Welcome to the Q&A Forum

Hi all,

I just noticed a potential issue with our websites. We have two ecommerce websites, one is a very large one selling all sorts of products, while the microsite focuses on a small segment of products. All products sold on the microsite are also sold on the main website. In the beginning of September, we upgraded the microsite to the same script that the main website uses to make it mobile friendly and update the design. They now look very similar.

Before, both websites used to rank on page 1 for a specific keyword. I have noticed that since we upgraded the microsite, the two websites have been taking turns ranking for the keyword. For a few weeks the microsite ranks and the main website doesn't rank for the keyword. Then for a few weeks only the main website ranks and the microsite doesn't.

I think the reason this is happening is that Google understands that the content is the same and the websites are both owned by the same company. Fair enough.

1. I remember reading an article about this phenomenon before but can't remember where. Does anyone know which article I'm talking about (it would have been on an SEO blog/website, e.g. Moz, SEJ, SE Roundtable etc)? I'm not even sure what this phenomenon is called.

2. If we can only have one of the pages rank, we would prefer it to be the microsite at all times. Would a canonical tag on the main website referring to the URL on the microsite fix this?

3. I think at the moment the product descriptions are either very similar or identical. Would it help to make them more different to get both to rank again if that is what we wanted to do? In the end it is still the same product being sold by the same company - after Google has already sort of merged the two, would they "un-merge" them if the content was more different?

Thanks in advance!

In your case it may be worth it to do this: Keep the pages accessible, show that the product is no longer in stock and then show similar products the user may be interested in instead.

If this is not possible (too complicated, site too big), I would 301 redirect to the closest category page, maybe with a notification at the top that the product is no longer available and that is why the user was redirected to the category page.

Have a look at this blog post: https://econsultancy.com/blog/65562-seo-considerations-for-discontinued-products-in-ecommerce/

I agree. It's more important for your text to be natural sounding than to squeeze in every keyword. Use topics rather than exact match keywords in your copy and make sure it reads naturally so your users aren't put off. Guy #2 is right, guy #1 is potentially hurting your performance and CTR.

I do believe I've read somewhere that Bing still places more emphasis on exact match keywords.

It's worth changing it in my opinion. When I joined my company, they also didn't use H1 tags and essentially used Hx tags as styling tags (which they are not). Even just all the basic on-page SEO has had a very noticeable effect on traffic.

That said, if you are working to a deadline, I would say you have to weigh your priorities. Not using correct Hx structure won't kill your SEO, it's just a missed opportunity. So do it, but if it has to wait a while, it's not going to be a massive issue.

Changing the styling should be fairly easy. Simply copy the styling used for H2 to H1 and then update your website (template) to use H1 for the main header on each page.

H1 and page title don't have to be absolutely identical but it makes sense for them to be fairly similar. So yes, you can use keyword variations, but if they are too different that might make it look spammy - both need to accurately describe the content of the page.

For optimisation purposes I would only look at the body as you can't really help repetitions in navigation, breadcrumbs etc on an ecommerce website, but Moz is probably counting all occurrences within the page source code but outside of tags (so not page titles or meta tags or alt attributes but all occurrences within the menu whether visible or not - they are visible in the source code).

Hi, I can see a few more. 6 in the text main body + 2 in the menu in the main body + 2 in the sidebar menu + 1 breadcrumb + 1 in the top level navigation + 6 in the opened top level navigation. That's 18.

Oh I see what you mean, I opened some of the results and they are just 301 redirects (presumably set up by your company?). Were these previous URLs used before your change of direction?

I do get some of the same results (like for the www search, e.g. a link to https://www.couponshop.co.uk/category/Jewellery on page when looking through the 13,100 results but they just seem to be one every once in a while among u.couponshop.co.uk redirects.

When did you relaunch your website?

I just did a screenshot of a section of the results, but it does show more results in the actual SERP, see attached image, that's positions 3-10.

gFSScgS

When I do a search for site:couponshop.co.uk, I get 13,100 results. Is that roughly what you would expect to see? For site:www.couponshop.co.uk I get 13,000 results.

ww0KqVS

You can see a list of the reasons you got this spam score here: https://moz.com/researchtools/ose/spam-analysis/flags?subdomain=www.naturesouq.com

Low MozTrust or MozRank Score
The site link profile is not trustworthy.

Ratio of Followed to Nofollowed Domains
The ratio of followed to nofollowed domains linking to this subdomain is outside the normal range of others in our index.

Ratio of Followed to Nofollowed Subdomains
The ratio of followed to nofollowed subdomains linking to this subdomain is outside the normal range of others in our index.

Large Site with Few Links
We found very few sites linking to this site, considering its size.

Small Proportion of Branded Links
Links to this subdomain have low amounts of branded anchor text.

Thin Content
A subset of pages within this subdomain have little content.

Large Number of External Links
A subset of pages within this subdomain has a large number of external links.

External Links in Navigation
There's a large number of external links within sidebars and footers.

How can you lower your spam score? Some might be easier, i.e. removing external links in the navigation. Based on the points listed, it sounds like you might have some issues with your backlink profile, on-site links and (thin) content.

Google suggests you should, not for Penguin but for other reasons: https://www.seroundtable.com/google-disavow-penguin-use-22798.html

@ Sean: Have a look at this article. Google now devalues spammy links instead of demoting the website.

Depends on the attribution model and number of channels involved in the conversion, for example when using the linear attribution model, each channel involved in the checkout process will get equal "credit" for the conversion. So if 4 channels were involved, each channel would get 25% which is 0.25 conversions counted for that conversion while if only 2 channels were involved each would get 50% (so 0.5). That's what the weighted number of conversions refers to.

This article helps explain how much credit is given depending on the attribution model you choose. The percentage also depends on the number of channels receiving credit for 1 conversion (i.e. how much is it diluted).

Yeah, I'd do the same. Another option would be (if it is your video) to re-upload the video to YouTube, that way it gets a new very recent date.

It was mentioned yesterday on SE Roundtable, seems that Google are aware of it, see here.

Hi, someone posted about this on Moz Q&A the other day and somebody else suggested it was to do with YouTube videos embedded on the affected pages. See this link.

Are you looking at the right property? I get this when looking at the property for the non-www version of our domain, I do get data when looking at the www version, though, which is our preferred domain.

It needs to open with <link... as="" it="" is="" a="" <link="">tag and rel and href are attributes.</link...>

I don't think it makes any difference if you put  or . Based on my understanding of HTML, the order of attributes doesn't make a difference but the first option seems to be more commonly used (rel attribute first, href attribute second).

So your second example is wrong as it would mean it is a <href>tag (which doesn't exist).</href>

I wouldn't necessarily agree with that, probably depends on your search term. For SEO as a search term here in the UK I am getting 6 with www., 2 without and 1 using another subdomain. For "us election" all but one use www (the one that doesn't uses another subdomain). I am not aware of any preference by Google, however, you should be consistent in your use (in canonical tags, sitemap etc), and make sure the one you don't opt for 301 redirects to the other.

Have a look at this blog post here: https://moz.com/blog/keyword-explorer-upgrades-new-features-more-lists-and-better-metrics It says KWE has moderate coverage for languages of western countries. So that's probably the issue. Depending on topic, I also get a lot of keywords with "No data" and I'm in the UK where coverage is significant.

You can define the locale if you upload the keywords using a list format, but I am not sure, KWE already supports Austrian volume data. See here for how to upload the list using a .csv file (see option 2, you need to create a column for keywords and then one next to it for the locale - de-at): https://moz.com/help/guides/keyword-explorer#frequently-asked-questions