Website strategy, redesign and migration: the complete guide

A good business website lets a visitor understand what you do, trust you and take the next step quickly. In practice that means clear pages for each service, fast loading on a phone, honest proof such as reviews and case details, and an obvious way to contact you. Google’s documentation says its ranking systems prioritise relevant, helpful content, and that good page experience helps mainly when many useful pages compete for the same query. So content and structure come first, and speed, security and mobile layout support them. Design polish matters, but only after those basics work.

There is no fixed interval. Redesign when evidence shows a problem that small changes cannot fix, not because a certain number of years has passed. Nielsen Norman Group advises making the smallest change that solves the problem and considering a major overhaul only when incremental fixes hit diminishing returns, the technology cannot support key needs, or the structure is incoherent. Many sites that look dated still perform well, and many new-looking sites convert poorly. Our guide on [when to redesign a website](/resources/learn/websites/when-to-redesign/) gives a decision framework you can apply with your own data.

Most redesigns take from a couple of months for a small site to six months or more for a large one. The length depends on the number of page types, how ready the content is, how many people approve work, and how complex the build is. Content and approvals cause more delay than design or code. Treat any timeline as a range with stated assumptions, not a promise. The [redesign process guide](/resources/learn/websites/website-redesign-process/) breaks the work into stages and explains what drives each range, so you can check a proposed schedule against reality.

It can, but careful planning limits the risk. Google’s site move documentation says to expect ranking fluctuations for weeks or longer after a move, depending on site size. Losses usually come from avoidable errors: changed URLs without redirects, pages removed, metadata lost, or a staging block left on at launch. Mapping every old URL to a new one, using permanent redirects and monitoring afterwards protects most of the value. No one can promise rankings stay identical. Read [redesigning without losing SEO](/resources/learn/websites/seo-safe-redesign/) for the full sequence.

Improve step by step when the site mostly works and your data points to specific problems, such as one weak page or a slow template. Redesign when the cause is structural: an outdated platform, a tangled structure, a changed brand or a site that fails on mobile everywhere. Step-by-step changes are cheaper, carry less risk and show which change helped. A redesign makes sense when many small fixes would cost more than starting over. Often the honest answer is a mix, where you fix the worst pages first and redesign only what evidence supports.