Loading…
Online · AI assistant
Ask about services, timelines or how we work, or describe your project. Type, or tap the mic and just speak.
Popular questions
AI can make mistakes; our team confirms the details. Privacy · Call us
A redesign is the most reliable way to lose organic traffic overnight — and the loss is almost always preventable with work done before launch, not after.
Organic traffic survives a redesign when every old URL either stays the same or permanently redirects to the closest equivalent page, and when the content that earned the rankings is carried across rather than shortened. Most redesign traffic losses trace to missing redirects or content cut for visual reasons.
Rankings do not attach to your website in the abstract. They attach to specific URLs carrying specific content and specific inbound links. A redesign that changes all three simultaneously and quietly is not a redesign from Google's perspective — it is a new site with no history.
The most common failure is the URL change nobody flagged. A new site structure renames pages sensibly, old addresses return a 404, and every ranking and inbound link pointing at them evaporates. This is invisible during the build because nobody is testing old URLs against the new site, and it is fully visible three weeks after launch.
The second failure is subtler and harder to argue against: content removed for design reasons. A page that ranked because it answered a question thoroughly gets rewritten to fit a cleaner layout, loses two-thirds of its text, and stops being the best answer available. The URL is intact, the redirect is unnecessary, and the ranking still goes — because the thing that earned it is gone.
The third is a staging site that was blocked from crawling and shipped with the block still in place. It is rarer, it is catastrophic, and it takes an alarmingly long time to notice if nobody is watching Search Console.
Almost all of the protective work happens before the new site goes live. Once traffic has dropped, you are recovering rather than preserving, and recovery takes months.
Export every URL that currently exists, with its title and status. Pull the last twelve months from Search Console and Analytics so you know which pages actually earn traffic and which merely exist. You cannot protect rankings you never measured, and this export is also the only reliable way to prove what changed afterwards.
Ranked pages, pages with inbound links, and pages with meaningful traffic. These are the ones that constrain the redesign — everything else is free to change. In practice this is usually a much shorter list than expected, which makes the constraint easier to accept.
Every old URL mapped to its new equivalent, agreed before development finishes. Map to the closest genuine match — redirecting everything to the homepage is treated as a soft 404 and preserves nothing. Where no equivalent exists, decide deliberately whether to recreate the page or let it go.
If a page ranks because it answers something thoroughly, the new version must answer it at least as thoroughly. Redesign the presentation freely; do not quietly cut the substance. This is the step most likely to be lost in a disagreement about visual density, so settle it early.
Run these within an hour of the new site going live. Each one catches a failure that is trivial to fix immediately and expensive to discover in a fortnight.
Check robots.txt is not blocking everything and that no stray noindex survived from staging. This is the single most damaging thing that can ship, and it takes thirty seconds to rule out.
Take the top fifty URLs from your export and request each one. They should return a single permanent redirect to a working page — not a chain, not a temporary redirect, not a 404. Chains dilute the signal and are easy to introduce accidentally.
Each page should declare itself canonical unless it is deliberately a duplicate. A template that hardcodes the homepage as canonical across the whole site is a known and quietly devastating bug.
Update and resubmit in Search Console to accelerate discovery. Keep the old sitemap available briefly so Google re-crawls the old URLs and finds the redirects rather than inferring their disappearance.
Tracking codes are routinely lost in a rebuild. If you find out a month later, you have a permanent hole in the data exactly where you most need to compare before and after.
A small dip in the first two to four weeks is normal even on a well-executed migration. Google has to re-crawl and re-assess every page, and that process is not instant. The question is whether it recovers on the expected curve.
Watch Search Console rather than rankings. Coverage tells you whether the new URLs are being indexed and whether anything is erroring. Impressions on your priority pages tell you whether Google still considers them relevant. Both move before rankings do, which makes them the early warning you actually want.
If traffic has not returned to its previous level within about six weeks, treat it as a defect rather than as patience. Re-crawl the site, compare against your pre-launch export, and look specifically for URLs that ranked before and now return a 404 or redirect somewhere unhelpful. In our experience the cause is nearly always in that comparison, and it is nearly always a redirect that was missed rather than anything about the new design.
Keep the redirects permanently. There is no point at which removing them becomes safe — old links on other people's sites do not expire, and they keep sending both visitors and ranking signal for years.
What clients ask before committing to a rebuild.
If they rank and the structure is not actively causing problems, yes — the safest migration changes nothing it does not need to. Restructure when there is a real reason, and accept the redirect work as part of that decision rather than as an afterthought.
A permanent redirect to a genuinely equivalent page passes essentially all of it. Value is lost when the target is not equivalent — redirecting a detailed service page to a generic homepage is treated as a soft 404 and passes little. Relevance of the target is what matters.
Yes, and it is lower risk. Migrating section by section limits the blast radius of any mistake and makes the cause obvious when something moves. It takes longer and is worth it on sites where organic traffic is commercially important.
Usually, if the cause is redirects or indexation. Find the ranked URLs that now 404, redirect them properly, and rankings typically return over several weeks. Recovery is slower where content was cut, because that requires rebuilding the substance rather than restoring a mapping.
Working on this for your own business? See how we run SEO in Delhi — scope, timelines and what we report each month.
We do this work for a living — technical SEO, performance and builds that hold their rankings. Tell us what you're dealing with.
Prefer to talk? +91 87960 72717
What happens next