Website Redesign

A redesign fails in one of three ways: the rankings go, the content gets mangled, or the new site is so awkward to update that within a year it is as stale as the one it replaced. This service is organised entirely around not doing those things.

Start with an inventory, not a moodboard

The first deliverable is a spreadsheet, not a design. Every URL on the current site, what it is for, what traffic it gets and what links to it. Then a decision on each one: keep, merge, or retire. That inventory is what makes the redirect map possible, and the redirect map is what makes the rankings survive.

It is also where most sites discover they have eleven pages competing for the same query and three that nothing links to.

Getting content out of a builder

Builder content is not content — it is markup pretending to be content: nested wrapper divs, inline styles, and text held inside a serialised blob. Migration extracts the actual words, headings, images and links, and rebuilds them as clean semantic HTML that any future editor can work with. Where a section only existed because the builder needed it to, it goes.

Custom content — case studies, products, team members, services — moves into proper custom post types with real fields, so it becomes queryable and reusable instead of retyped on every page that mentions it.

Building alongside, not on top

The new site is built on a staging URL while the live site keeps working. You can open it whenever you like, page by page, and sign it off before a single visitor is affected. Staging is set to discourage indexing so the two never compete for the same query.

Launch day

The cutover is a checklist, run in order: full backup, deploy the theme and plugin, import content, activate, put the redirects in, purge the cache, flush rewrite rules, then re-crawl the whole site to confirm every URL returns 200 or 301 and nothing is orphaned. Search Console gets the new sitemap the same day.

And before any of that, the rollback is tested — not written down as an intention, actually tested. Launch day is not the moment to find out whether the backup restores.

The week after

Rankings wobble briefly after any migration; that is normal, and it settles. What is not normal is a 404 in the top ten landing pages, and that only shows up if someone is watching. Search Console coverage and the server logs get checked through the first week, and anything the data surfaces gets fixed inside the price.

The result you should expect

A site that is dramatically faster, that your team can update without fear, and whose content is finally structured rather than pasted. The Playbook migration is a good example: custom editorial architecture, moved onto the client’s own hosting, with the publishing workflow intact.