What is the effect of CloudFlare CDN on page load speeds, hosting IP location and the ultimate SEO effect?
-
Will using a CDN like CloudFlare.com confuse search engines in terms of the location (IP address) of where the site is actually physically hosted especially since CloudFlare distributes the site's content all around the globe?
I understand it is important that if customers are mostly in a particular city it makes sense to host on an IP address in the same city for better rankings, all things else being equal?
I have a number of city-based sites but does it make having multiple hosting plans in multiple cities/ countries (to be close to customers) become suddenly a ridiculous thing with a CDN? In other words should I just reduce it down to having one hosting plan anywhere and just use the CDN to distribute it?
I am really struggling with this concept trying to understand if I should consolidate all my hosting plans under one, or if I should get rid of CloudFlare entirely (can it cause latency in come cases) and create even more locally-based hosting plans (like under site5.com who allow many city hosting plans).
I really hope you can help me somehow or point me to an expert who can clarify this confusing conundrum. Of course my overall goal is to have:
1. lowest page load times
2. best UX
3. best rankingsI do realise that other concepts are more important for rankings (great content, and links etc.) but assuming that is already in place and every other factor is equal, how can I fine tune the hosting to achieve the desirable goals above?
Many thanks!
Mark -
Well good luck with 100/100 but you should be good to get a 90-ish score.
-Andy
-
Hey thanks Andy,
I have learned a few things regarding the Page Speed Insights and was able to make a few changes regarding expiry times on cache, which increased the score.
I will tweak the suggestions some more to try to get the score to 100/100 for Mobile and Desktop, if that's possible to get 100/100.
Thanks for letting me know about that.
Regards,
Mark -
Hi Mark,
Have you run the site through Page Speed Insights? What does Google suggest? It might be that you can get some good speed increases just by making .htaccss changes.
-Andy
-
Thanks Andy, sorry I just noticed the settings button where you can pick the international locations, it wasn't obvious to me before.
Whichever page I choose, http://theatrebuddies.us it is taking between 2 and 4 seconds to load, which doesn't seem great?
I am not sure if this is a hosting issue or something to do with javascript on the page which affects load time, it's very hard to say.
I wonder if anyone can recommend a Web host with the fastest possible shared hosting, if that has anything to do with it and I could set up a plan and load one page and do a comparison.
Many thanks again.
- Mark
-
With most of these you can't Mark. You can set a different country on Pingdom, but that is about as much flexibility that you get.
-Andy
-
Hi Andy,
How can I test page load speed if when using the tool above, it does not recognise where I am? It seem to think I am in a different location globally each time I try.
Thanks so much,
Mark
-
P.S. All 148 of my sites are hosted in the US. I guess the reason I am asking is that it has been hideously time-consuming setting up all these sites that to break them up into all sorts of locally-based hosting plans would take me weeks of fiddling, down-time etc. which I've already gone though earlier this year. I moved things via CloudFlare because some sites were hacked so it sent me down a wormhole of trying to learn Web security and put best practices in place. So it's not like I have one site to move somewhere ... any answer may have huge implications in terms of the effort required to put best practices in place for all my sites. Many thanks again!
-
Hi Josh,
Thank you very much for your kind assistance! Makes sense.
I posted a reply to another response above, where I guess I fleshed out my concerns perhaps a bit better (sorry it's a very confusing thing for me to sort out here). I would be very interested and grateful to hear your comments on it.
Have a great day.
Mark -
Hi Andy,
Thanks so much for that.It is really hard for me to tell what the user's experience is since I am in Australia and users are in 7 other countries and I am not sure how to determine the UX of a person in another city.
To me it seems the sites are serving fast enough even for me here; but since the Web is so hyper-competitive, I am just trying to fine tune in every conceivable area as much as possible, with the belief that lots of little things added up is a good thing, even if they are not the top 3 determining factors of ranking but still important.
In some cases, certain sites only allow users to join if they are within a particular city so the site is totally city-specific. In this case it may make sense to not use CloudFlare but just buy hosting within that city with that city's IP address. Would that be true? It seems to me that a site for people in a certain city only would get better rankings if it's server's IP address was within the same city?
CloudFlare does have a server in this city as well.I have experienced seeing some subdomains serving so slowly through CloudFare that a warning page comes up that it could not be served (page fail) ... yet with CF suppressed, the pages serve fine, so this is a concern, but maybe it was a random instance I'm not really sure.
Thing is, because of the protection of CloudFlare it feels better to have it in place(?)
My main concern is also centred around the idea that if content is cached and distributed over dozens of servers on a CDN ... and we do a "Who's hosting" the domain and it says 'CloudFlare' then for SEO purposes in terms of serving the most relevant content to people in their local area, then how does the IP address of the hosting affect ranking within a local area? In other words, if some content is cached (e.g. images) and served via CF and the rest of the content is HTML and not cached and served from the origin server ... in effect, 'where is the home of the site' i.e. if the site is hosted in the UK because most of its customer base is in the UK (but also in other countries) then how do rankings for keywords work in the UK compared to other countries? I would think without CF the site would appear more UK-specific and UK-centric; but with CF in place ... the context of the hosting is totally lost / confused?
Many thanks for your assistance.
Best regards,
Mark -
Hi Mark,
First of all, have you tested your sites to see how they are in terms of speed? I would suggest checking with...
My own personal experience is that it can cause real issues in terms of page load times - to the point I have had clients drop it altogether. Out side of that, I can't advise on the best way to utilise it.
Are you finding that your experience is a good or bad one so far?
-Andy
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
-
Research on industries that are most competitive for SEO?
I am trying to see if there is a reputable / research-backed source that can show which industries are most competitive for search engine optimization. In particularly, I'd be interested in reports / research related to the residential real estate industry, which I believe based on anecdotal experience to be extremely competitive.
Local Website Optimization | | Kevin_P3 -
Mysterious Location Based SERP Disappearance
Hi Everyone, I've got a bit of a confusing SEO issue which I'm hoping you'll be able to help with. Apologies in advance for the long post, I've put an abridged version below also. We have one main keyword and it seems to have disappeared in some locations. The main keyword is "clothing manufacturers" and up until recently we had stability for almost a year. We're based in London, England and we regularly check "clothing manufacturers" to see where we're showing in search, and we usually see between 3rd - 5th. We use AHREFS to track rankings and noticed recently that "clothing manufacturers" had disappeared totally. We asked some people in different areas of the country to check where we were showing in search - one in Somerset, one in Liverpool, one in Beckingham and we used a VPN in Manchester. In all of these areas we aren't ranking for our main keyword at all. In London though we're 5th which is the lower end of normal. We then checked other keywords and it turns out "Clothes manufacturers" is one we're also not ranking for outside of London. However for "clothing manufacturers uk" and "clothes manufacturers uk" we are ranking for in every location we have tried. "Clothing manufacturers uk" is currently the keyword which brings us the most traffic. There are no manual penalties in webmaster tools, but looking at analytics it looks like our impressions for the main keyword have been down over the past 90 days, so we think we have had a problem and not realised for some time. Around a week before we see that our traffic for "clothing manufacturers" dropped, we made some structural changes to the website homepage, where we added LSIs, more H2s, more long tail keywords and more content, taking the copy from around 500 words to around 1100 words. This was in an effort to make the homepage less keyword stuffed and more natural. As a result of this we saw an overall increase in traffic and enquiries, and that's the reason we didn't notice for so long that traffic from "clothing manufacturers" has dropped so badly. Our first thought is that this might be something to do with Schema. Our website was until last week using a schema which included our "postal address" which is our physical office location in London. The schema was implemented in June 2017 and we have noticed that 3 months after implementing the schema, in October, our traffic fell dramatically for our main keyword, "clothing manufacturers". At the same time, our traffic for "clothing manufacturers uk" increased dramatically. Interestingly, the schemas used by our competitors don't include their office addresses and they show up all over the country for "clothing manufacturers" and "clothes manufacturers". One of our competitors is physically within half a mile of us. Have you guys seen a schema limit a company to searches only in one locality before? We have now removed the address from the schema to see if we start ranking all over the country again, like we used to before we implemented it. If this is the problem then it could take 3 months to turn around like it did for us to get in to this situation (Schema implemented June 2017, traffic fell October 2017). We're therefore trying to investigate every possibility to ensure we leave no stone unturned. Do you have any thoughts on the problem and if it could be schema related, or possibly something else? Thank you in advance! TL:DR Keywords "clothing manufacturers" and "clothes manufacturers" no longer ranking around the UK. Still ranking in London where we are based. Still ranking well for "clothing manufacturers uk" and "clothes manufacturers uk". Traffic for "clothing manufacturers" dropped 3 months after implementing schema and one week after making changes to website homepage (increased word count, added long tail keywords, LSIs and H2s). Schema included "postal address" which we notice none of our competitors have. They rank all over the country for "clothing manufacturers". One of our competitors is based within half a mile of us in London. Could having the address in the schema limit us to one locality? Could it be something else entirely?
Local Website Optimization | | rswhtn0 -
Do duplicate street addresses on 2 website affect SEO?
Hi, We have 2 websites built for one client that has 2 companies running from the same physical location. Would having the same address listed on both websites affect their SEO rankings? The 2 websites mentioned are linked below: http://anastasiablinds.ca/ http://www.greenfoxwindows.ca/ Thanks for your help!
Local Website Optimization | | Web3Marketing871 -
How Google's Doorway Pages Update Affects Local SEO
Hey Awesome Local Folks! I thought I'd take a proactive stance and start a thread on the new doorway pages update from Google, as I feel there will be questions coming up about this here in the forum: Here's the update announcement: http://googlewebmastercentral.blogspot.com/2015/03/an-update-on-doorway-pages.html And here's the part that will make local business owners and Local SEOs take a second glance at this: Here are questions to ask of pages that could be seen as doorway pages: Do the pages duplicate useful aggregations of items (locations, products, etc.) that already exist on the site for the purpose of capturing more search traffic? I think this will naturally lead to questions about the practice of creating local/city landing pages. At this point, my prediction is that this will come down to high quality vs. crummy quality pages of this type. In fact, after chatting briefly with Andrew Shotland, I'm leaning a bit toward seeing the above language as being strongly geared toward directory type sites and large franchises. I recommend reading Andrew's post about his take on this, as I think he's on the right track: http://www.localseoguide.com/googles-about-to-close-your-local-doorway-pages/ So, I'm feeling at this point that if you've made the right efforts to develop unique, high quality local landing pages, you should be good unless you are an accidental casualty of an over-zealous update. We'll see! If anyone has thoughts to contribute on this thread, I hope they will, and if lots of questions start coming up about this here in the community, feel free to link back to this thread in helping your fellow community members 🙂 Thanks, all!
Local Website Optimization | | MiriamEllis9 -
SEO for local business directory type site
I am thinking about creating a local business directory type website that lists all local Tattoo Shops. I am familiar with both local and global SEO and how to differentiate between them, however, I am not sure how I should approach this type of website. It isn't an actual business, but I want to target local searches that are looking for tattoo shops. In other words, when someone types in "tattoo shops" or "tattoo shops near me", or "tattoo parlors", I want the website to appear. Is this something that is manageable, or will the individual Tattoo Shop websites always show before mine since they are real local businesses with google+ pages?
Local Website Optimization | | brfieger0 -
Different page for each service at each location? Where does it end!
If we have 15 different locations and 10 different services, do we need to make keyword targeted landing pages for each combination? Is that actually the best method or is there some alternative? For example, if we are a law office specializing in slip and falls and car accidents, do we need a page for EACH location for each service (ie. Miami Car Accident Lawyer, Miami Slip and Fall Lawyer, Orlando Slip and Fall Lawyer, Orlando Car Accident Lawyer) etc. to maximize our ranking potential in each location? Is there a better way or are we left with this until Google gets "smarter"?
Local Website Optimization | | RickyShockley0 -
One location performing worse than the rest despite no major difference in SEO strategy
Hi all, I'm flummoxed. I'm dealing with a business that has 15 or so offices in three cities, and one city is performing horribly (this includes every office therein). The other two cities have shown consistently stellar results with massive traffic increases month over month for the past year; the city in question dropped unexpectedly in June and hasn't ever recovered. We didn't perform any major website changes during or immediately prior to that time period, and the website in general hasn't been negatively affected by Hummingbird. All locations for the business are optimized in the exact same way and according to best practices; there's no significant difference in the number of local listings, reviews, G+ fans, social signals, etc across locations. All meta data and content is optimized, NAPs are all consistent, we've built links wherever we can: the SEO for every location has been by-the-books. We've run a competitor audit in this particular city that included pulling our top competitors and exploring their domain authority, meta data, on-page keyword grade for the term we're trying to rank for, number and type of inbound links, social signals, and more; and we didn't spot any patterns or any websites that were significantly outperforming us in any area (besides actual rankings). It's frustrating because the client is expecting a fix for this city and I can't find anything that needs to be fixed! Have any multi-local SEOs out there run into a similar problem? What did you do about it?
Local Website Optimization | | ApogeeResults0 -
How slow can a website be, but still be ok for visitors and seo?
Hello to all, my site http://www.allspecialtybuildings.com is a barn construction site. Our visitors are usually local. I am worried about page speed. I have been using Google Page Insight, and Gtmetrix. Although I cannot figure out browser leveraging, I have a 79 / 93 google score and for gtmetrix 98/87 score. Load times vary between 2.13 secs to 2.54 secs What is acceptable? I want to make sure I get Google love for a decent page speed, but for me these times are great. Bad times are like 7 seconds and higher. I have thought about a CDN, yet I have read horror stories too. I have ZERO idea of how to use a CDN, or if I need it. I just want a fast site that is both user and Google speed friendly. So my question is, what is a slow speed for a website? Is under 3 seconds considered ok? or bad for seo? But any advice is greatly appreciated.
Local Website Optimization | | asbchris0