Website Redesign vs Rebuild vs Migration: How to Choose

The short version #

"We need a new website" usually means one of three different projects. A redesign changes visuals and copy. URLs, CMS, and database stay where they are. A rebuild swaps information architecture, CMS, templates, and probably URLs along with them. A migration moves an existing site to a new domain, host, or platform, and the whole point is to keep the SEO equity you already have. The budgets and timelines aren't comparable. Treating all three as the same line item is how a project goes 2x over budget and still ships late. The decision table at the bottom is the version we walk through with clients in kickoff meetings.

A buyer called us last month and opened with "we need a new website." Forty minutes in, the actual scope was: move the English site off a neglected Hong Kong host onto AWS Frankfurt, freshen the homepage colors, and leave everything else alone. That's a migration plus a redesign, not a rebuild. He'd been quoted for a rebuild and was about to spend nearly twice what the work needed.

This article is for that conversation. If you're talking to vendors or scoping an internal "website overhaul," start by being honest about which of the three you actually need.

Three projects #

Redesign #

A redesign updates visuals and copy. Page structure, URLs, CMS, and database stay where they are.

Typical triggers:

• Brand refresh: new logo, new palette, new typography.
• Homepage needs a hero video and a few new case-study tiles.
• Service-page copy is two years stale and the team wants to rewrite it without touching the URL.

The upside is that SEO risk is close to zero. URLs don't change, the sitemap doesn't change, your backlinks still resolve. The downside: if the real problem is a confusing nav, broken mobile, or an eight-second time-to-first-paint, a redesign just gives a broken car a fresh coat of paint. The mechanical issues are still there.

Budget and timeline: 2 to 4 weeks, often deliverable by a small two-person team.

Rebuild #

A rebuild changes information architecture, tech stack, templates, and content all at once. You may swap the CMS (an unmaintained legacy CMS to WordPress or a headless setup), restructure the page hierarchy (split a single "Products" hub into "Solutions" and "Case Studies"), and almost certainly change URLs.

Typical triggers:

• The current site was built by an outsourced shop ten years ago, the original developer is gone, and adding a single product page takes the vendor two weeks.
• Mobile experience is bad enough that you need to rebuild mobile-first.
• The site still has the legacy "News + Products + About" portal layout, but the business now runs on a B2B services funnel (Home, then Solutions, then Case Studies, then Contact).

The biggest risk in a rebuild is SEO. If URLs change and you ship without 301 redirects, an updated sitemap, or refreshed internal links, you can lose 40 to 60 percent of organic traffic in three months. We have a separate piece on this exact failure mode: How to Preserve SEO During a Website Rebuild.

Budget and timeline: 6 to 12 weeks, with content, design, engineering, and SEO all on the project.

Migration #

A migration is moving the site somewhere new. Maybe to a new domain (oldbrand.cn to newbrand.com), a new host (a domestic data center to AWS or DigitalOcean overseas), or a new platform (Wix to WordPress, WordPress to Shopify). Visuals and content stay roughly the same. The whole point is that traffic, backlinks, and indexing survive the move.

Typical triggers:

• The export team is being spun out from the parent company and needs a standalone .com.
• Hosting contract is up, or the team is moving from a domestic host to an overseas CDN-fronted node.
• Acquisition or rebrand forces a domain change.

Migration looks like the smallest of the three because the visible site barely moves. In practice it carries the highest SEO risk, because you're touching the foundation. The 301 map, hreflang, backlink outreach, and Search Console resubmission are all load-bearing. Skip one and traffic suffers. The step-by-step is in Website Content Migration Checklist for Pages, Images, Cases, and Old URLs.

Budget and timeline depend on size. Under 100 pages: two to three weeks. Over 1,000 pages: six to ten weeks, because the URL map needs row-by-row review.

How to decide #

The shortcut we use internally is "ask three questions." It pins about 80% of clients to a clear scope inside half an hour.

Question 1: Are the URLs, domain, or hosting causing problems?

• All fine: you're probably looking at a redesign or a rebuild, not a migration.
• Domain has to change, host has to move, or platform has to switch: there's a migration in scope, and it deserves its own project line.

Question 2: Is the current information architecture and admin still usable?

• Admin works, only the visuals are dated: redesign is enough.
• Admin is broken (vendor gone, CMS too old, every new page needs a developer): rebuild required.
• The architecture itself is wrong (products, services, and case studies tangled together so overseas buyers can't find the entry point): rebuild required.

Question 3: Are the conversion paths broken?

• Forms, WhatsApp, phone all work, inquiries are just slow: this is usually a content and SEO problem, not a website problem. Run the website renovation audit before scoping a project.
• Forms drop submissions, mobile buttons don't tap, no WhatsApp integration: this is a sub-task inside a rebuild.

Answer those three and the right scope is usually obvious: redesign, rebuild, migration, or some combination.

Common combinations #

In real projects, you rarely get one of the three in isolation. The two combinations we see most often:

Redesign plus migration: existing site is mostly fine, but the brand is moving to a new .com for the export business. Resist the urge to "fix the IA while we're at it." Change one variable at a time, or you can't diagnose what broke after launch.

Rebuild plus migration: the old site needs a full rebuild and you're also changing host or platform. This is the highest-stakes combination. Run it in two phases. Stand up the new site on the new domain or host first, prove it works, then 301 the old site over. We've watched too many big-bang switchovers fail.

Redesign masquerading as rebuild: client says they want a rebuild. Forty minutes in, the actual ask is a hero refresh and a new video on the homepage. Quote it as a redesign even if "rebuild" sounds bigger. The client always finds out, and the relationship is harder to recover than the line item.

SEO trade-offs #

The SEO impact is order-of-magnitude different across the three:

If your site pulls more than 5,000 organic visits a month, both rebuild and migration need an SEO owner inside the project team. Not a consultant called in at week six. The worst recovery we've watched: an industrial brand changed every URL from /products/xxx to /solutions/xxx during a rebuild, shipped without 301s, and watched core-keyword rankings drop from position 3 to 47 inside a quarter. Recovery took most of a year.

Budget and timeline #

Medians from projects we've shipped over the last two years. Use as a sanity check, not a quote:

The spread is wide because of page count, native-English review, ERP/CRM integration, and the size of the URL map for migration work. The breakdown is in Website Renovation Cost and Timeline.

Decision table #

Print this. Walk it row by row in your scoping meeting. Most projects clarify themselves inside ten minutes:

FAQ #

If we only swap the design template, do URLs always stay the same? #

Not always. When a WordPress site changes themes, a new theme can ship with different URL slug rules — for example, moving category archives from /category/news/ to /news/. Export the current URL list before you swap, and diff it against the new theme's defaults.

Can we keep the old URLs during a rebuild? #

In most cases yes, but you have to write it into the requirements before kickoff. Engineers will default to "semantically clean" new URLs unless told otherwise. If you want the old URLs preserved, that's an explicit routing-map requirement. Raise it at scoping, not two weeks before launch.

How long after a migration does SEO recover? #

The pattern we see: with a clean 301 map, a resubmitted sitemap, and outreach to your top backlink sources, Google reindexes about 80% of pages inside two to four weeks. Organic traffic recovers to 90%+ of pre-migration levels in three to six months. A small dip in the first two weeks is normal. Don't panic-fix it.

We have a 50-page site. Do we really need the full rebuild process? #

Page count isn't the same as risk. If five of those 50 pages carry most of your organic traffic, those five still need a per-URL 301 map, internal-link refresh, and schema validation. You can compress the process. You can't skip the SEO leg.

Get a diagnosis #

If you're trying to decide between redesign, rebuild, and migration, bring your current site, target markets, and admin status. We'll run the decision table with you in a free initial review under our website rebuild service, and tell you the project type, budget range, timeline, and SEO risks in one pass. That way you don't scope the wrong project and discover it halfway through build.