Website replatforming: when it's worth it, what it costs and how to choose

The short answer

Website replatforming means moving a site to a different CMS or platform, such as WordPress to Webflow, or Webflow to Astro with a headless CMS. It is worth doing when the current platform forces the decision, or when staying costs more than moving. The forcing reasons have dates: Webflow sites in Freelancer and Agency Workspaces move to its 2026 prices at renewals from 16 November 2026 (Webflow Help), Drupal 10 reaches end of life on 9 December 2026 (Drupal.org), and Sitecore 10.0 and 10.1 lose extended support on 31 December 2026 (Sitecore).

Licence savings alone rarely pay for a move. In the example below, the yearly software bill differs by a few hundred dollars from one platform to another, while published migration packages start in the thousands. The business case usually lives elsewhere: developer hours spent on routine edits, a site that fails Core Web Vitals, software that no longer gets security fixes, or a feature the platform can't do.

If you decide to move, keep URLs wherever you can, change one thing at a time and test every redirect before launch. Google advises planning changes "one after the other, not everything at the same time" (Google Search Central). Our website migration SEO checklist covers the execution. This guide covers the decision.

Replatforming, redesign, rebuild or migration?

The words get used loosely, and the difference changes both the budget and the risk.

A move between site builders is always a rebuild of the templates. Framer says it "does not offer HTML export for self-hosting" (Framer Help), Wix says a site "must run on Wix's servers" (Wix Help), and Webflow's code export leaves out CMS content, forms and site search (Webflow Help). Content can travel; layouts and features get built again.

Replatforming and redesign are often sold as one project. Google's site-move guide recommends making changes one after the other and warns to "expect temporary fluctuation in site ranking during the move" (Google Search Central). With two changes at once, you can't tell which one caused a drop. A like-for-like move first and a redesign second is the safer order; our website redesign checklist covers the second phase.

Signals that justify replatforming

Each signal below has a fix that doesn't require moving. Replatform when the fix in place costs more than the move, or doesn't exist.

A licence, plan or support date you can't avoid

Vendors change prices and retire versions on their own schedule. These dates were in force when we checked on 7 November 2026:

A date forces a decision, not necessarily a replatform. Drupal 10 sites can upgrade to Drupal 11, and Sitecore sites to 10.4 or 10.5, whose extended support runs to the end of 2030 and 2032 (Sitecore). Price both paths: the upgrade and the move. For Webflow, the company's FAQ adds an option: a site billed monthly that switches to yearly billing before the change reaches it locks in its current plan for one more year (Webflow Help). That buys a year to decide calmly. Our Webflow pricing 2026 guide and Webflow renewal calculator price the change for your site.

Software that no longer gets security fixes

PHP 8.1 and every older version are past end of life, and PHP 8.2 gets security fixes only until 31 December 2026 (php.net). php.net warns that users of an end-of-life release "may be exposed to unpatched security vulnerabilities". On 7 November 2026, WordPress.org's statistics showed 32.4% of the WordPress sites that report to it running PHP 8.1 or older, and another 22.6% on PHP 8.2 (WordPress.org). WordPress recommends PHP 8.3 or newer (WordPress requirements).

Plugins are the larger exposure. Patchstack counted 11,334 new vulnerabilities in the WordPress ecosystem in 2025: 91% in plugins, 9% in themes and 6 in WordPress core. For 46% of them, the developer had no fix ready by the time they were disclosed (Patchstack).

Updating PHP and pruning plugins is maintenance, not a replatform. It becomes one when the theme or a plugin the site depends on blocks the upgrade, or when keeping the stack patched takes a developer every month. Website maintenance covers the first case.

A site that fails Core Web Vitals

Google's field data from real Chrome visitors, published monthly by the HTTP Archive, shows how sites on each platform fare. The latest month available on 7 November 2026 was September 2026. Share of US sites passing Core Web Vitals on mobile:

These are shares of all sites on each platform, not a forecast for yours. Small brochure sites dominate the site builders, large script-heavy sites dominate the enterprise platforms, and Next.js passing at 51.6% shows that a modern headless stack isn't fast by default. Check your own field data in Search Console or PageSpeed Insights first. If the failures come from images, fonts or third-party scripts, speed optimization on the current platform costs less than a move. Our platform speed report shows the same data by country.

Editors who can't publish without a developer

The most expensive signal rarely shows up on an invoice: a marketing team that waits days for a developer to change a page, can't preview a draft, or keeps translations in spreadsheets. Log those hours for a month before deciding. Then check the destination's limits, because a hard ceiling is a signal too:

Integrations and features the platform fights

Logged-in areas, product data from another system, a CRM that needs more than a form post, search across thousands of items: when every new requirement turns into a workaround or another third-party script, the platform has become the constraint. Choose the destination for what the site has to do in two years, not for what it does today. If the requirement is really an application, see web app development or client portals.

What doesn't justify replatforming: a dated look (that's a redesign), one slow page, a single plugin conflict, or a new team member's preference for another tool.

The business case: the cost of staying vs the cost of moving

Put both sides on one page over the same period, usually three years.

Then: payback in months = one-off cost of moving ÷ (yearly cost of staying − yearly cost after moving) × 12. If the bottom of that fraction is close to zero, the move has to be justified by something other than money: security, a support date or a feature you need.

What the software costs a year

An example: a 40-page marketing site with a 300-post blog, three people editing and one extra language. List prices read on the vendors' pages on 7 November 2026, in US dollars, billed yearly:

The widest gap in the table, Framer at USD 840 a year against Astro with Sanity's free plan at USD 60, is USD 780 a year. Published migration packages cost more than that: on Contra, a Webflow partner's WordPress-to-Webflow package starts at USD 2,500 for two weeks (Contra), and on Dribbble an agency's migration to Webflow starts at USD 7,500 for four weeks (Dribbble). At USD 780 a year, a USD 7,500 move takes almost ten years to repay on licences alone. Our website migration cost guide has published prices by platform and site size.

Licence-driven cases do exist at the top of the market, where enterprise licences are quoted rather than listed. Ask for the renewal quote in writing before anyone scopes a migration; that number, not a list price, belongs in the model.

Risks and how to contain them

The URL list is where most moves go wrong. When we moved the music school Gissi to a new domain in October 2026 (a domain move, not a replatform), the sitemap held 2,144 pages but 4,379 old addresses needed a redirect; Search Console, older redirects and the site's own files supplied the rest (how we did it). Our free website migration planner reads your sitemap before the new site exists, groups the URLs into patterns and marks which keep their path on the platform you choose. Once both sites exist, the redirect map generator matches old URLs to new ones.

Accessibility deserves its own line because a replatform rebuilds every template. WebAIM's 2026 scan of the top million home pages found detectable WCAG 2 failures on 95.9% of them, with an average of 56.1 errors per page (WebAIM). Most of those errors live in components, so a rebuild is the cheapest moment to fix them. Check the templates against our WCAG 2.2 checklist.

Choosing the destination

Start from who edits the site and what it has to do, then pick the platform. Dardo builds on Astro and Cloudflare by default, and on Webflow, Framer, WordPress or Next.js when a team or a feature needs them, so read our view with that in mind.

Leaving Webflow specifically? Webflow alternatives prices each option and what leaving takes. Our migration pages cover each pair we build: WordPress to Webflow, WordPress to Astro, WordPress to Framer, Webflow to Astro, Webflow to Framer, Framer to Astro, Wix to Webflow and Squarespace to Webflow. For the CMS itself, we compare the options for SaaS marketing sites in SaaS CMS. Larger organizations with approvals, SSO and many sites should also read our enterprise CMS comparison.

Replatforming readiness checklist

Our website brief and RFP builder puts most of these on one document you can send to studios.

How long replatforming takes

There's no public benchmark, only what providers publish:

• Two weeks for a small site: a Webflow partner's WordPress-to-Webflow package on Contra, from USD 2,500.
• Two to four weeks: an agency's migration-to-Webflow package on Dribbble, listed at four weeks, from USD 7,500.
• Three to six months: what Pantheon, a hosting vendor, says a replatforming project "commonly takes", with larger enterprise programs taking six months or longer. It cites no data for the range.

Search engines need their own time after launch. Google says "a small to medium-sized website can take a few weeks for most pages to move", larger sites take longer, and redirects should stay in place for at least a year (Google Search Central). Content and approvals, not code, set most schedules; our guide to how long it takes to build a website explains why. If the move comes with a redesign, add the phases in our website redesign timeline.

What to do next

Run your sitemap through the website migration planner to see how much of the site keeps its URLs on each platform. If you'd like Dardo to take the move on, website migration covers the inventory, the redirect map, the rebuild and the reviews after launch. We quote each project in a written proposal after seeing the URL inventory, as our pricing page explains. If the move also needs a new design or a new content model, see website redesign and headless CMS development.