Change platform
without losing your rankings.

Changing platform is a project where success is invisible and failure very visible. If all goes well, nobody notices anything: visitors keep arriving, pages keep ranking, enquiries keep coming. If the redirection plan is incomplete, you lose in one night what the site took years to build.

That is why we treat a migration as a continuity project before treating it as a design project. The new site will be prettier and faster, that is the easy part. The part that demands rigour is the correspondence between the old and the new.

We migrate to Webflow when it is justified, and we say so when it is not.

It is the only deliverable of a migration whose absence is paid for immediately and lastingly. We deal with it first, before the design.

Why a lost address is a lost asset

Every page of your current site has a history: it may be cited elsewhere, it may rank for a search, it may be bookmarked by a client. When it disappears without a redirect, all of that disappears with it, and nothing warns you.

The classic symptom arrives three weeks after launch: traffic has fallen, nobody knows why, and it turns out that forty addresses return an error. At that stage, repair is possible but the ground lost takes months to come back.

How we draw it up

• We survey every existing address, not only those in the menu. A ten-year-old site always contains more than its owner imagines.
• We cross-reference with the pages that actually receive traffic and those that receive links from outside. That cross-reference sets the priorities.
• Every old address gets a destination, one by one. Redirecting everything to the home page is the false good idea par excellence: it avoids the visible error and loses the value anyway.
• We test the plan before the switch, then check it again afterwards. An untested plan is always discovered too late.

The cases that need a decision

A page that no longer exists on the new site because the offering has gone. A duplicated page that was competing with another for the same search. A blog category emptied of its content. These cases are not handled mechanically: they are decided, and we put them to you rather than settling them alone.

The second classic mistake is to transfer everything. A migration is the least expensive opportunity to do the sorting you have been putting off for years.

What we find on a site over five years old

Pages describing an abandoned offering. Blog articles published for the sake of publishing, with no readers. Duplicates created by accident, depriving each other in the results. And pages nobody has ever read, sometimes since they were created.

Transferring all of that amounts to importing the mess into a new site, with a measurable effect: quality is assessed at the scale of the site, and weak pages weigh on the good ones.

Our sorting method

Three categories, decided with you on real usage data and not on attachment. What works is carried over and improved. What has a history but no up-to-date content is rewritten or merged, keeping the address. What has neither traffic nor links nor use is removed, with a redirect to the nearest page.

Content that does not export cleanly

Formatting done by hand in an editor, tables, attached documents, forms and their recipients. Those are the items that lengthen a migration, and they are surveyed during framing rather than discovered along the way.

So we cost a migration on the number of templates and on the state of the content, not on the number of pages. Three hundred pages built on five templates cost less than forty pages that are all different.

We also deliver WordPress and Shopify, which lets us answer without a commercial agenda.

What Webflow gains

Native performance, which does not require stacking optimisation plugins. Independence for the team, which publishes without depending on a supplier once the templates are in place. And no upkeep of a foundation: no updates to apply, no plugins ageing, no known vulnerabilities left open.

That last point is the most underestimated. It moves a recurring workload into a subscription, which is not free but becomes predictable.

What it loses

The plugin ecosystem, when your need genuinely depends on it. Flexibility on very particular content structures. And reversibility: leaving Webflow means rebuilding, whereas a self-hosted site moves more easily.

The three situations where we advise against migrating

• A site that works, that the team knows how to edit, and that nobody complains about. Migrating would be a comfort expense.
• A need that rests on a business plugin with no equivalent, typically complex booking or a client area.
• A substantial retail catalogue, where other platforms are better suited. We have set out that decision in Shopify or PrestaShop.

The full comparison between the two platforms most concerned by this work is in Webflow or WordPress, and the closest alternative in Framer or Webflow.

A migration does not end at launch. That is in fact the moment when the watching has to be tightest.

Before the switch

The new site is checked on a separate environment: every template in its states, the forms right through to the message actually being received, the display on a real phone, and the redirection plan tested address by address.

We also prepare the rollback. Knowing exactly how to return to the old site within minutes changes the nature of switch-over day: you decide instead of enduring.

On the day itself

The switch happens at a quiet time, never on a Friday evening. We then check, in order: the site responds, the redirects work, the forms arrive, the measurement tracking records, and the indexing directives file has not kept the block from the test environment.

That last point is the most frequent and most expensive error in the whole trade: a new site put live while still closed to search engines, and discovered weeks later.

The following two weeks

We watch the addresses returning errors, the rankings on the searches that matter, and the volume of enquiries received. A successful migration produces none of those three signals. A migration needing correction produces them within days, and that is precisely why we watch.

The upkeep that follows then belongs to website maintenance, and the groundwork on visibility to the SEO and performance page.

They recur from one project to the next, whatever the platform you start from. None is technical in the difficult sense: they are oversights, and they are prevented by a list.

1. The indexing block left in place

The new site is built on a test environment, closed to search engines so it is not indexed before its time. On switch-over day, that block has to be lifted. When it is not, the new site stays invisible and nobody notices for several weeks. It is the most expensive error in the trade, and the most mundane.

2. Forms that arrive nowhere

A form that displays and sends is not a form that works. You have to check that the message actually arrives, with the right person, and that it does not end up in junk. A silent form produces no error: it produces an absence of enquiries, blamed on the economy for weeks.

3. Images carried over as they are

Migrating by copying the original visuals also transfers their weight. You end up with a new site slower than the old one, which is the opposite of the point. Images are reworked to the real display format and size, during the migration, not afterwards.

4. Measurement tracking forgotten

The measurement tools and the Search Console property are reconnected, and the gap is documented. Without that, comparing before and after becomes impossible, and you can no longer demonstrate that the migration went well, nor detect that it went badly.

5. Addresses with and without a trailing slash

Platforms do not treat that detail the same way. One page reachable at two addresses makes search engines work twice on the same content. The rule is chosen once and applied everywhere, redirects included.

6. Content approved after launch

The most frequent organisational trap. Switching over with provisional text on the assumption it will be fixed afterwards amounts to publishing an unfinished site, which search engines will discover in that state. We would rather push the switch back a week than put live content that is not the right content.

The checklist, shared with you

Those six points, plus the checks particular to the project, form a list we go through together before the switch. It is written, it is ticked off, and it is what lets you decide to switch or to postpone on facts rather than on a feeling.

Two migration quotes cannot be compared without these three variables, and none of them is the number of pages.

The number of templates

It is the main variable. A site of three hundred pages built on five models costs less than a site of forty pages that are all different. It is also the simplest lever for adjusting a budget without degrading the result: reduce the number of models, not the quality of each.

The state of the content

Content that is clean, structured and exportable is carried over quickly. Content formatted by hand in an editor, with tables and attached documents, needs manual work that bears no relation to the number of pages.

The features to rebuild

A simple form is remade in an hour. A client area, a booking system or a connection to a management tool are projects in themselves. We survey them during framing and cost them separately, so that you see the decision being made.

What is never included

The platform subscription, which is in your name. Writing new content, which is a separate trade. And buying visuals or a photo shoot, which are decided independently of the technical work.