UI/UX Design · Redesign & Replatform · SMB

Website Redesign Without Losing Your Rankings

Most redesigns that lose traffic lose it for avoidable reasons — unmapped URLs, deleted content that was ranking, and a launch nobody measured. We redesign around what already works, and prove it afterwards against your own numbers.

Benchmarks verified 29 July 2026 Redirect map before launch, not after Staged, reversible launches Fixed quote after a free scoping call
52%Of mobile sites fail Core Web Vitals — only 48% passed in 2025 (HTTP Archive)
≤2.5sGoogle's "good" LCP threshold, measured on your real visitors
75%Share of real page loads that must hit "good" for a URL to pass (Google)
2015Redesigning and rescuing SMB websites since
Start here

Four situations, four different projects

"Redesign" covers work ranging from a two-week visual refresh to a full platform migration. Find yours — the scope, risk and price are not the same.

Looks dated, but works
Typically 3–5 weeks

Same pages, same URLs, same CMS — new visual system, better typography and layout, faster loading. The lowest-risk option because nothing structural moves.

Details below ↓See scope
Needs rebuilding, but you can't afford to lose search traffic
Typically 6–10 weeks

New structure, new templates, new content — with a full URL inventory, redirect map and content-parity check done before launch, and rankings monitored after it.

Details below ↓See scope
Changing CMS, domain or hosting
Typically 6–12 weeks

Moving to WordPress, Shopify, BigCommerce or a headless build. Every URL changes, which is exactly why the redirect strategy is designed first and tested on staging.

Details below ↓See scope
Traffic already dropped after a relaunch
Diagnosis in 1–2 weeks

Someone relaunched and organic traffic fell. We find what broke — missing redirects, removed content, blocked crawling, changed titles — and fix it in priority order.

Details below ↓See scope

Not sure a redesign is what you need? Often it isn't. If the site's structure is sound and only conversion is weak, conversion rate optimization or landing page optimization costs a fraction of a rebuild and carries none of the SEO risk. We'll tell you that on the call rather than after the invoice.

The analytical part

Why redesigns lose traffic — and how that's avoided

Five failure patterns account for nearly every "we redesigned and our traffic collapsed" story. None of them are mysterious.

01

Unmapped URLs are the single biggest cause

A redesign that changes URL structure without a complete old-to-new mapping serves 404s to every page Google had indexed and every external site that linked to you. Google's own site-move documentation is explicit about the sequence: build the URL mapping first, update internal links to the new URLs, save a sitemap of the new URLs alongside a list of sites linking to your old ones, then plan the redirect strategy, then test the redirects before going live.

That order matters. Redirects designed after launch are archaeology — you're reconstructing what used to exist from memory and crawl logs, while the clock runs.

The trap: redirecting everything that moved to the homepage. It resolves the 404 and destroys the relevance signal. Each URL should point to its closest real equivalent, and pages with no equivalent need a decision, not a catch-all.

02

Short-term fluctuation is normal; a permanent drop is a mistake

Google states plainly that with any significant change to a site you may see ranking fluctuations while it recrawls and reindexes, and that as a general rule a medium-sized website takes a few weeks for most pages to move in the index — larger sites longer. Submitting a sitemap speeds up discovery.

So a dip in week one is expected and is not evidence of failure. A dip still there in month three is evidence of something broken, and by then the diagnosis is harder. This is why we agree in advance what normal looks like and what triggers investigation — and why we prefer to launch during a recurring traffic dip in your own analytics, which is also Google's advice, so fewer people are affected and more server capacity is free for crawling.

Hosting changes have their own normal, too. Google notes that a temporary drop in Googlebot's crawl rate immediately after a hosting move is expected, followed by a steady climb over the following days — sometimes to a higher rate than before. Knowing which dips are routine is most of what stops a team panicking into changes that make things worse.

Worth knowing: moving a site in sections rather than all at once is acceptable to Google and reduces blast radius. For large sites we usually recommend it.

03

The redesign that "looks cleaner" often deletes what was ranking

Modern design taste favours less text. Search visibility frequently depends on that text. The pattern is predictable: a 1,400-word service page that answered twelve real customer questions becomes a 200-word page with a big photograph, and the rankings that page held go with the words. Nobody notices for two months because the site looks so much better.

We run a content-parity check before launch: every page with existing organic traffic or backlinks gets compared old-to-new, and anything substantive that's being cut has to be a deliberate decision rather than a casualty of the layout. Where the old copy was genuinely poor, our copywriting and content teams rewrite it to be better and complete, not shorter.

04

Speed is the bar you're measured against, and most sites don't clear it

HTTP Archive's Web Almanac found that in 2025 only 48% of mobile websites and 56% of desktop sites had good Core Web Vitals — meaning more than half the mobile web fails. The trend is improving (mobile was 36% in 2023 and 44% in 2024), but the bar is real: largest contentful paint within 2.5 seconds, interaction to next paint under 200 milliseconds, minimal layout shift, assessed on the 75th percentile of your actual visitors rather than a lab test on a fast laptop.

A redesign is the cheapest moment to fix this, because you're rebuilding templates anyway. It's also the easiest moment to make it worse — heavy hero video, a page builder generating bloated markup, four new tracking scripts. We build against the thresholds on mid-range mobile hardware and treat a regression as a launch blocker.

05

Without a baseline, nobody can tell whether it worked

Redesign success is usually argued about rather than measured, because the numbers from before the launch were never recorded properly. We take a baseline first: organic sessions and rankings by page, conversion rate, Core Web Vitals field data, top landing pages and their traffic, and a full crawl of the existing site. That crawl doubles as the URL inventory the redirect map is built from, so the measurement work and the safety work are the same work.

Analytics and conversion tracking get rebuilt and verified on the new site before launch too. A redesign that silently breaks goal tracking looks like a traffic problem for weeks.

Side by side

Which engagement fits

Scroll horizontally on mobile. Timelines are typical ranges from past projects, not commitments — yours goes in your scope document.

Comparison of D Cloud Solutions website redesign engagements: Design Refresh, SEO-Safe Rebuild, Replatform and Migration, and Post-Redesign Recovery
Design RefreshSEO-Safe RebuildReplatform & MigrationPost-Redesign Recovery
Best forDated look, sound structureSite needs rebuilding, traffic mattersChanging CMS, domain or hostTraffic dropped after a relaunch
Typical timeline3–5 weeks6–10 weeks6–12 weeks1–2 weeks to diagnose
URLs changeNoSomeMost or allAlready did
SEO risk levelLowManagedHighestDamage control
Baseline recorded firstYesYesYesReconstructed
Redirect mapNot neededBuilt & testedBuilt & testedRebuilt
Content parity checkIncludedIncludedIncludedAudited
New CMS or platformOptionalIncluded
Core Web Vitals workIncludedIncludedIncludedAssessed
Post-launch monitoring4 weeks8 weeks12 weeksThrough recovery
PricingFixed quote after scopingFixed quote after scopingFixed quote after auditFixed quote for diagnosis

Every engagement includes the baseline, the pre-launch crawl and the monitoring window — those aren't upsells. Monitoring lengths reflect how long reindexing realistically takes at each risk level.

01 — Lowest risk

Design Refresh

A new look on the same foundations — no URL changes, no structural risk.

Typical timeline3–5 weeks

We keep your pages, URLs and CMS exactly where they are, and rebuild the visual layer: type scale, colour, spacing, imagery, navigation clarity, and the templates that carry your key pages. Speed work comes with it, since a refresh is the natural point to strip accumulated script weight and oversized images. Because nothing structural moves, the search risk is close to zero.

What you get

Designs approved before build, a consistent component set across templates, mobile-first layouts, measurable speed improvement, and a site your team can still edit the same way they do today. Informed by our UI/UX design practice.

What it's not

Not a fix for a bad information architecture. If your navigation is confusing, your pages are in the wrong places, or the site is missing pages your customers need, a refresh makes that better-looking rather than better. We'll say so during scoping and recommend the rebuild instead.

Best for: sites three to five years old that still function, still rank, and simply look their age.

02 — The most common engagement

SEO-Safe Rebuild

A genuinely new site that keeps the search visibility the old one earned.

Typical timeline6–10 weeks

New information architecture, new templates, rewritten content where it needs it — built on a documented URL inventory. Every page with organic traffic or inbound links is accounted for: kept at the same URL, redirected to its closest equivalent, or consciously retired. Titles, meta descriptions, headings, structured data and image alt text carry across. The redirect map is crawl-tested on staging, and we launch during a low-traffic window with rollback ready.

What you get

A findings and architecture document you keep, designs signed off before code, a tested redirect map, verified analytics, and eight weeks of post-launch monitoring for 404s, index coverage and Core Web Vitals regressions. Pairs with SEO services when you want to grow rather than just hold position.

What it's not

Not a guarantee of zero fluctuation — Google says a medium site takes a few weeks to reindex, and no agency controls that. Not a traffic-growth service either: protecting what you have and increasing it are different projects with different budgets, and we price them separately rather than implying one buys the other.

Best for: businesses whose site is a real acquisition channel, where a bad launch would be felt in revenue rather than pride.

03 — Highest risk, handled deliberately

Replatform & Migration

Changing CMS, domain or hosting without handing your rankings to a competitor.

Typical timeline6–12 weeks

Moving to WordPress, Shopify, BigCommerce or a headless stack changes your URL structure by definition, so the redirect strategy is designed at the start and not bolted on. We migrate content, media, products and customer data, rebuild templates on the new platform, then follow Google's documented sequence — mapping, internal-link updates, redirect testing with the URL Inspection tool, and a Change of Address submission in Search Console where the domain itself is moving. For large sites we usually move in sections rather than all at once.

What you get

A platform recommendation with the trade-offs written down, a staged site you review before the DNS switch, a documented redirect map you own, and twelve weeks of monitoring. If you're weighing platforms, our e-commerce design and development page compares five of them honestly.

What it's not

Not lossless. Plugin-generated content, review histories tied to a specific app and some custom fields have no clean equivalent on the destination platform — those need decisions, which we surface in the audit before you sign rather than during the build. Migrating to a new domain also resets some accumulated signals no matter how well it's executed.

Best for: teams spending more on maintaining their current platform than it returns, and businesses outgrowing a builder that can't do what they now need.

04 — Damage control

Post-Redesign Recovery

Your traffic fell after a relaunch. This finds out why, fast.

Diagnosis1–2 weeks

We work through the usual suspects in order of how often they're the culprit: missing or wrong redirects, pages quietly dropped in the rebuild, content cut below the threshold that ranked, a stray noindex or robots.txt rule left over from staging, canonical tags pointing at the old domain, titles and headings rewritten past recognition, broken internal linking, and Core Web Vitals regressions. Search Console's index coverage and crawl data plus a full crawl of both versions tell most of the story within days.

What you get

A ranked findings document — cause, affected pages, estimated traffic impact, fix effort — and a quote for the fixes you want us to make. You keep the document whether or not you hire us for the remediation, and you can hand it to your existing developer.

What it's not

Not a guarantee of full recovery. If pages were deleted and the content is genuinely gone, or a migration moved to a new domain badly months ago, some of the loss may be permanent. We'd rather say that in week one than bill you for six months of hope. Recovery also isn't instant once fixes ship — reindexing takes weeks.

Best for: anyone whose organic traffic dropped after a relaunch, whether we built the site or not. Most of this work is on sites we didn't touch.

What's included

In every redesign we ship

These are baseline, not line items you negotiate back in.

Elements included in every D Cloud Solutions website redesign engagement
AreaWhat it means in practice
Baseline firstOrganic sessions and rankings by page, conversion rate, Core Web Vitals field data and a full crawl recorded before anything changes — so "did it work" is a measurement, not an argument.
URL inventoryEvery existing URL catalogued with its traffic and inbound links, which becomes both the redirect map and the content-parity checklist.
Redirect map, testedOld-to-new mapping built before launch and crawl-tested on staging. Each URL points to its closest real equivalent, never a blanket redirect to the homepage.
Content parityPages with traffic or links compared old-to-new so substantive content is cut deliberately, not accidentally by a layout decision.
Speed to the thresholdsBuilt and tested against LCP ≤2.5s, INP ≤200ms and low CLS on mid-range mobile hardware. A regression against your baseline blocks launch.
Mobile-first designDesigned for the phone first, then scaled up — because that's where the majority of your visitors and Google's assessment live.
Metadata carried acrossTitles, meta descriptions, heading structure, structured data, canonical tags and image alt text migrated rather than regenerated from scratch.
Analytics rebuilt and verifiedTracking and conversion goals reinstalled and tested before launch, so a measurement break can't masquerade as a traffic problem.
Editable by your teamLayouts, banners, copy and images you can change yourself without a developer for everyday updates.
Launch window and rollbackGo-live scheduled during a recurring dip in your own traffic data, with a tested rollback path and a backup taken first.
Monitoring window404s, index coverage, rankings and Core Web Vitals watched for four to twelve weeks depending on engagement, with a written close-out against the baseline.
How we work

The sequence, and why it's in this order

Most redesign damage is done in the first two steps, by skipping them.

  1. Free scoping call. Thirty minutes on the current site, your analytics and what's actually wrong. If the honest answer is that you need optimization rather than a redesign, you hear it here.
  2. Baseline and crawl. Traffic, rankings, conversions and Core Web Vitals recorded; the whole site crawled to produce the URL inventory. Nothing is designed until this exists.
  3. Written scope and fixed quote. What's included, what's explicitly excluded, timeline, price. No work starts from a verbal brief.
  4. Architecture. Page structure, navigation and URL strategy agreed before design — deciding which URLs move, and where each one goes, on paper.
  5. Design. Templates for the pages that matter, reviewed on real devices and signed off before code.
  6. Build on staging. Developed on a password-protected copy, with the redirect map implemented and crawl-tested there, not in production.
  7. Pre-launch checklist. Redirect crawl passes, no stray noindex or staging robots rules, canonicals correct, metadata carried across, analytics firing, speed verified against baseline, backup taken.
  8. Launch in a traffic dip. DNS TTL lowered about a week ahead so the change propagates quickly, then the release switched during your quietest recurring window. Staging crawl blocks removed, sitemap submitted, Change of Address filed if the domain moved. Large sites move in sections.
  9. Monitor and close out. Active watch on 404s, index coverage, rankings and Core Web Vitals through the monitoring window, ending in a written comparison against the step-two baseline.
Transparency

What's verified, and what we won't claim

This page makes specific factual claims. Here's where each one comes from, and where our commitments stop.

Verified on 29 July 2026. The Core Web Vitals pass rates — 48% of mobile and 56% of desktop websites in 2025, up from 44% and 55% in 2024 and 36% and 48% in 2023 — come from HTTP Archive's Web Almanac performance chapter. The thresholds (LCP within 2.5 seconds, INP under 200 milliseconds, low CLS) and the rule that assessment uses the 75th percentile of real page loads come from Google's web.dev documentation. The site-move guidance quoted here — build the mapping first, update internal links, test redirects, submit a Change of Address, expect a medium site to take a few weeks to reindex, move during traffic dips, moving in sections is acceptable — is from Google Search Central's own migration documentation, linked below.

Not verified, and stated as our own. Timelines on the engagement cards are typical ranges from past projects, not commitments; yours goes in your scope document. Monitoring windows are our service commitments. The failure patterns in the section above are our observations from redesign and recovery work, not published research — we've labelled them as analysis rather than dressing them up with invented percentages.

What we will not claim. We do not promise zero ranking fluctuation, because Google itself says fluctuation accompanies any significant site change and no agency controls the recrawl. We do not promise a traffic or conversion increase from a redesign — protecting existing visibility and growing it are different projects, and conflating them is how agencies sell rebuilds that didn't need to happen. We do not guarantee full recovery on a botched relaunch we didn't perform; if some of the loss looks permanent we say so in the first week. And we don't publish redesign prices, because a five-page brochure refresh and a two-thousand-URL replatform aren't the same job — you get a fixed written quote after the free call, with exclusions named so change requests are a conversation rather than a surprise.

When we'll talk you out of it. A redesign is the most expensive answer to a lot of website problems. If your structure is sound and only conversion is weak, CRO or product UX consulting is cheaper and safer. If the site is fine but nobody finds it, the constraint is visibility, not design. You keep the baseline document either way.

Questions

Website redesign FAQs

What comes up most often on scoping calls.

Will a website redesign hurt my SEO?

It can, and whether it does comes down to preparation rather than luck. The damage comes from changing URLs without a complete redirect map, deleting or drastically shortening content that was ranking, leaving staging directives like noindex or a blocking robots.txt rule in place, and losing titles, headings and structured data in the rebuild. Google's own guidance is to build the URL mapping first, update internal links, test the redirects before launch and submit a sitemap to speed discovery. Handled that way, a redesign typically causes a short fluctuation while Google recrawls — Google says a medium-sized site takes a few weeks for most pages to move in the index — rather than a lasting loss. Handled carelessly, the loss can be permanent.

How long will my traffic take to recover after a redesign?

Expect some fluctuation immediately, and expect most of it to resolve as Google recrawls and reindexes. Google's stated general rule is that a medium-sized website takes a few weeks for most pages to move in its index, with larger sites taking longer, and the speed depending on how many URLs you have and how fast your server responds. Submitting an updated sitemap helps discovery. If organic traffic is still meaningfully down after roughly two to three months, that is no longer normal reindexing and something is broken — usually redirects, index coverage, or content that went missing.

How much does a website redesign cost?

It depends on the number of pages and URLs, whether the platform is changing, how much content needs rewriting, and how much custom functionality exists — a five-page brochure refresh and a two-thousand-URL replatform are entirely different projects. We quote after a free scoping call, itemized and fixed-scope, with exclusions written down. Be wary of published flat rates for redesigns: they generally assume a template with your logo dropped in, and they never include the URL inventory, redirect mapping and monitoring that stop the project costing you traffic.

Do my URLs have to change?

No, and where possible they shouldn't. A design refresh keeps every URL exactly as it is, which is why it carries almost no search risk. URLs only need to change when the site's structure is genuinely wrong, when you're moving to a platform that enforces its own patterns — Shopify's product and collection paths, for instance — or when you're changing domain. Even then, we change only what has to change, and every moved URL gets a redirect to its closest real equivalent, tested before launch. "Let's tidy up all the URLs while we're in here" is the single most expensive optional decision in a redesign.

How long does a website redesign take?

Typical ranges from past projects: three to five weeks for a design refresh, six to ten weeks for an SEO-safe rebuild, and six to twelve weeks for a replatform or migration. A recovery diagnosis takes one to two weeks. What actually determines whether those ranges hold is how quickly we get your material — content, images, brand assets, and access to analytics, Search Console, hosting and DNS — plus timely sign-off at each approval point. Projects slip waiting on content and decisions far more often than on development, which is why the scope document names who owns each input and by when.

My traffic dropped after a relaunch. Can you fix it?

Usually a large part of it, and the first step is diagnosis rather than fixes. We crawl both the old and new versions, work through Search Console's index coverage and crawl data, and check the common causes in order: missing or misdirected redirects, pages dropped in the rebuild, content cut too thin, leftover noindex or robots.txt rules from staging, canonicals pointing at the old site, and rewritten titles and headings. Most of that is diagnosable within days. Whether recovery is complete depends on what happened — if content was deleted and is genuinely gone, or a domain move went badly months ago, some loss may be permanent, and we'll tell you that early rather than bill you for optimism.

Can you redesign without changing my CMS?

Yes, and it's often the right call. If your current CMS does what you need and your team knows how to use it, changing it adds cost and risk for no benefit. We redesign on WordPress, Shopify, BigCommerce and other platforms in place, keeping your editing workflow intact. We'd only recommend replatforming when the current system is genuinely blocking you — unsupported, insecure, unable to do something the business now needs, or expensive to maintain relative to what it returns. That's a conversation with reasons attached, not a default recommendation.

Do I actually need a redesign, or just improvements?

Often just improvements, and it's worth establishing which before spending redesign money. If your site is structurally sound and reasonably fast but visitors aren't converting, targeted conversion rate optimization or landing page work costs a fraction of a rebuild and carries none of the SEO risk. If the site is fine but nobody finds it, the constraint is visibility, not design. A redesign genuinely earns its cost when the information architecture is wrong, the platform is holding you back, the site fails on mobile or speed, or the brand has moved on and the site hasn't. We assess which of those applies on the free call.

Should I redesign on a new domain at the same time?

Only if you have a business reason beyond the redesign — a rebrand, a legal name change, consolidating multiple sites. Combining a domain move with a full redesign stacks two risky changes into one launch, which makes diagnosis much harder if traffic falls: you won't know whether the cause was the move or the rebuild. If both genuinely have to happen, we sequence them rather than shipping them together, and a domain move requires its own Change of Address submission in Search Console for the site and each subdomain. Expect a longer settling period than a same-domain redesign.

Find out whether you need a redesign

Thirty minutes on your site and your analytics. We'll tell you what's actually wrong, what it would take to fix, and whether a redesign is the right answer — including when it isn't.

Book a free consultation

Sources