Traffic Dropped After a Website Redesign: What Actually Broke

Every agency that has ever shipped a redesign has had this conversation. Traffic falls in the two weeks after launch, somebody checks the update tracker, finds a rollout that happened around the same time, and the diagnosis is settled before it starts.

It is almost always wrong. Redesigns break things mechanically, and the mechanical causes are findable in an afternoon. What follows is the order to check them in, which is roughly the order of how catastrophic and how fast to fix each one is.

Is it the redesign or an algorithm update?

Four signals separate them, and you can read all four in about fifteen minutes.

Signal	Points at the redesign	Points at an update	Measurement (where to look)
Timing	Decline starts within days of launch	Decline starts inside a confirmed rollout window	Drop analyzer, then the date checker
Indexing	Excluded pages jump in the Pages report	Pages stay indexed, positions fall	Search Console, Indexing, Pages
Shape	Specific URLs vanish entirely	Most pages lose a similar share	Page scope analyzer
Crawl	404 and redirect errors rise sharply	Crawl stats look normal	Settings, Crawl stats

The Indexing row is the decisive one. A core update reassesses pages it still has. A broken redesign removes pages from the index entirely, and the Pages report names the reason for each exclusion. That is a diagnosis rather than a hypothesis.

It is also possible to have both, and that combination is what makes some redesign drops so hard to unpick. Fix the mechanical faults first regardless, because they cap everything else.

If you also changed domain, platform or URL structure, that is a migration rather than a redesign and it carries different risks: traffic dropped after a site migration covers redirect mapping at scale and what transfers.

How do I check for a staging noindex tag?

This is the first thing to check because it is the most destructive and the fastest to confirm. A staging site is usually blocked from search. When it is promoted to production, the block sometimes travels with it.

1. Open your live homepage and view source.
2. Search the source for the word noindex. If it appears in a robots meta tag, that is your answer.
3. Repeat on two or three template types, because a CMS can apply it selectively to posts, archives or products.
4. In WordPress, check Settings, Reading, and confirm the discourage search engines box is unticked.
5. Visit yoursite.com/robots.txt and confirm there is no blanket Disallow: / line.
6. Use the URL Inspection tool on a key page and check the indexing allowed status.

If you find one, remove it and request indexing on your most important pages. Recovery follows recrawling, typically one to four weeks depending on site size, and nothing is permanently lost. This is the best possible outcome for a redesign drop.

How do I find URLs that lost their redirects?

The second most common cause, and the one that does permanent damage if left. When a redesign changes URL structure, every old URL needs a redirect to its closest equivalent. Missing redirects throw away the links, rankings and history attached to those addresses.

• Get your old URL list. Why: Export the Pages tab from Search Console for a period before the launch. That is the list of URLs Google knew about and considered worth showing.
• Test each one. Why: Anything returning 404 is a lost page. Anything returning 200 on a generic page is worse, because a soft 404 passes nothing and looks fine in a browser.
• Check for redirect chains. Why: Old URL to intermediate to final wastes crawl budget and dilutes signals. Point every old URL directly at its final destination in one hop.
• Do not redirect everything to the homepage. Why: Google treats mass homepage redirects as soft 404s. A redirect to an irrelevant page passes nothing, so a genuine 404 is more honest and no worse.
• Prioritise by what earned. Why: Sort your old export by clicks and fix the top fifty first. The long tail matters less than the pages that were actually working.

Did the redesign remove content that was ranking?

This is the cause that surprises people, because the site looks better and nobody thinks of it as a loss.

Modern redesigns are usually cleaner, which in practice means shorter. Text that felt cluttered gets cut. FAQ sections get collapsed into accordions or removed. Long explanatory copy below the fold is trimmed because it looked untidy in the mockup. Every one of those decisions removes the specific content that was matching queries.

• Compare word counts on your top twenty pages before and after. Why: Use an archived copy from the Wayback Machine. A page that lost half its text lost half its query coverage.
• Check whether content is in the HTML or loaded by JavaScript. Why: Content that appears only after a click or a script may not be indexed. Use the URL Inspection tool and read the rendered HTML rather than trusting the browser.
• Check whether your H1s survived. Why: Designers replace headings with styled text blocks routinely, and a page with no heading structure is harder to interpret.
• Look for removed FAQ blocks. Why: These often carried a disproportionate share of long-tail impressions and featured snippets.

Did you lose internal links?

The least visible cause on this list and one of the most damaging over time. A redesign changes navigation, and navigation is how authority moves around a site.

A simplified menu, a removed footer link block, a sidebar that no longer exists, or a related-posts module that was dropped can each cut hundreds of internal links. Pages that were three clicks from the homepage become six, or become orphaned entirely, and Google reads that as a statement about their importance.

Check the Links report in Search Console under Internal links. Compare the top-linked pages against what you would expect. A key commercial page that has dropped out of the top of that list has lost its internal support, and that is fixable this week rather than next quarter.

How long should a redesign dip last?

Even a correctly executed redesign usually dips, because Google has to recrawl and reassess a changed site. The shape of that dip is what tells you whether it is settling or damage.

Weeks after launch	Normal settling	Something is wrong
1 to 2	Dip of 10 to 25 percent, indexing counts stable	Sharp fall, excluded pages rising
3 to 4	Recovering toward the previous level	Still falling, or flat at the bottom
6 to 8	At or near the previous baseline	No recovery trend at all
Beyond 8	Settled	Stop waiting and diagnose properly

A dip that deepens after week six is not settling. That is the point at which continuing to wait costs more than investigating.

This is not your cause if

• The decline started before the launch date. Check the exact break date rather than trusting the launch date. A redesign shipped during an existing decline often takes the blame for it.
• Your URLs did not change and your content is intact. A purely visual refresh on identical URLs rarely causes a meaningful drop.
• Indexing counts are stable and positions fell evenly across templates. That is an algorithmic pattern, not a mechanical one.
• Impressions held while clicks fell. That is a click problem rather than a visibility problem, covered in traffic dropped but rankings the same.
• Only one section fell. Worth checking whether that section was changed at all before blaming the redesign as a whole.

What order should I fix things in?

Fastest and most damaging first. Each step either resolves the problem or rules out a cause, and none of them requires waiting on a reassessment cycle.

1. Remove any noindex tag or robots block, then request indexing on your top pages.
2. Restore redirects for the old URLs that earned the most clicks, one hop each, to relevant destinations.
3. Put back removed content on pages that were ranking, starting with the highest earners.
4. Restore internal links to important pages, particularly anything now orphaned.
5. Confirm content renders in the HTML rather than only after JavaScript.
6. Then, and only then, consider whether an algorithm update is also involved.

Establish your exact break date first with the drop analyzer, because if it predates the launch this whole page is the wrong diagnosis. The page scope analyzer will show whether specific URLs vanished or everything fell together, which is the distinction that separates mechanical damage from an algorithmic one.

Frequently asked questions

Why did my traffic drop after a website redesign?

Usually a mechanical fault rather than an algorithm change: changed URLs without redirects, a staging noindex tag shipped live, removed content, lost internal links, or content that now loads only through JavaScript. Check the Indexing report for a jump in excluded pages.

How long does it take to recover from a redesign drop?

One to four weeks after the fault is fixed, because recovery follows recrawling rather than any reassessment cycle. A correctly executed redesign typically dips for two to four weeks and then returns without intervention.

Should I redirect old URLs to the homepage?

No. Google treats mass homepage redirects as soft 404s, so they pass nothing. Redirect each old URL to its closest equivalent, and where none exists a genuine 404 is the more honest answer.

Can a redesign cause a Google penalty?

No. A penalty means a manual action, which appears in Search Console and is a human decision about policy violations. A redesign causes mechanical indexing and ranking problems, not penalties.

Does changing my design affect rankings?

Visual changes on identical URLs with identical content rarely do. What affects rankings is what a redesign changes underneath: URLs, content volume, heading structure, internal linking and how content is rendered.

How do I know if content is missing after a redesign?

Compare your top pages against archived copies in the Wayback Machine and check word counts. Then use the URL Inspection tool to view the rendered HTML, since content loaded by JavaScript may be visible to you and absent to Google.