Welcome to the Q&A Forum

I'd say that 5kb is definately worth it! Even on bandwidth alone, if you get that image downloaded 10 times that's 50kb in bandwidth saved.

The answer shouldn't be "what does it take to get a green light" it should be "what's the quickest we can get it for our user". The ideal speed should be under two seconds to load a page.

Personally I think a 301 redirect would be the best.

They've had a few posts on some domain forums:

http://www.domainnameforum.org/showthread.php?t=97639

https://www.namepros.com/threads/problem-with-tradenames-com.755117/

Just got this status page email: http://health.moz.com/incidents/p9q8yk49kh25

However, the canonical is only an advisory tag. I've had few cases where people have relied on their canonical tag when their site has numerous product url types (as above with category in the url and just product url) which has many references to these different urls elsewhere (onsite and offsite) and they are now indexed as both versions, which is not always ideal. It also means that reporting tools such as Screaming Frog only show the true URLs on the site. It's also saving crawl budget as it doesn't have to crawl the category produced url and the canonical url.

Whilst it's not a major issue, it's something I would look at changing.

A tool that I use for some reporting is: https://agencyanalytics.com/ however I only use this with some automated reports for customers who don't want to use GA or search console so I'm not sure what their social analytics is like.

They do have a free trial so perhaps worth giving it a go?

What I've suggested will be avoiding these duplicate urls? Here's some actual examples, going via a tier two category I get the following product url:

https://www.symectech.com/epos-systems/customer-displays/pole-mounting-kit-94591.html

With a canonical of:

https://www.symectech.com/pole-mounting-kit-94614.html

Yet when going from https://www.symectech.com/epos-systems/?limit=32&p=2 (a tier 1 category) I get the canonical url.

So if there are products listed in multiple tier two categories then that's multiple urls for the same product. With the suggestion I made, there would only be one variation of this product url (the canonical)

Just going through Laura's list as a checklist for ones that are applicable:

• Have you checked your robots.txt file or page-level meta robots tag to see if you are blocking or noindexing anything?

Nothing that I can see, that's causing a major issue.

• Is it a large site? If so, check for issues that may affect crawl budget.

The main thing I can see is that the product urls and canonicals are different, is there anyway of listing the product urls as their canonical versions in the category?

Just incase you haven't seen, I've added a link to a useful Yoast guide for amp seo in wordpress: https://yoast.com/setting-up-wordpress-for-amp-accelerated-mobile-pages/

This is a plugin that you may be interested in: https://wordpress.org/plugins/amp/

Another post to add, https://yoast.com/setting-up-wordpress-for-amp-accelerated-mobile-pages/

Absolutely

Hi Rachel,

Glad the resources were helpful, re-reading my response, I did forget to add that you can also specify your target here: https://support.google.com/webmasters/answer/62399?hl=en

Best of luck with the multilingual site, anymore questions please do ask.

Tom

Just a thought, is it possible to get this video in a text format too? I've worked in offices that don't allow speakers/headphones, so there is no way of watching this video. So you'll miss out on the information you've provided.

I'm not sure if this is of any help to you? https://moz.com/community/q/without-robots-txt-no-crawling

Are you able to post the domain here? I'd be happy to take quick look around it for you.

Just to expand on this, I found the following quote from a Moz member of staff:

"I wish I could give you IP addresses, but they change all the time since we host Roger in the cloud. There's not even a reliable range of IPs to give you. You can totally whitelist the useragent rogerbot, but that's the only reliable information about the crawler you can go off of. I hope that helps but let me know if there's any other solution you can think of. Thank you! "

From a previous question: https://moz.com/community/q/is-there-a-whitelist-of-the-rogerbot-ip-addresses

Unfortunately this is all I can find: https://moz.com/help/guides/moz-procedures/what-is-rogerbot with no IP range shown.

Not a problem Britney!

Probably not the right place to ask the question, but it's related to what you posted, have you got a source for the below? I would love some more information on it.

• Logos with transparent areas can get tricky if you can only view them over colored backgrounds. Google has been known to decline use of white/transparent images in their Knowledge Graphs (because their background is white, rendering the logo pretty invisible). If this is the case, be sure to have alternative logos with a colored background that you can also use for social logos.

Thanks,

Tom

Hi Andrew,

You can use the hreflang tag, this will tell Google that this is your site in a different language. Useful post for you here: https://hreflang.org/what-is-hreflang-a-primer/

My biggest question though, is how you are translating, will it be human translation or machine translation? See: http://searchengineland.com/how-machine-translation-has-a-habit-of-mangling-multilingual-seo-91759 for a bit more detail.