Getting found online · Updated October 4, 2026
Why did traffic fall after we launched a new website?
Updated October 4, 2026
Traffic usually falls after a relaunch because old page addresses were not redirected, pages that brought calls were cut, or the new site still blocks Google with a leftover noindex or robots.txt rule. A short dip while Google recrawls is normal. Confirm the drop in Search Console first, since a missing analytics tag can fake one.
First, confirm the drop in Search Console, not in your analytics
If your analytics or your web designer's report says traffic fell, open Search Console's Performance report before you change anything. It counts "The number of times a user clicked your site from Google Search results," so it does not depend on a tag in your pages. If clicks held while analytics shows a cliff, the drop is in the measuring.
Where to look first
Search Console
It counts clicks from Google's results, so a tag missing from the new site cannot make it fall.
A common fake drop is a missing tag. Google Analytics asks you to paste the Google tag "immediately after the <head> on each page of your website," and a site builder has its own field for the tag ID, which starts empty on a new site. Open a page of the new site, then the Realtime report, which shows the past 30 minutes. If your visit is missing, with ad blockers off and the cookie banner accepted, the tag is probably not running there.
Then check calls and form leads, not only sessions. A new site can lose the call tracking script, the form email or the Google Ads conversion tag. Count booked jobs before and after launch. If jobs held and only reports fell, fix the tracking and stop there.
If Search Console clicks fell too, the drop is real. Work down this page in order: what fell, whether Google is blocked, redirects, missing pages, then changes to the pages that survived.
- Open Performance, then Search results. Set the date range wide enough to cover several weeks before launch and the weeks since.
- Choose Weekly above the chart so weekends do not look like drops. Read Total clicks and Total impressions.
- If both fell from launch week and stayed down, the drop is real. If they dipped and are climbing back, Google is still recrawling; recheck in two weeks.
Then compare before and after by page to see what lost the clicks
Open the date filter, choose Compare, and set the weeks since launch against the same number of weeks before. On the Pages tab, sort by Clicks Difference, which Google's drop guide suggests to find "pages that lost the most traffic."
For example
A roofing company's site earned 600 clicks in the four weeks before launch and 445 in the four weeks after. Analytics showed about two thirds of visits gone, because the tag was only on the home page. Search Console showed the real loss: 155 clicks. On the Pages tab, the old /roof-repair page went from 180 clicks to 10 and the new /services/roof-repair page earned 15, while the home page (200) and the other pages (220) held. The whole loss sits on one page. Google was still showing the old address, and nothing told it the new page had replaced it. A missing redirect, not a penalty.
How to read the pattern on the Pages tab
Each pattern points to a topic below.
Every page fell, home page included. Google may be blocked from the whole site.
Old addresses lost clicks; new versions gained little. Redirects are missing or wrong.
An old address lost its clicks and no new page covers that job. The page was cut.
Impressions held but clicks fell. Google's guide says that pattern can mean "you might not be generating the best page title and snippet."
One group of kept pages sank in position. Compare what changed on them.
The causes, most damaging first
If the whole site fell, remove the noindex or robots.txt block left over from building it
If every page lost clicks together, check whether the new site still tells Google to stay away. Google's move guide warns that "some site owners block all crawling while in development," and its troubleshooting list starts with: "Don't forget to remove any noindex or robots.txt blocks that were only needed for the migration."
Look in the site builder first. Site builders often have a search visibility setting. Google says your CMS "might expose a search settings page or some other mechanism to tell search engines whether or not to crawl your page." If it was switched on for the build, switch it off.
Then check what Google saw. Paste the home page and one service page into the URL Inspection bar at the top of Search Console and click Test live URL. Read two lines in the result: Crawl allowed? and Indexing allowed? Either one saying No is the block. One trap: Google says that when robots.txt blocks a page, Indexing allowed? "will always be 'Yes'" because Google cannot see the noindex, so a Yes there means nothing until Crawl allowed? is Yes too. Then open the Page indexing report. Google says a fall in indexed pages "without a corresponding increase in errors" can mean you are "blocking access to your existing pages via robots.txt, 'noindex' or a required login."
Don't skip that report because the site is small. Its help page says a site under 500 pages "probably" doesn't need it. Google also says error spikes "might be caused by a change in your template," which is what a relaunch is: one setting can block every page at once, and this report shows it as a single spike.
A noindex drop is slow, so it can hide. Google notes that a misplaced noindex depends on Google recrawling each page, "meaning there would be a slower drop in traffic." A gradual slide after launch fits this cause too.
Why fixing robots.txt alone can leave pages out
Robots.txt stops Google fetching a page; noindex is a line on the page telling Google to drop it. Google says that if a page is blocked by robots.txt, "the crawler will never see the noindex rule." So a site launched with both can look fixed once robots.txt is cleared, until Google fetches each page, finds the noindex, and drops it.
Remove both. Then, in the robots.txt report, choose Request a recrawl on the file. Google says this "doesn't guarantee an immediate recrawl of unblocked URLs," so also click Request indexing in URL Inspection for the home page and your top service pages. There is a daily limit on inspection requests, so spend them on the pages that brought calls.
If old addresses lost their clicks, 301 each one to the matching new page
Make a list of every old page address, write the new address that covers the same job beside each one, and set a permanent (301) redirect for every pair. A redesign usually changes addresses (/water-heaters.html becomes /services/water-heater-repair), and Google's move guide says it "frequently" sees people "redirecting to the wrong (non-existent) URLs on the new site."
Without a redirect, the old address returns an error and Google cannot tell the new page replaces it. Google is direct that "301 and other permanent redirects don't cause a loss in PageRank."
Not everything to the home page. Google's guide says: "Don't redirect many old URLs to one irrelevant single URL destination, such as the home page of the new site. This can confuse users and might be treated as a soft 404 error." An old drain cleaning page goes to the new drain cleaning page. The exception Google allows: several old pages merged into one new page can all point to it.
Permanent, not temporary. Google shows the new address in results for a permanent redirect and the old one for a temporary (302) redirect. Use the builder's own redirect feature if it has one (Google says platforms "may already have built-in redirect solutions") and choose the permanent option.
Point straight at the final page. Google advises "redirecting to the final destination directly," which matters most when this relaunch sits on top of an older one.
Test every pair in a browser. Google's move guide suggests URL Inspection for testing redirects, but its help page says the live test checks "the redirect target without indicating that there is a redirection." A green live result on an old address only says the destination works; the indexed result is the one that shows the redirect. Open each old address in a browser and confirm it lands on the right page, and have whoever built the site run a redirect check that reports the status code. Google's own move guide points to the crawler Screaming Frog for this. In the Page indexing report, old addresses under Page with redirect are working as intended; old addresses under Not found (404) still need one.
How long do the redirects have to stay?
Google gives two answers. The site move guide says "generally at least 1 year." The Change of Address help page says "at least 180 days." Use the longer one, and longer still if you can. Change of Address help adds "longer if you still see any traffic to them from Google Search," and the move guide says "From users' perspective, consider keeping redirects indefinitely." It also notes that "redirects are slow for users," so update the links you control to point at the new addresses. Judgment: links in directories, emails and bookmarks work only while the redirect does, so keep the rules as long as the site runs.
Should a page you dropped on purpose redirect anywhere?
Only if a new page covers the same job. Otherwise Google says it should "correctly return an HTTP 404 or 410 error response code." Sending it to the home page is what Google warns "might be treated as a soft 404 error."
If pages that brought calls were cut, put them back
Find the old addresses on the Pages tab that earned clicks before launch and have no counterpart on the new site, and rebuild those pages. A redesign that merged six service pages into one tidy list removed the pages Google ranked for those six jobs.
Google's drop guide gives the rule for missing searches: "If there are queries missing from your traffic, check if you have content on that subject and make sure it's being crawled and indexed." Switch to the Queries tab in the same comparison. Searches like "emergency furnace repair" that fell to zero name the content that went missing.
Shortened pages count as cut pages. If the old page had prices, photos and answers and the new one has two sentences and a button, put the substance back.
Rank the rebuilds by calls, not clicks. If call tracking or form leads show which pages people landed on before they called, restore those first. A page with 40 clicks a month and three booked jobs matters more than a blog post with 400 clicks and none.
What if Google changed its rankings at the same time?
Check Google's list of ranking updates around launch, as its drop guide suggests, and widen the report to Last 16 months to spot a yearly dip. A drop that starts in launch week on the pages that changed is the relaunch.
If kept pages slipped, compare their titles, headings and links with the old site
Put the old and new version of each slipping page side by side and compare the title, the main heading and the links pointing to it. New templates often swap descriptive titles for generic ones and drop menu links.
Titles. Google calls the title link "often the primary piece of information people use to decide which result to click." It warns against vague titles like "Home" and asks for distinct text describing each page. "Services | Company Name" on every page throws away what "Water Heater Repair in Mesa" told searchers.
Links. Google says it can "generally" crawl a link only "if it's an <a> HTML element ... with an href attribute," and that "Every page you care about should have a link from at least one other page on your site." A menu built from buttons that run a script can leave pages Google cannot reach. Ask the builder whether the menu uses ordinary links.
Domain changes, the old site, and how long to wait
If the domain changed, file a Change of Address; if only the design did, don't
Decide which kind of launch you had; Google's move tool fits only one of them.
- Same domain, new design
- The domain stays, and pages may get new addresses. Google says not to use the Change of Address tool for "Moving some pages from one location to another within your site", for http to https, or for www to non-www. Redirects and an updated sitemap do the work.
- Everything on this page above applies. Nothing else needs filing.
- New domain
- Set up the redirects first, then open Change of Address in Search Console for the old domain. You must own both properties with the same Google account, and Google asks you to file it for every version of the old domain, www and non-www, "even if you're not actively using these variants."
- Google says to keep paying for the old domain for at least a year so nobody else can buy it.
No Change of Address
Change of Address
A new domain plus a redesign is the riskiest launch. Google's Change of Address help says that if you combine a site move with a redesign of content and URL structure, "you will probably see some traffic loss as Google might need to relearn and reassess the individual pages." Its move guide says to change one thing at a time: "move to a new domain, then change your site's layout." If both happened, the fixes are the same; expect a slower recovery.
Mind stale names in Google's own guide. The site move article still uses old names: the "Index Coverage report" and "Index Status report" are now the Page indexing report, and "Links to your site" is now the Links report.
Update every link you control, starting with your ads and Business Profile
Change the website link on your Google Business Profile, the final URLs in your Google Ads, and your listings on directories and social profiles to the new addresses. Google's move guide lists "Ad campaigns to point to the new landing pages" and profile links among the links to update right after a move.
Ads are the costly one. Google Ads disapproves ads whose destination returns an HTTP error (Destination not working) or whose final URL redirects to a different domain (Destination mismatch). A disapproved ad "can't show anywhere," Google says, so an ad pointing at a deleted page can stop serving, and in the reports that looks like fewer calls from the website. Check your ads' status for disapprovals the week you launch.
The Business Profile link carries calls too. Google's Business Profile help asks you to "Make sure that Googlebot isn't blocked on your site," which a leftover robots.txt block breaks.
If the old site is gone, rebuild the list of old addresses from Search Console and the Wayback Machine
If nobody saved the old page list, export it from Search Console's Performance report: set the date range to the months before launch, open the Pages tab, and export it. Google says "Many reports provide an export button to download both the chart and table data." That list is ranked by clicks, so it starts with the addresses that matter most.
Search Console keeps 16 months of data, so this works for a launch up to about a year back. The table shows at most 1,000 rows, enough for most local business sites. On a domain change, export from the old domain's property; it must still be verified in your account.
- Search Console Pages tab. The best list for redirects, ranked by what earned traffic.
- The Wayback Machine. The Internet Archive says the way to see everything it saved for a site is web.archive.org/*/ followed by your domain and /*. It also shows what each old page said. The archive says it "can't guarantee" any site was saved, so treat it as a second source.
- Old sitemaps, analytics and the Links report. Google's move guide names your sitemaps, "server logs or analytics software," and the Links report in Search Console as places to find old addresses. Analytics still lists old page paths for the dates before launch.
After the fixes, give Google a few weeks, then rerun the same comparison
Rerun the before-and-after comparison on the Pages tab three to four weeks after the fixes, and judge it on calls and booked jobs as well as clicks.
Google says "To consider a site move complete, Googlebot will have to visit every URL on your old and new site at least once," and that "a small to medium-sized website can take a few weeks for most pages to move." It adds that "visibility of your content in Search may fluctuate temporarily during the move." A page that was blocked by noindex returns only when Google recrawls it, and Google says that "Depending on the importance of the page on the internet, it may take months for Googlebot to revisit a page." Request indexing for the pages that matter most rather than waiting, and submit the new sitemap in the Sitemaps report, as Google's move guide says. If a builder like Wix or Squarespace runs the site, Google says it probably manages the sitemap for you; open it once and check it lists the new addresses.
The judgment call. Google's own timeline is "a few weeks or more" for a medium-sized site, so a dip that recovers in that time is the recrawl it describes. A drop that still holds after a month, on the same pages, with the fixes in place, is worth rechecking rather than waiting out. Recheck Page indexing for old addresses still under Not found (404), and the Pages tab for pages still missing from the new site.