Proper CDN Implementation
-
I am working with a site that is using the Cloudflare CDN to deliver their images. The problem is that the URLs for the images are all cloudflare URLs (cloudflare.com/example-filename.png). We want to use a cname to change the image URLs to be on a subdomain of their site to take advantage of the SEO strength.
Currently, they get a high percentage of traffic from image search and Google is directing users to their site when returning images hosted on Cloudflare, so I am apprehensive about making changes. I'm curious if anyone has any experience with this or has any insight on how Google will handle it.
Will the longer term benefits out way any short term drop in traffic from image search? WIll there be a drop in image search?
Any thoughts or experiences are appreciated.
Thanks in advance.
-
Hi Lisa
This is interesting.
As far as I can tell, everything is configured 100% correctly.
Few questions:
1. Do you keep all images in /sites/default/files/?
Were all (most) of your errors related to items using that path?2. This could be a downtime issue. Did you experience any availability problems?
3. Do you see any changes in the error report? (different results for the same URL)
If so, this makes my downtime theory more plausible.Also, the site loaded VERY slowly for me.
I've run Google Page speed and saw that you have issue with Cache headers (low expiration date that makes it harder to leverage local and CDN caching capabilities)
This should be looked at. Beyond the immediate ux and seo implications, this can contribute to server load and can be have a negative effect on website's availability. (i.e. the above mentioned downtime)
Best
Igal
-
Igal-
Related question. We just moved to a CDN for our images. Our site is www.mnn.com and our CDN is http://images.mnn.com/. The only thing we’re allowing Google to crawl are the images, so we shouldn’t have any duplicate content issues.
However, what I AM seeing is several thousand warnings in Google Webmaster Tools for our video sitemap. Our video sitemap contains about 618 videos, but we have 2,472 errors citing “Sitemap contains urls which are blocked by robots.txt” which I believe is related to the video icons in search.
When I look for a video in Google, I still see our icons related. I can’t seem to figure this one out.
Any thoughts?
Lisa
-
I`ve recently covered this in a blog post about SEO & CDN Myth busting. The short answer is "No", as Google is well aware of CDN structures.
I think the image traffic drop is un-related and it may have something to do with Bot Blocking features that prevent access from Google Image bot (which, as we recently discovered, can use non-US IPs and thus may also be considered as "fake" ).Having said that, Google official statement says that speed factors only affect 1% of SERP results. If your image traffic is high or/and valuable I would re-think this.
-
The typical way of doing this is to setup a subdomain, specifically for the content that is going to be served from a CDN, and then use a CNAME to map the ugly CDN url to the clean subdomain that you setup. I've only done this with Rackspace and AWS. If you want a good description of how this is done on AWS, see this link:
http://www.labnol.org/internet/setup-content-delivery-network-with-amazon-s3-cloudfront/5446/
Apparently, cloudflare has a way to map your website url to the CDN, as long as you're hosting with a cloud server system that they support. I haven't tried this before, but it might actually be a cleaner way of doing this, if you're hosting setup supports it.
http://blog.cloudflare.com/zone-apex-naked-domain-root-domain-cname-supp
It sounds like this would be a big win for your client, as they get a large percentage of traffic from image traffic. Using a CDN service is going to improve the speed of loading images and reduce their bandwidth bill, and mapping the CNAME should give their domain the credit for being the source of the image.
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
-
Did We Implement Structured Data Correctly?
Our designer/developer recently implemented structured data on our pages. I'm trying to become more educated on how it works since I'm the SEO marketing specialist on the team and the one that writes and publishes the majority of our content. I'm aware it's extremely important and needs to be done, I just don't know how to do it yet. The developer was on our team for over a year, we recently let him go. Now, I'm going through all the pages to make sure it's done correctly. I'm using the structured data testing tool to look at the pages and have been playing with the structured data markup helper. I would REALLY appreciate it if one of my fellow MOZ fans & family can help me determine if it's done correctly. We do not currently have any schema plugs installed that I know of. So I'm not sure how he implemented the Schema code. I would like to know what I need to do moving forward to the additional content we publish as well as what to do to correctly implement Schema if not already. When I manually look at one of our FAQ pages I see multiple schema data formats detected... I'm not sure if we're supposed to have multiple or just one----> https://www.screencast.com/t/TjHphL7jsI I also noticed in the Question schema data for that same page... the accepted answer is empty. I would image that should have the short version of the answer to the question in it?--->https://www.screencast.com/t/e6ppXkhXd7QS Here's a screenshot of our structured data info from Google search console---> https://www.screencast.com/t/KHj4BGgdrZ4m HELP please! Our website consists of 25-30 "product" pages https://www.medicarefaq.com/medigap/ https://www.medicarefaq.com/medicare-supplement/ https://www.medicarefaq.com/medigap/plan-f/ https://www.medicarefaq.com/medicare-supplement/plan-f/ We currently have about 75 FAQ pages and adding 4-6 per month. This is what brings in most our traffic. https://www.medicarefaq.com/faqs/2018-top-medicare-supplement-insurance-plans/ https://www.medicarefaq.com/faqs/2018-medicare-high-deductible-plan-f-changes https://www.medicarefaq.com/faqs/medicare-guaranteed-issue-rights We have 100 state specific pages (two for each state) https://www.medicarefaq.com/medicare-supplement/florida/ https://www.medicarefaq.com/medigap/florida/ https://www.medicarefaq.com/medicare-supplement/California/ https://www.medicarefaq.com/medigap/California/ We have 20ish carrier specific pages https://www.medicarefaq.com/medicare-supplement/humana/ https://www.medicarefaq.com/medicare-supplement/mutual-of-omaha/ Then we have about 30 blog pages so far and are publishing new blog posts weekly https://www.medicarefaq.com/blog/average-age-retirement-rising/ https://www.medicarefaq.com/blog/social-security-benefit-increase-announced-2018 https://www.medicarefaq.com/blog/new-california-bill-force-drugmakers-explain-price-hikes
Intermediate & Advanced SEO | | LindsayE0 -
Json LD e-commerce site with Excellent implementation of all markup features
Hi all I am looking for some really good clear examples of sites that have excellent JSON LD markup. Not just the basics but packed to the teeth with markup for every element. I am particularly interested in e-commerce applications as I am re skinning our e-commerce platform written from scratch in house. It is far from perfect, not mobile friendly and well a bit backward but links into everything we have in a seamless way all the way to our manufacturing plant. Take a look have a little laugh and then take pity 🙂 https://www.spurshelving.co.uk/shop/shop.aspx Thanks Pete
Intermediate & Advanced SEO | | Eff-Commerce0 -
Just implemented microdata, what should one typically expect?
Should I expect my rankings to do any weird things over the next couple of days or weeks?
Intermediate & Advanced SEO | | Edward_Sturm0 -
Duplicate content across similar computer "models" and how to properly handle it.
I run a website that revolves around a niche rugged computer market. There are several "main" models for each computer that also has several (300-400) "sub" models that only vary by specifications for each model. My problem is I can't really consolidate each model to one product page to avoid duplicate content. To have something like a drop down list would be massive and confusing to the customer when they could just search the model they needed. Also I would say 80-90% of the market searches for a specific model when they go to purchase or in Google. A lot of our customers are city government, fire departments, police departments etc. they get a list of approved models and purchase off that they don't really search by specs or "configure" a model so each model number having a chance to rank is important. Currently we have all models in each sub category rel=canonical back to the main category page for that model. Is there a better way to go about this? Example page you can see how there are several models all product descriptions are the same they only vary by model writing a unique description for each one is an unrealistic possibility for us. Any suggestions on this would be appreciated I keep going back on forth on what the correct solution would be.
Intermediate & Advanced SEO | | The_Rugged_Store0 -
How To Implement Pagination Properly? Important and Urgent!
I have seen many instructions but I am still uncertain. Here is the situation We will be implementing rel prev rel next on our paginted pages. The question is: Do we implement self referencing canonical URL on the main page and each paginated page? Do we implement noindex/follow meta robots tag on each paginated page? Do we include the canonical URL for each paginated page in the sitemap if we do not add the meta robots tag? We have a view all but will not be using it due to page load capabilities...what do we do with the viewl all URL? Do we add meta robots to it? For website search results pages containing pagination should we just put a noindex/follow meta robots tag on them? We have seperate mobile URL's that also contain pagination. Do we need to consider these pages as a seperate pagination project? We already canonical all the mobile URL's to the main page of the desktop URL. Thanks!
Intermediate & Advanced SEO | | seo320 -
What is the Proper Inbound Linking methodology? Is PR Sculpting gone?
Hi, I know that once there were discussions about "navigating" all links top wise to the main pages - for example to category pages in e-commerce sites. What is the proper methodology today? One example: Should I link out from a category main page which shows my products (a serious landing page) to a page that explains about the product? It makes sense but I don't want my about page to become the more important page (even though it does link back to the products category page). Thanks
Intermediate & Advanced SEO | | BeytzNet0 -
Will implementing a 'Scroll to Div Anchor' cause a duplicate content issue?
I have just been building a website for a client with pages that contain a lot of text content. To make things easier for site visitors I have created a menu bar that sticks to the top of the page and the page will scroll to different areas of content (i/e different Div id anchors) Having done this I have just had the thought that this might inadvertently introduce duplicate content issue. Does anyone know if adding an #anchor to the end of a url will cause a duplicate content error in google? For example, would the following URLs be treated as different:- http://www.mysite.co.uk/services
Intermediate & Advanced SEO | | AdeLewis
http://www.mysite.co.uk/services#anchor1
http://www.mysite.co.uk/services#anchor2
http://www.mysite.co.uk/services#anchor3
http://www.mysite.co.uk/services#anchor4 Thanks.0 -
How was cdn.seomoz.org configured?
The SEOmoz CDN appears to have a "pull zone" that is set to the root of the domain, such that any static file can be addressed from either subdomain: http://www.seomoz.org/q/moz_nav_assets/images/logo.png http://cdn.seomoz.org/q/moz_nav_assets/images/logo.png The risk of this configuration is that web pages (not just images/CSS/JS) also get cached and served by the CDN. I won't put the URL here for fear of Google indexing it, but if you replace the 'www' in the URL below with 'cdn', you'll see a cached copy of the original: http://www.seomoz.org/ugc/the-greatest-attribution-ever-graphed The worst-case scenario is that the homepage gets indexed. But this doesn't happen here: http://cdn.seomoz.org/ That URL issues a 301 redirect back to the canonical www subdomain. As it should. Here's my question: how was that done? Because maxcdn.com can't do it. If you set a "pull zone" to your entire domain, they'll cache your homepage and everything else. googlebot has a field day with that; it will reindex your entire site off the CDN. Maybe the SEOmoz CDN provider (CloudFront) allows specific URLs to be blocked? Or do you detect the CloudFront IPs and serve them a 301 (which they'd proxy out to anyone requesting cdn.seomoz.org)? One solution is to create a pull zone that points to a folder, like example.com/images... but this doesn't help a complex site that has cacheable content in multiple places (do you Wordpress users really store ALL your static content under /wp-content/ ?). Or, as suggested above, dynamically detect requests from the CDN's proxy servers, and give them a 301 for any HTML-page request. This gets complex quickly, and is both prone to breakage and very difficult to regression-test. Properly retrofitting a complex site to use a CDN, without creating a half-dozen new CDN subdomains, does not appear to be easy.
Intermediate & Advanced SEO | | mcglynn0