Welcome to the Q&A Forum

Hi there!

What you see is very common when searching results for a different region.
For example, being in the US and searching for something in UK it highly probable that you´ll get a mix of US and UK results. 
Google does understand the localization of your IP thus will serve you some results for the device's current country.

As far as hreflang tags, having ccTLD is enough for google to understand that your site serves content for that specific country. Also, as you say that 70% is duplicate content, having a correct hreflang configuration helps google to understand even more and gives some more authority to each site. As you are telling google that for a page there are others for other country.

That tags that you said, i think are incorrect just because you are adding the autoreferential hreflang and it needs the other hreflang tags for the other versions of the same page.
More information, here:
Hreflang generator - Aleyda Solis International SEO - Moz Learning Center The Guide to International Website Expansion: Hreflang, ccTLDs, & More! - Moz Blog The International SEO Checklist - Moz Blog

Hope it helps.
Best luck.
GR

Hi there,

Yes, always autogenerated pages with thin content are a risk. As long as you can identify and manage those pages, you have some options:
1- Delete them
2- Noindex
3- Canonical to the correct
4- Correct the issue in the theme

Hope it helps.
Best luck.
GR

Hi Nutric,

Google is not doing anything wrong. You have the canonical tag present in those pages and point to the page without parameters.
Check the images im attaching.

What you should do is to disallow the canonical in for that parameters or create the subfolder for these locations.

Best luck.
GR.

TGjyWDl t5Zqc82

Hi angela,

Those 5 out of 64 URLs.. Is that a report in Search Console? or only 5 URLs appear when searching in Google?
Search Console usually takes a little longer to update its reports on index status.

Have you tried a site: search? Also using _inurl: _parameter.
For example: site:domain.com inurl:/category-noindexed/

Hope it helps.
Best luck.
GR

What sometimes happens is that when some URLs are marked as noindex, googlebot reduces its crawling frequency as they interpret that you really don't want that page to be indexed thus has value or so ever.

What you just did is tell GoogleBot to crawl specifically that page and "force" it to analyze and render that page. So GoogleBot now understood that the noindex is no longer set and that page should be indexed.
I'd wait a few days so that googlebot naturally crawls all your site again and eventually index every page that deserves to be indexed.

If that doesnt happen in about 2 weeks, then there is a tool in the old Search Console, where you can tell GoogleBot to Crawl a single page and its links. That is under Crawl-> Fetch as Google. Request an URL to fetched, after a few minutes it a button: _Request indexing_will appear, there you'll have the option to "Crawl this URL and its direct links". This image might came handy: https://imgur.com/a/y5DbUVw

I'm glad it helped previously and hope the last helps you even more.
Best luck.
GR

You're welcome. 
We as a community are here to help.

If your issue is now fixed, you could mark this question as answered

Best luck.
GR

Hi there!

1. No, at all. There is no issue there, as long as changes do make sense.
2. Noup, there is no such thing as "too much content".
3. Think to Google other of your readers/clients. Wich frequency would be better for them?
4. No, there aren't any negatives as long as you keep the content coherent and don't create duplicate content

Hope it helps.
Best luck.
GR

Hi,

As said below by Robin... I'm suggesting you think about the frequency that would be better for users/readers/clients. In the end, Google is another reader.
Hope it helps.
Best luck.
GR

Hi Risto,

Google said it many times that they are pretty good at detecting bad links thus they encourage us to disavow links only when a manual penalty has hit your site or when you dont trust Google's algorithms and/or do know that those links are really hurting you.

That said, my personal opinion, don't bother at all. 
Also, please keep in mind that private metrics as PA, DA, TF, CF or whatever do not reflect what google sees. Those companies with their algorithms try to simulate what Google understands so don't fully rely on them.

Hope this helps.
Best luck.
GR

Just here to add some to Will almost complete answer:
The 'site:' often shows results that won't be displayed in Google search results and don't represent the entirely nor precisely the pages that are indexed.

I'd suggest to you:
1- If those pages are already serving 404 or 410, then wait for a little. Google won't show them in search results and eventually won't be seen in a site: search. You can check whether those URLs are being shown in searches through search console.
2- There is a script made by a webmaster that helps you using the GSC URL removal tool for a big list of URLs. Please, use it carefully and try it first within a riskless GSC property

Hope it helps.
Best luck.
Gaston

Hi Frankie,

My guess is that the traffic you were losing was because of its traffic driven by /search pages.

The questions you should be asking are:

1. Are those /search pages getting traffic?
2. Are them important to users?
3. After being disallowed, which pages were losing traffic?

As a general rule, Google doesn't want to crawl nor index internal search pages, unless they have some value to users.

On another matter, the crawlability of your product pages can be easily solved with a sitemap file. If you are worried about the size of it, remember that it can contain up to 50k URLs and you can create several sitemaps and list them in a sitemap index.
More info about that here: Split up your large sitemaps - Google Search Console Help

Hope it helps.
Best luck,
Gaston

If you have a site with, at least 30k URLs, looking at only 300 keywords won't reflect the general status of the whole site. If you are looking for a 10% loss in traffic, I'd start by chasing the pages that lost more traffic, then analyzing whether they lost rankings or if there are some other issues.

Another way to find where there is traffic loss is in search Console, looking at keywords that aren't in the top300. There might be a lot to analyze.

It's not a big deal having a lot of pages blocked in robots.txt when what's blocked is correctly blocked. Keep in mind that GSC will flag those pages with warnings as they were previously indexed and now are blocked. That's just how they've set up flags.

Hope it helps.
Best luck.
Gaston

FOR THE LOVE OF GOD DONT REDIRECT  404s TO THE HOME!

This is terrible advice. Doing that you'll turn those 404s into soft 404s, making them more problematic than ever.

Hi there,

I'm considering that you have over 500k URLs, to be worrying about crawl efficiency. If you have less than that, please don't worry.

Having 404s is completely fine, and google will eventually lower their crawl frequency to those pages. 
Blocking them in robots.txt will cause to google stop crawling them, but never to never remove them from the index.
My advice here: don't block them in robots.txt

As Rajesh pointed out, you could force those 404s into 410 to tell Google that they are gone forever. Yet, Google said that they treat 404s and 410s as the same.
John Mueller said over a year ago that 4xx status codes don't incur in crawl wastage. You can check it our in these Webmasters hangout notes - Deepcrawl

Hope it helps,
Best luck.
Gaston

Hi there,

I don't fully understand what's your issue with canonicals as you made it look like the issue is related to hreflang tag.

So, answering your questions:

1. Canonical tags should point to a page that means the same INSIDE the same website. If that's not the case, its a common practice to have self-referential canonical tags.
2. Not having canonical tags at all **will not impact **your rankings nor its ability to rank higher. With this, you are telling Google that you don't know if there is a better page and let google choose for itself what's the best canonical (if any)
3. Having different domains for different countries and even with different letters/words is not an issue at all. You have an advantage having ccTLD domains for each country, this means that Google already understands that your website is targeted to that country. If between both websites, you share content and have a set of pages that mean the same but for different countries, then you should add hreflangs so G gets the idea that both sites relate.

Do these comments answer your questions?

Hope it helps,
Best luck.
Gaston

Hi there,

This is exactly the case where Google recommends to use canonical, on this resource page: Consolidate duplicate URLs - google Search Console Help.

Keep in mind that canonicals are efficient when different URLs have the same content.
I'd avoid redirections because that would be hurt user experiences when navigating the website, and we know that hurting UX upsets Google.

Hope it helps
Best luck.
Gaston