of sites with a known CMS; Webflow is at 0.8 and 1.2 percent (W3Techs, November 2026). That share is why every MarTech vendor ships a WordPress plugin first. What WordPress does not give you is a finished product: a purchased theme with a page-builder plugin and forty others is a different thing from a custom block framework with a locked-down editor, and most complaints about WordPress are complaints about the former.

Flat-file (Grav) is the one comparisons leave out. Pages are Markdown files with YAML settings in a git repository; there is no database. Our builds add a block system and a visual builder, so editors click into a heading and type, reorder sections and add them from a library. Since Grav 2.1 every page is also served as Markdown by appending .md to the URL or sending Accept: text/markdown, "built for AI agents and other text clients" (Grav 2.1.0 release notes). Hosting is a small PHP server. We covered WordPress versus flat-file on its own in an earlier post; here it is the third column.

What does Webflow cost once the team is real?

Webflow simplified its plans in May 2026 and the headline numbers are modest. From the pricing page: Basic is $15 a month billed yearly (300 static pages, 10 GB bandwidth, no CMS); Premium is $25 a month billed yearly (20,000 CMS items, 40 collections, bandwidth tiers to 2.5 TB), or $39 month to month (Flow Ninja, updated October 2026).

The site plan is the small line. What scales with a marketing team is:

  • Seats. A Workspace plan (Core $19, Growth $49 a month, billed yearly) includes one full seat. Each extra full seat is $39 a month and each limited seat (content and page building with components) is $15, billed yearly.
  • Localisation. Localize Essential is $9 a month per locale for up to three locales and 10,000 machine-translated words a month; Advanced is $29 per locale for up to ten locales.
  • The Team plan. Publishing workflows, single-page publishing, an activity log and API, and enhanced security sit on the new Team plan at $2,500 a month on an annual contract, with five full and five limited seats, two locales and 100,000 AI credits a year. Enterprise is custom.
  • AI credits. Webflow AI and AEO agents draw on a credit pool; 2,000 credits a month is $20.

None of this is hidden. The point is that Webflow prices the people and the markets, not just the site. For three people and one English site it is cheap; for twelve people and four regional sites it is not, and the month you need workflows or SSO the bill moves from tens of dollars to $30,000 a year.

Custom WordPress is priced the other way round: a larger one-time build, then hosting and a care plan. WP Engine's Essential plan is $28 a month billed annually and Business is $400 (WP Engine); Kinsta's single-site plans start at $35 (Kinsta). Seats, locales and editors cost nothing extra. The cost that does not go away is the developer who owns updates, security and the block framework.

A flat-file site has neither per-seat pricing nor a managed-WordPress bill. DigitalOcean's smallest Droplet is $4 a month (DigitalOcean) and a brochure site does not need more. The care plan is lighter because there is no database and little to patch.

Who changes the layout?

This question genuinely favours Webflow. A designer who knows the Designer can rebuild a landing page on a Tuesday afternoon with no ticket. If your team has that person and the site is mostly pages, that autonomy is worth real money.

The counterpoint is that most marketing teams want to publish a page that looks like the rest of the site in ten minutes, not unlimited layout freedom. A well-built WordPress block framework gives editors a constrained set of on-brand blocks, and the site stays consistent across sixty pages and three years. Our flat-file builds work the same way. In both, a genuinely new layout is developer work. In Webflow it is designer work: faster if you have one designer, a governance problem if you have five.

Can you leave? Ownership and export

Webflow lets you export code on a paid Workspace plan, but its own documentation says CMS, User Accounts and Ecommerce content, code components and functionality, and localised pages, elements and content are not included (Webflow Help). You get static HTML, CSS, JavaScript and images for the primary locale; collection lists export empty, forms stop working because they depend on Webflow's backend, and interactions on Webflow's runtime may need rework (Flow Ninja). Leaving Webflow is a rebuild, and Webflow's pricing is set by a company that knows it.

With WordPress you own the code, the database and the host, and you can change agency or host without asking anyone. With a flat-file site the content is already Markdown in git, the most portable format there is, and the whole site is a folder.

Integration depth: forms, gated content and the MarTech stack

The B2B integration list is predictable: marketing automation (Marketo, HubSpot, Pardot), intent or ABM (6sense, Demandbase), GTM, a consent manager (OneTrust, Cookiebot), a careers feed (Greenhouse, Lever), a scheduler and chat. All of them embed on all three platforms. The difference is what happens when "embed" is not enough.

On Webflow, forms post to Webflow's handler and are forwarded by integration, or you drop in the vendor's embed; gated content means a hidden page, a Webflow Cloud app or a third-party tool; server-side logic lives outside the platform. On custom WordPress, forms are server-side, so a Marketo submission can be validated, enriched, posted to the API, mirrored to the CRM and gated behind a progressive-profiling rule in code you own; Greenhouse jobs arrive through the API as real pages with structured data. On flat-file, embeds work and forms post server-side to vendor APIs; the question to ask early is what "gated" means. A form before a PDF is fine. Accounts and personalised content are WordPress, or an application layer.

Performance and security, with the numbers

Field data favours the hosted builders, and we will say so. Search Engine Journal's reading of HTTP Archive's Core Web Vitals Technology Report in May 2026 had WordPress last among major platforms at roughly 49 percent of origins passing all three Core Web Vitals, behind Drupal at about 64, Wix at about 80 and Duda at about 85 (Search Engine Journal); the Web Almanac's 2025 CMS chapter put WordPress at 45 percent (HTTP Archive). Figures circulated from the same May 2026 CrUX data put Webflow at about 69 percent on mobile, though the only copies we could open were on Webflow agency blogs (Webtonic), so treat that one as directional.

Those are platform averages, and the WordPress average includes millions of cheaply hosted theme-and-plugin builds. A custom WordPress build on good hosting passes Core Web Vitals; we have not shipped one that does not. A flat-file site has less to get wrong: no database on the request path, no plugin stack. The build makes the site fast, not the platform. The builders make it harder to build a slow one, a fair point for teams without a developer.

Security is the mirror image. Webflow removes patching entirely, a real benefit for a team with no technical owner. WordPress has the largest attack surface because it has the largest install base, almost all of it in plugins and unpatched installs; a lean custom build with a care plan has a small, well-understood surface. A flat-file site has no database to inject into and no SQL credentials to leak.

The AI-readability question

AI assistants now read marketing sites to decide whether to mention you, and most do not run JavaScript. Vercel's analysis of crawler traffic across its network found that "none of the major AI crawlers currently render JavaScript," including GPTBot and ClaudeBot, which fetched JavaScript files on 11.5 and 23.8 percent of requests but did not execute them (Vercel).

All three platforms serve server-rendered HTML, so none is disqualified. On Webflow, check content behind tabs, sliders and "load more" on collection lists. WordPress can serve Markdown, llms.txt and structured data with plugins and a developer to set them up. A flat-file site on our kit does it by default: every page has a .md twin, llms.txt indexes them, and an MCP server lets agents read content under the admin's permissions. Webflow AEO measures how often answer engines cite you and recommends fixes, which is useful, but it is Enterprise-only, and measuring visibility is not the same as serving pages agents can read cleanly.

The comparison table

Webflow Custom WordPress Flat-file (Grav)
Cost to build Lowest if you have a Webflow designer; agency builds priced like any design project Highest: design, block framework, integrations Lower than WordPress: block system and builder exist; work is design and content
Cost to run Site plan from $25/mo yearly; plus seats ($39 full, $15 limited), locales ($9 to $29 each), bandwidth blocks; Team at $2,500/mo Managed hosting from about $28 to $35/mo for a small site, up to hundreds; plus a care plan A small server (from $4/mo) and a light care plan
Who edits layout Designers, in the Designer, no developer Editors within on-brand blocks; new layouts are developer work Editors within blocks, inline editing; new layouts are developer work
Content modelling Collections, reference fields, 20,000 items on Premium Custom post types, taxonomies, relationships, unlimited Pages, collections and taxonomies in files; no query engine
MarTech integration Embeds, native HubSpot/Salesforce via Campaigns, APIs; server-side logic lives elsewhere Deepest: server-side forms, APIs, plugins for every vendor Embeds and server-side forms to vendor APIs; no plugin ecosystem
Forms and gated content Webflow forms; gating via hidden pages, Cloud apps or third parties Native, with progressive profiling, CRM sync and real access control Forms yes; simple gating yes; accounts no
Localisation Add-on, metered per locale and per site Plugin licence, unlimited locales Native multilanguage, no per-locale cost
Performance Strong platform averages; hosted CDN Average is poor; a good build passes CWV easily Fastest to get right; no database on the request path
Security surface Smallest for you: Webflow patches everything Largest install base; manageable with a lean build and a care plan No database; PHP, web server and admin login
Ownership and export Export excludes CMS content, localised content and backend features; paid Workspace required You own code, data and hosting Content is Markdown in git; the site is a folder
AI and agent readability Server-rendered pages; AEO analytics on Enterprise Server-rendered; Markdown and llms.txt with plugins Every page served as Markdown, llms.txt, MCP server, by default
Talent pool Growing, concentrated in agencies Largest of any platform Small; any PHP or front-end developer learns it fast
Best for Design-led teams, page-heavy sites, no developer on call Content-heavy sites, deep MarTech, many editors, long-lived corporate sites Brochure and campaign sites that must be fast, cheap to run and AI-readable

Pick Webflow if, pick WordPress if, pick flat-file if

Pick Webflow if you have a designer who lives in it, the site is mostly pages with a blog, you run one or two languages, embeds and a form forwarder cover your integrations, and you would rather pay a predictable monthly bill than own anything. Budget for seats and check where the Team plan threshold falls before you commit.

Pick WordPress if the site is content-heavy (hundreds of pages, a resource library, many authors), you need Marketo or HubSpot forms with logic rather than embeds, you run several languages, you want to own the code and move freely between agencies and hosts, or procurement wants a platform it can hire for anywhere. Insist on a custom block framework, not a theme, and price in a care plan from day one.

Pick flat-file if the site is a brochure, campaign or professional-services site of up to a few hundred pages, a handful of people edit it, nobody logs in except editors, and leadership cares about speed, security, hosting cost and whether AI assistants can read the site. It is the cheapest of the three to run and the easiest to leave.

The questions to ask before you sign

  1. How many people need full access versus content-only access, and what does that cost per year?
  2. How many languages in three years, and how is each one priced?
  3. What exactly comes out in an export, and could another agency rebuild the site from it?
  4. Which integrations need server-side logic rather than an embed, and where will that logic live?
  5. What does gated content mean to us: a form before a PDF, or accounts?
  6. Who owns updates and security, and what does that cost per month?
  7. What does the site look like to a crawler that does not run JavaScript?
  8. At what point do we hit a plan threshold (workflows, SSO, CMS items, bandwidth), and what does the next tier cost?

What to do next

We build custom WordPress and flat-file sites and we do not build Webflow, so if the answers above point to Webflow we will say so and you can take that to a Webflow partner. If they point to WordPress or flat-file, send us the brief and the current site and we will come back with both quotes and the one we would pick for your team. Book a call.