Welcome to the Q&A Forum

Hamid - the fluctuation is not necessarily coming from something you did wrong. The main takeaway from Rand's article is to consider the PA/DA as a relative measure - to see how you score compared to your competitors (see quote below). It tries to predict how well your site is doing in the search results - but (as all tools on the market) it will never be perfect.

"My strongest suggestion if you ever have the concern/question "Why did my PA/DA drop?!" is to always compare against a set of competing sites/pages. If most of your competitors fell as well, it's more likely related to relative scaling or crawl biasing issues, not to anything you've done. Remember that DA/PA are relative metrics, not absolute! That means you can be improving links and rankings and STILL see a falling DA score, but, due to how DA is scaled, the score in aggregate may be better predictive of Google's rankings."

There was an update of the Moz index yesterday - so it's not really abnormal that there are some fluctuations. If you want to read more about that - check this post from Rand https://moz.com/community/q/da-pa-fluctuations-how-to-interpret-apply-understand-these-ml-based-scores which explains quite well what's happening.

Dirk

Sorry Peter - I was first - the reward is mine

The site did exist before - check https://web.archive.org/web/20141117163048/http://www.(your domain)/ - so quite possible it had a manual action (if the type of content was as low quality then as it is now) erasing it from the index.

It's a very thin affiliate site with 0% original content (all content = Amazon). On top of that - its quite heavy to load, has no optimisation whatsoever (H1/Meta/..etc); several elements on page that return 404 status, low pagespeed scores and as it is new, no incoming links.

You could check the logs - it's quite possible that Googlebot hasn't discovered the site yet. If it has visited, it probably considered the site too low quality to index. If not visited, you could register in Search Console and do a "fetch like Google".
It will probably put some pages in the index - but there is no chance that with the current content this site is going to rank.

Dirk

Bonus: You indicate that you don't have problems with 404's - but your site isn't exactly clean concerning the internal links. Can't crawl using Screaming Frog given the 503 I get - but manually checking few pages it seems you have mal formatted links:

Classic Car Insurance (the .com is missing) - it's not a 404 but DNS lookup error

I would do a full scan with Screaming Frog to check what other technical issues you have on the site.

Just added the screen copy from the Pagespeed Insights result for sake of completeness.

06WdsNk

Difficult to check - given the very strict firewall rules you apply (I get a 503 out of Belgium : Reason: Access from your area has been temporarily limited for security reasons)

I assume you are sure that your firewall is allowing Googlebot to access your site? (the sentence "For example, if you were blocked because it was detected that you are a fake Google crawler, then disable the rule that blocks fake google crawlers" scares me a bit)

When I use Pagespeed Insights I get the message that they are seeing a 503 as well (and this is Google after all). You might want to check your logfiles to see that Googlebot is indeed able to crawl your pages (with all the IP addresses it is using)

Other than the config of your firewall - you might want to check loadtimes - according Webpagetest you page takes 25sec to load so there is probably some room for improvement. (I know from other cases where very slow sites were the main reason for indexing issues)

Let me know how to get the reward
(if one of these reasons is causing the indexing issues)

Dirk

I don't fully agree with the answer above - having 404's is not a direct cause for demoting. Check https://googlewebmastercentral.blogspot.be/2011/05/do-404s-hurt-my-site.html . Another interesting video on expired e-commerce products is here: http://searchengineland.com/googles-matt-cutts-seo-advice-unavailable-e-commerce-products-186882. Redirecting to the homepage (or any other non related page) is certainly not a good thing - this would be considered a soft 404.

It is however best practice to redirect products that remain in the catalog to the newer versions of the page. Even for those that don't remain in catalog you might consider to redirect them to a closely related new product. You indicate that given the number of products this would be not feasible. Are there no common patterns in the url to redirect these pages in bulk? Is it possible to put the redirect in the page header?

Redirects (properly implemented = going to the corresponding content) are both good for users & Google. If you don't redirect it will take some time before the new products are indexed and replacing the old ones in the ranking - which will almost certainly lead to rank loss: the 404 results will quickly disappear from the results and it will take a lot longer for the new ones to take their positions, it they manage to get back to these positions at all.

If you happen to have links point to these old product pages - you will loose these as well if they go to a 404.

In case that redirects is indeed not a valid option - make sure that you have a custom 404 pages so that user can easily find what their looking for. It might reduce the damage a bit.

If it was my site  - I would try to find a solution to implement 301's.

Dirk

I do understand that you want to keep a copy of your old site - but don't really understand why you allow this site to be indexed by Google? (robots.txt allows full indexing).

The reason you kept it seems a bit off - redirects are made on domainnames. So keeping the site accessible under the IP address will not help you to check if the redirects are properly working.

Why would you need to install Wordpress on the old server - redirects are made in the htacess file. Absolutely no need to do this via Wordpress. Check here or just google it (if you also changed the url's the redirect links might be more complicated than the example behind the link).

Duplicate content - not too difficult to spot:

Examples:

https://www.cheaptubes.com/product-category/industrial-carbon-nanotubes-products/conductive-nanotubes-composite-additive/
https://www.cheaptubes.com/product-category/industrial-grade-carbon-nanotubes/conductive-nanotubes-composite-additive/

seem quite similar to me.

Some pages are also a bit strange - check this url

https://www.cheaptubes.com/product-category/functionalized-carbon-nanotubes/www.cheaptubes.com/product-category/functionalized-carbon-nanotubes/fluorine-functionalized-carbon-nanotubes/ - link is coming from https://www.cheaptubes.com/product-category/functionalized-carbon-nanotubes/ - I guess caused by a wrong formatted relative link.

Try running Screaming Frog to check these issues - it's an incredible tool to help you with issues like this.

Edit: you use canonicals on the strange url's above so they shouldn't really count as duplicates

Dirk

A canonical url always applies to a "full page" = content accessible under a specific url so it's not possible to apply the canonical url to only a part of the page.

Duplication of content is not always a reason for "punishment" by Google - check https://support.google.com/webmasters/answer/66359?hl=en : "unless it appears that the intent of the duplicate content is to be deceptive and manipulate search engine results" that you will get an action from Google. In most cases - "we (=Google) do a good job of choosing a version of the content to show in our search results".

You could consider to make one main page for each type of product and then list the specifics for each model number on the same page but it could lead to lower traffic.

You could also try to make for each detailed page a piece of "original" content, but in cases like yours I guess this would be virtually impossible.

Don't think everyone would agree on this advice - but given that your pages are currently ranking quite well and they are answering a certain customer need I wouldn't do touch them. The risk you run seems limited.

Dirk

To quote Google: Duplicate content generally refers to substantive blocks of content within or across domains that either completely match other content or are appreciably similar.

Having one image on two different posts - even when alt & title are identical will hardly qualify as duplicate content. A lot of sites (even well known publishers) are reusing the same images on their pages - if you're talking about stock photos - these are used on millions of sites. You won't be able to rank them for Image Search - but they won't hurt you either.

Dirk

Was a bit to quick in my first answer - on first sight it looks ok - but you messed up the canonicals.

All variations of http://www.key.co.uk/en/key/plastic-storage-boxes#orderBy:5&pageView:list& have a canonical url http://www.key.co.uk/en/key/plastic-storage-boxes - which is great as it's strips away the parameters. It also has a rel next pointing to http://www.key.co.uk/en/key/plastic-storage-boxes?page=2 - which is fine as well.

However - the second page (if you start from the ugly url http://www.key.co.uk/en/key/plastic-storage-boxes#orderBy:5&pageView:list&) has url http://www.key.co.uk/en/key/plastic-storage-boxes#productBeginIndex:30&orderBy:5&pageView:list& - this page has a canonical url http://www.key.co.uk/en/key/plastic-storage-boxes - in fact that page should have the canonical url http://www.key.co.uk/en/key/plastic-storage-boxes?page=2

The coding on the "clean" (=canonical) versions of your url are fine. Check https://support.google.com/webmasters/answer/1663744?hl=en - there is an example listed which is quite similar to your configuration.

Dirk

Don't know if you did changes in the mean time - but if I check the url you seem to do it ok:

• you use canonical url's to strip the parameters from the faceted navigation
• you use rel next/previous to navigate between pages

The second url you mentioned will probably not appear in the index due to the canonical.

Dirk

Great work guys!

Would be nice if Moz would include this info  yearly stats (https://moz.com/blog/the-best-of-2015) - activity in the Q&A section is now completely missing in this overview.

Comparing with total Moz points isn't always the best indicator:

• thumbs up vs total comments makes more sense than thumbs up vs total Moz points
• good/endorsed answers vs total questions answered (idem - don't think that this info is easy to scrape)

As measure for activity in general:

• PPM for the last 12 months (rather than beginning of subscription - little personal bias as I only became active last year - while I created my account in 2008)

Dirk

Even if you implemented all the tags correctly there is no guarantee that Google will show the them in the SERP's - the choice whether or not to show them is entirely up to them.

I guess you already checked if all the tags are properly implemented (if not - check here or here)
You could also check this page for the usage guidelines (bottom of the page).

Again - even if you are doing everything correct & you meet the guidelines - it's entirely up to Google if the snippets will be shown or not. You can't control it.

Dirk

The best way to deal with this is using canonical url's. Choose one of the 2 pages are "main" page - use a canonical on the second page to point to the "main" version.

Example: yourdomain.com/category1/article.htm => main page
yourdomain.com/category2/article.htm -> in the head section of this article put:

This way Google will know it only has to index yourdomain.com/category1/article.htm.
To learn more about canonicals: 
https://support.google.com/webmasters/answer/139066?hl=en

https://moz.com/learn/seo/canonicalization

Dirk

You don't have to put index & follow - that the default behaviour. It's only when you don't want to be indexed or links to be followed that you have to indicate it.

Rel next/previous have to be in the header - check https://support.google.com/webmasters/answer/1663744?hl=en  - not on the links themselves.

Dirk

Personal experience - redirecting from subdomain to main domain shouldn't have a big impact (maybe some temporary fluctuations).
Changing the design can have a huge impact. We did a relaunch recently where both url / ux where changed and we had a massive drop of traffic - even when all url's where properly redirected. We did notice however pageviews/visit go down & bounce rate increase which indicated that our visitors had issues with the new (responsive) design. It took us about 4/5 months to return to previous traffic levels - but we still haven't recovered some of our top positions for very competitive keywords.

Dirk

Are you using Cloudflare? I know Moz used to have issues with https & cloudflare - not sure if these have been solved.
You might want to check these threads as well here on Moz:

https://moz.com/community/q/804-https-ssl-error and this one https://moz.com/community/q/error-code-804-https-ssl-error-encountered

If this doesn't help - contact help@moz.com.

Dirk