Technical SEO that survives a redesign

The pattern is familiar. A company relaunches its site, the design is better in every way, and eight weeks later organic traffic is down 40%. The new site isn't worse. It's just a different site as far as Google is concerned, and the equity the old one earned didn't carry over.

The things that must not move

• URLs. If a URL has to change, it gets a 301 to its closest equivalent — not to the homepage. A redirect map is part of the launch, not a follow-up ticket.
• Primary content. The H1, the body copy, the internal links that make a page rank — those survive the redesign even if the layout around them is new.
• Structured data. Schema that was driving rich results gets ported, tested in the Rich Results tool, and checked again after launch.
• Crawl surface. The sitemap, robots rules and pagination behaviour stay coherent. A new framework's default routing can silently orphan pages.

Instrument before you touch anything

Before the first component is built, export a full snapshot: every indexed URL, its impressions and clicks, its top queries, its Core Web Vitals. That snapshot is the acceptance test. After launch you compare against it page by page, and anything that regressed gets fixed while the change is still fresh.

On the Knwdle engagement we wired Search Console sync into the platform itself, so ranking movement is a dashboard the team watches — not a report someone remembers to pull once a quarter.

The redesign is a chance, not just a risk

A rebuild is the cheapest time to fix crawl waste, collapse thin pages, and get the site fast. Done with the search surface in hand, a redesign should end with rankings flat-to-up on week one and climbing after — not a recovery project.