How to de-index old URLs after redesigning the website?
-
Thank you for reading.
After redesigning my website (5 months ago) in my crawl reports (Moz, Search Console) I still get tons of 404 pages which all seems to be the URLs from my previous website (same root domain).
It would be nonsense to 301 redirect them as there are to many URLs. (or would it be nonsense?)
What is the best way to deal with this issue?
-
Thank you Clever PhD, really valuable insights!
-
I completely agree with all of the above - I've taken her point more like my own. Where receiving thousands of annoying 404 errors from pages that haven't existed for many months just gets annoying!
-
I respectfully disagree with all of the above. Please repeat after me, 404s are not bad, they are diagnostic, 404s are not bad, they are diagnostic, 404s are not bad, they are diagnostic.
After redesigning my website (5 months ago) in my crawl reports (Moz, Search Console) I still get tons of 404 pages which all seems to be the URLs from my previous website (same root domain).
**Part 1 Internal links that 404s from Moz Crawl: **The 404s that show up in the Moz crawl are only going to be from an internal link on your website. The Moz crawl only looks at internal links and not links from other website. In other words, if you see 404s in your Moz crawl, that means, somewhere, you are linking to those pages and that is why the 404s are showing up. Download the CSV and you will find them in your Moz crawl. Other tools such as screaming frog, Botify, Deep Crawl, will show you a similar analysis.
Simple solution. Go through your code and remove the internal links on your site that direct the Moz crawler to those pages and the 404s will go away. (FYI this same approach will work for any internal 301s) These 404 errors in the Moz report are great diagnostic signals on where to fix your site. It is bad for users to click on a link within your website and get sent to a page that does not exist.
**Part 2 external links from Search Console: **The 404s that show up in Search console can come from your internal links on your site AND external links from other sites. Google will keep trying to crawl these links due to other sites linking to pages on your site and your own internal links. For internal link fixing - see suggestion above. For external links you need a different approach.
Look at the external links, where are they coming from? Are they from quality websites? Do they go to formerly important pages on your websites (ie pages that were good converters? If so, then use the 301 redirect to send them to the correct replacement page (and this is not always the home page). You get users to the correct page and also any link equity is passed along as well and this can help with your site rankings. If the link goes to former page on your site that was not any good to start with and the links that come into it are poor quality, then you just let the page 404. Tools such as Moz Open Site Explorer or Ahrefs or Majestic can help with this assessment - but usually you can just look at a site linking to you and tell if it is crap or not.
You need to consider the above regardless of if you want to get the pages that are 404ing in question out of the Google index as if you get Google to remove the page from the index, it will then see the internal link on your site and then find the 404 again. If you have removed the links to the 404 pages on your site, eventually Google will stop crawling them and drop out of the index.
Important note regarding the use of robots.txt. Blocking Google from crawling the 404s will not remove the pages from the index, Google will just stop crawling them. Google has to be able to crawl the URL to see the 404 and then see that it is a bad page and then remove the page from the index. Blocking with robots.txt stops Google from doing that. As soon as you take the page out of robots Google will recrawl and the 404 shows up again. Robots.txt treats a symptom that is a red herring, allowing the 404 to occur takes care of the issue permanently.
Dead pages are a natural part of the web. Let Google see the 404 (if it truly is a page that should 404 and has no link equity that should be passed along with a 301). Google will crawl the 404 several times, you will see it in search console several times. It is ok. You are not penalized for X number of 404s. You may lose ranking if you 404 a page that Google used to rank well, but this is just because Google will not keep a page highly ranked that does not exist :-). Help Google out by cleaning up your internal link structure so when it sees that you do not link to the page any more, then that is a signal that the page should 404. Google knows that due to the nature of the web, pages will time out on occasion and show an error. Google will continue to recrawl a page just to make sure, it wants to give you the benefit of the doubt. Therefore, you have to give clear directives by not linking to dead pages so that after Google double and triple checks the page, it will finally drop it. You will see the 404 in your Search Console for several months then it will eventually go away.
Hope that makes sense. Good luck!
-
Hey Lana, If you really think that 301 does not make sense in that case you can always add the URLs in the robots.txt file and once Google will recrawl your website, Google will de-index the pages from the index.
Another thing you can do is using the de-index feature in Google webmaster tool. You can do that by getting in to your GWT, Optimization > Remove URLs and do that accordingly.
Hope this helps!
-
I see the point. Thanks Liam. As the most of our 404 pages starts with /en-GB/ i will do like this:
Disallow: /en-GB/
-
Hi Lana,
I've been having the same problem on one of our websites. I've been 301 redirecting over 5,000 URL's but still receive a lot of 404 errors. One of the main reasons for these 404 errors still appearing is other bots such as Bing Bot that is still crawling the old URL's.
To resolve this, I would just block them in your robots.txt file. We blocked our old product URL's that were under a "product directory like this:
User-agent: *
Disallow: /product/
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Domain name in URL
Hi we are targeting local cities across the UK with landing pages for our website. We have built up a few links for https://www.caffeienmarketing.co.uk/bristol/ and recently advised that I should change the URL to https://www.caffeienmarketing.co.uk/marketing-agency-bristol/ and 301 directing the old one 2 questions really: 1. is there any benefit in doing this these days in that this is the main keyword target we have for this page? 2. Do I get 100% benefit for all the links built up on the old page
Intermediate & Advanced SEO | | Caffeine_Marketing0 -
SEO issues? New functionality added to website and now hash (in URL) - fragments
Hi All! We have new nice functionality on website, but now i doubt if we will have SEO issues. Duplicate content and if google is able to spider our website. See: http://www.allesvoorbbq.nl/boretti-da-vinci-nero.html#608=1370
Intermediate & Advanced SEO | | RetailClicks
With the new functionality we can switch between colors of the models (black / white / red / yellow).
When you switch with Ajax the content of other models is fetched without refreshing the page. (so the url initial part of url stays the same (for initial model) only part behind # changes. The other models are also accessible by there own url, like the red one: http://www.allesvoorbbq.nl/boretti-da-vinci-rosso.html#608=1372 So far so good. But now the questions: 1. We use to have url like /boretti-da-vinci-nero.html - also our canonical is that way But now if we access that url our system is adding automatically the #123-123 to the url to indicate which model(color) is shown. Is this hurting SEO or confusing google? Because it seems that the clean url is not accessible anymore? (it adds now #123-123) 2. Should we add some tags around the different types (colors) to prevent google from indexing that part of website? Every info would be very helpfull! We do not want to lose our nice rankings thanks to MOZ! Thanks all!
Jeroen0 -
Google and PDF indexing
It was recently brought to my attention that one of the PDFs on our site wasn't showing up when looking for a particular phrase within the document. The user was trying to search only within our site. Once I removed the site restriction - I noticed that there was another site using the exact same PDF. It appears Google is indexing that PDF but not ours. The name, title, and content are the same. Is there any way to get around this? I find it interesting as we use GSA and within GSA it shows up for the phrase. I have to imagine Google is saying that it already has the PDF and therefore is ignoring our PDF. Any tricks to get around this? BTW - both sites rightfully should have the PDF. One is a client site and they are allowed to host the PDFs created for them. However, I'd like Mathematica to also be listed. Query: no site restriction (notice: Teach for america comes up #1 and Mathematica is not listed). https://www.google.com/search?as_q=&as_epq=HSAC_final_rpt_9_2013.pdf&as_oq=&as_eq=&as_nlo=&as_nhi=&lr=&cr=&as_qdr=all&as_sitesearch=&as_occt=any&safe=images&tbs=&as_filetype=pdf&as_rights=&gws_rd=ssl#q=HSAC_final_rpt_9_2013.pdf+"Teach+charlotte"+filetype:pdf&as_qdr=all&filter=0 Query: site restriction (notice that it doesn't find the phrase and redirects to any of the words) https://www.google.com/search?as_q=&as_epq=HSAC_final_rpt_9_2013.pdf&as_oq=&as_eq=&as_nlo=&as_nhi=&lr=&cr=&as_qdr=all&as_sitesearch=&as_occt=any&safe=images&tbs=&as_filetype=pdf&as_rights=&gws_rd=ssl#as_qdr=all&q="Teach+charlotte"+site:www.mathematica-mpr.com+filetype:pdf
Intermediate & Advanced SEO | | jpfleiderer0 -
Is it worth redirecting an old domain name which was hacked to my new website?
I had a website which got hacked and malware added to it. I have since closed that website down but I still have the domain name. That domain name prior to the malware was incredibly well ranking for its niche and had a good range of high quality links to it and a domain age of 6 years. I'm now creating a new website which is similar to the old one (the same but with a different platform and layout). Is it a good or bad idea to redirect the old domain name to the new website?
Intermediate & Advanced SEO | | james.rose0 -
Google showing high volume of URLs blocked by robots.txt in in index-should we be concerned?
if we search site:domain.com vs www.domain.com, We see: 130,000 vs 15,000 results. When reviewing the site:domain.com results, we're finding that the majority of the URLs showing are blocked by robots.txt. They are subdomains that we use as production environments (and contain similar content as the rest of our site). And, we also find the message "In order to show you the most relevant results, we have omitted some entries very similar to the 541 already displayed." SEER Interactive mentions that this is one way to gauge a Panda penalty: http://www.seerinteractive.com/blog/100-panda-recovery-what-we-learned-to-identify-issues-get-your-traffic-back We were hit by Panda some time back--is this an issue we should address? Should we unblock the subdomains and add noindex, follow?
Intermediate & Advanced SEO | | nicole.healthline0 -
Search traffic decline after redesign and new URL
Howdy Mozzers I’ve been a Moz fan since 2005, and been doing SEO since. This is my first major question to the community! I just started working for a new company in-house, and we’ve uncovered a serious problem. This is a bit of a long one, so I’m hoping you’ll stick it out with me! ***Since the images aren't working, here's a link to the google doc with images. https://docs.google.com/document/d/1I-iLDjBXI4d59Kl3uRMwLvpihWWKF3bQFTTNRb1R3ZM/edit?usp=sharing Background The site has gone through a few changes in the past few years. Drupal 5 and 6 hosted at bcbusinessonline.ca and now on Drupal 7 hosted at bcbusiness.ca. The redesigned responsive design site launched on January 9th, 2013. This includes changing the structure of the URL’s, such as categories, tags, and articles. We submitted a change of address through GWT shortly after the change. Problem Organic site traffic is down 50% over the last three months. Below, Google analytics, and Google Webmaster Tools shows the decline. *They used the same UA number for Google analytics, so that’s why the data is continuous Organic traffic to the site. January 2011 - Dips in January are because of the business crowd on holidays. Google Webmaster Tools data exported for bcbusiness.ca starting as far back as I could get. Redirects During the switch, the site went from bcbusinessonline.ca to bcbusiness.ca. They were implemented as 302’s on January 9th, 2013 to test, then on January 15th, they were all made 301’s. Here is how they were set up: Original: http://www.bcbusinessonline.ca/bcb/bc-blogs/conference/2010/10/07/11-phrases-never-use-your-resume --301-- http://www.bcbusiness.ca/bcb/bc-blogs/conference/2010/10/07/11-phrases-never-use-your-resume --301-- http://www.bcbusiness.ca/careers/11-phrases-never-to-use-on-your-resume Canonical issue On bcbusiness.ca, there are article pages (example) that are paginated. All of the page 2 to page N were set to the first page of the article. We addressed this issue by removing the canonical tag completely from the site on April 16th, 2013. Then, by walking through the Ayima Pagination Guide we decided for immediate and least work choice was to noindex, follow all the pages that simply list articles (example). Google Algorithm Changes (Penguin or Panda) According to SEOmoz Google Algorithm Changes there is no releases that could have impacted our site at the February 20th ballpark. However - Sitemap We have a sitemap submitted to Google Webmaster Tools, and currently have 4,229 pages indexed of 4,312 submitted. But there are a few pages we looked at that there is an inconsistency between what GWT is reporting and what a “site:” search reports. Why would the submit to index button be showing, if it’s in the index? That page is in the sitemap. Updated: 2012-11-28T22:08Z Change Frequency: Yearly Priority: 0.5 *GWT Index Stats from bcbusiness.ca What we looked at so far The redirects are all currently 301’s GWT is reporting good DNS, Server Connectivity, and Robots.txt Fetch We don’t have noindex or nofollow on pages where we haven’t intended them to be. Robots.txt isn’t blocking GoogleBot, or any pages we want to rank. We have added nofollow to all ‘Promoted Content’ or paid advertising / advertorials We had TextLinkAds on our site at one point but I removed them once I satarted working here (April 1). Sitemaps were linking to the old URL, but now updated (April)
Intermediate & Advanced SEO | | Canada_wide_media1 -
Unable to Crawl my Website
Hi all, I have a website that I am trying to promote, but tried to add it here in SEOMoz and got the following message: We have detected that the root domain evolving-networks.co.uk does not respond to web requests. Using this domain, we will be unable to crawl your site or present accurate SERP information. Does anyone know why this website cannot be crawled? Please help. Thank you in advance!
Intermediate & Advanced SEO | | LSDigital0 -
Removing pages from index
Hello, I run an e-commerce website. I just realized that Google has "pagination" pages in the index which should not be there. In fact, I have no idea how they got there. For example, www.mydomain.com/category-name.asp?page=3434532
Intermediate & Advanced SEO | | AlexGop
There are hundreds of these pages in the index. There are no links to these pages on the website, so I am assuming someone is trying to ruin my rankings by linking to the pages that do not exist. The page content displays category information with no products. I realize that its a flaw in design, and I am working on fixing it (301 none existent pages). Meanwhile, I am not sure if I should request removal of these pages. If so, what is the best way to request bulk removal. Also, should I 301, 404 or 410 these pages? Any help would be appreciated. Thanks, Alex0