Welcome to the Q&A Forum

Hi,

Generally, people advice to use CDN for serving static content very fast on your site, that helps you improve your page speed and hence, there it adds value from SEO & UX perspective.

Now, the definition of CDN will help you understand how exactly it can help you improve your page speed. So, here you go:

"A CDN is essentially a network of geographically dispersed servers. Each CDN node (also called Edge Servers) caches the static content of a site like the images, CSS/JS files and other structural components. The majority of an end-user's page load time is spent on retrieving this content, and so it makes sense to provide these "building blocks" of a site in as many server nodes as possible, distributed throughout the world.

When a user requests your site, the node closest in proximity to user will deliver the static content, ensuring the shortest distance for the data to travel (reduced latency), therefore providing the fastest site experience."

Also, check the attached image to understand it well. Hope this helps

9tiFzvG.png

Well said Brian. The solution I suggested was for a scalable system which is not dependent on the number of product. IMHO, one shouldn't think about 5-15 products while working on a project like URL structure even if you're planning to not scale that much.

P.S Thinking about a scalable solution won't hurt you anyways, it'll rather add value if you're planning to scale at any time.

Hi SalSantaCruz,

I would like to add something on top of what John Cross mentioned here. I would suggest the following URL for this particular use case:

www.salshoes.com/shoes/canvas, and will keep "Men" & "Women" as on page filters.

Reason: You'll probably be looking to target this page for "canvas shoes" which is high volume search term than "canvas shoes for men" or "canvas shoes for women" which can be considered as secondary keywords for this page. Now, having "Men" and "Women" as filters on this page, helps you avoiding cannibalization issues as well i.e the situation when many of your pages (www.salshoes.com/men/shoes/canvas, www.salshoes.com/women/shoes/canvas, www.salshoes.com/kids/shoes/canvas etc.) will be fighting for a click of a user searching for "canvas shoes".

P.S its always good to think about site architecture and all the edge cases while thinking about such meta level things. Here, considering "Men" and "Women" as a category or a filter matters a lot.

Hope this helps. Cheers!

Hey,

This would be a nice experiment I would say. Give it a shot and please do share the results here. Looking forward.

IMO, linking from the external sources or inter-linking adds a lot of value from SEO perspective. Here, you're planning to not even consider these things for ranking. The SEO game has changed completely in last few years and today, its very difficult to rank in search engines without optimizing for them at all.

I don't know how exactly you're planning to promote this content, whether you're incorporating UGC in your product or your basic SEO elements are optimized (meta tags, header tags, url structure etc.) while doing this experiment. But trust me, this is going to be a very very tough game. Good luck for it.

I'll second Moosa here, very nicely explained.

Hi mrodriguez1440,

That's a very valid concern. Well, I think the primary reason behind this whole change is google's algorithm updates. Google has become very very smart these days and just pushing content might not get you good ranking these days.

And, may be when you published that content which is not a very common topic for the bloggers, there might have not enough decent quality content on the web (I'm not sure, just thinking aloud).

Well, 90% is way too high. As Umar mentioned here, you really need to make a content strategy first. Do a full fledge keyword & trends research before investing any time on writing the article.

Hi Ankit,

Though I have checked for the pages you're serving to bots, could you please have a look at your .htaccess file once? Does it contains something like:

RewriteEngine On 
RewriteCond %{HTTP_USER_AGENT} (google|yahoo) [OR] 
RewriteCond %{HTTP_REFERER} (google|aol|yahoo)

Do you have your code's copy in github or bitbucket or any other source code management tool? If yes, please scan last few commits thoroughly.

You can create a list of plugins installed recently. Remove them one by one and submit your home page URL to GWT for fetching a fresh copy it every time. Not sure what's the issue here, let's do hit-and-trial to deep dive a bit.

That's what. Not able to find any good information to go next-step for this. But, still checking random things with a "hope".

Well, this is probably 1 of the most interesting issues an SEO can come across with. Google is showing different cached version in different countries. For me, that's strange too. Is that usual thing?

That's right, its showing different cached versions in different countries. Just checked for US here. Screenshot attached.

55ymGkF.jpg

Ping and traceroute worked for me when I tried using my terminal (screenshot is attached).

Well, I agree that the problem is actually bigger. If you see its cached version on google, it was last cached on 16th Aug i.e after the issue of index.php/index.html was fixed by the admin (another screenshot attached).

I tried to see this page as googlebot as well, couldn't find the issue (wanted to check it for cloaking as well).

0xR3w7s.jpg r8i8Chi.jpg

Hi Keith,

I'm just curious, why would you like to fetch those element again & again? You have the list of your competitors (hope they aren't more than 4-5) and you can simply browse their product and pick a sample set of their pages, which are mapped to your pages for any given query. Now, do a on-page grade test for both the pages and you'll be able to compare them on each an every on-page element.

Assuming, each and every page on their (or yours for that matter) won't have absolutely unique template and hence, on-page grade report of any given page of one kind, would be more or less the same for other pages from that set.

Here's the task for you, pick he various types of pages you have for your site, take some sample instances (even one would be enough too, I guess) of each type. Now, pick the mapped landing pages on competitors sites and do the comparisons.

P.S Its an interesting requirement and I was thinking about building some interesting tool like this, thanks for your suggestion. I would love to work on this, if time permits

Cheers!

Well, not sure about how exactly you'll be putting these links on your product and how well bots will be able to crawl them. AFAIK, bots are smart enough to understand JS these days but upto some extent.

Anyways, if you're planning to build such awesome tool (as you're trying to explain), you should definitely go for it for the better UX. I hope you'll be having nice and useful content on your listing pages anyways (in well crawlable format). So, the problem is about inter-linking/exposing them to bots, right? XML sitemap, HTML sitemap, dynamic footer etc. can help you with that

Check the following, may be it'll help you resolve the issue:

https://moz.com/community/q/de-indexed-homepage-in-google-very-confusing

https://moz.com/community/q/site-de-indexed-except-for-homepage

Hey Chris,

Thank you so much, didn't notice that tiny link. Anyways, have shared the feedback with them now. I hope this will soon get fixed.

That's really strange. Could you please share the screenshot when you're trying to fetch it as google in the GWT?

I have the same doubt, when I search for "askme" which is the company I work for, its showing the wrong wiki page link which talks about a female trio from Chicago Illinois (PFA the screenshot).

How can I tell google to not link that wiki page with our brand?

J80ECKC.png

Nice answer Jane. This probably answered many questions (that I had in my mind before reading your answer) related to the best URL I can have for my website. Cheers!

I'm working on URL structure for the same business model. Loved your answer Jane

While thinking about the ideal URL structure, I did consider some of the blogs (including this one by Rand: https://moz.com/blog/15-seo-best-practices-for-structuring-urls, check point #11. Attaching a screenshot as well) and websites which were doing really good with their one level static URLs.

I actually did some keyword research on user's search pattern and google suggest data. Generally, our target search term comes before ("pizza huts" in this case) the geo location, may be people search things in a different way in India. Hence, I thought of keeping the URL structure that way.

A little confused about this point though "URL, breadcrumb both should match the sequence.  If one has one sequence, and the other has a different sequence, that confuses search algorithms". Because, have seen many website doing tremendously well who're not following these principles.

U2vC1Ua.png

Thank you so much once again Sir Alan.

Well, I'm just thinking aloud here. How about putting my primary keyword in the first level instead of having this well structured URL syntax? For instance:

• www.askme.com/pizza-huts-in-saket-delhi instead of www.askme.com/dehli/saket/pizza/pizza-hut/

Here,

• The complete primary keyword (or target search string) is closer to the domain. "Closer your keywords to the domain, better it is", I heard this somewhere. Is it still true and adds any additional value?
• We don't have deep URL directory structure and our primary keyword is together too. In the well structure URL (the one you suggested), the target keyword is broken into multiple pieces & the URL directories.
• But, I'm not exposing the hierarchy/navigation-flow via URL. I hope that's okay as far as I'm handling it cleanly from the breadcrumbs and rich snippets. What's your take on this?

I know there are chances of URL conflicts. For instance, if we have an area "foo" in the city "bar" vs a city "foo bar". I'll end up having the same URL for both the cases i.e /<search-query>-in-foo-bar. There are many such edge cases, I'm on it.</search-query>

Hi Mozzers,

I'm working at AskMe.com which is a local search engine in India i.e if you're standing somewhere & looking for the pizza joints nearby, we pick your current location and share the list of pizza outlets nearby along with ratings, reviews etc. about these outlets.

Right now, our URL structure looks like www.askme.com/delhi/pizza-outlets for the city specific category pages (here, "Delhi" is the city name and "Pizza Outlets" is the category) and www.askme.com/delhi/pizza-outlets/in/saket for a category page in a particular area (here "Saket") in a city. The URL looks a little different if you're searching for something which is not a category (or not mapped to a category, in which case we 301 redirect you to the category page), it looks like www.askme.com/delhi/search/pizza-huts/in/saket if you're searching for pizza huts in Saket, Delhi as "pizza huts" is neither a category nor its mapped to any category. We're also dealing in ads & deals along with our very own e-commerce brand AskMeBazaar.com to make the better user experience and one stop shop for our customers.

Now, we're working on URL restructure project and my question to you all SEO rockstars is, what can be the best possible URL structure we can have? Assume, we have kick-ass developers who can manage any given URL structure at backend.

Thank you so much Sir Alan. I really appreciate your efforts for compiling this detailed response to my questions. Have noted down all the points along with how better I can handle them, will soon come up with a better fat footer.

I have a very meta level question. Well, I was working on dynamic footer for the website: http://www.askme.com/, you can check the same in the footer. Now, if you refresh this page and check the content, you'll be able to see a different combination of the links in every section. I'm calling it a dynamic footer here, as the values are absolutely dynamic in this case.

**Why are we doing this? **For every section in the footer, we have X number of links, but we can show only 25 links in each section. Here, the value of X can be greater than 25 as well (let's say X=50). So, I'm randomizing the list of entries I have for a section and then picking 25 elements from it i.e random 25 elements from the list of entries every time you're refreshing the page.

Benefits from SEO perspective? This will help me exposing all the URLs to bots (in multiple crawls) and will add page freshness element as well.

**What's the problem, if it is? **I'm wondering how bots will treat this as, at any time bot might see us showing different content to bots and something else to users. Will bot consider this as cloaking (a black hat technique)? Or, bots won't consider it as a black hat technique as I'm refreshing the data every single time, even if its bot who's hitting me consecutively twice to understand what I'm doing.