Welcome to the Q&A Forum

Even if you disable JS the cited elements are there, so it must be a Moz crawler bug!

Oh dear what a kerfuffle. People are always searching for new negative-SEO assault deployments!

In February 2018 Moz actually covered the subject of cross-domain canonical tag usage:

• https://moz.com/blog/cross-domain-rel-canonical-seo-value-cross-posted-content

Apparently it can work but only under very specific circumstances:

• The content of the 'canonicaling' URL matches the canonical destination
• The headline of the 'canonicaling' URL's content matches that of the canonical destination
• In-Content Links of the 'canonicaling' URL match those of the canonical destination

So unless the content on their pages happens to be almost identical to the content on the page on your site which the tags are linking to; they're pretty much boneheads and this isn't a major concern.

In all likelihood nothing bad will happen but you could try using Google's Disavow tool (https://www.google.com/webmasters/tools/disavow-links-main) to send them a signal that you do not wish to be 'associated' with the spam-site in question. It is unknown (or at least I haven't heard) whether the disavow tool covers canonical links. I would think that it would at least prompt Google to fundamentally disassociate the two sites.

More than likely those behind the attack are fully aware that it is extremely unlikely to harm you. They probably have a network of machine-built sites launching the same (fake) attack against thousands of webmasters. My guess is that they are preying on ignorance, hoping that some webmasters will pay them to take the tags off.

DON'T fall for it

You can't do this from a rankings POV but you may be able to do it from a CTR POV, as you specify in your question

I reckon that if there is a way to do this, it's probably this: https://marketingplatform.google.com/intl/en_uk/about/optimize/

What this will do is serve traffic equally from Google, to two different versions of a web-page. The question is, when it does this magic - will it alter the SERP titles on Google's front end? I'm not certain, whether it's quite that advanced - but I'd at least give it a go

If that fails just run some PPC ads as a testbed and use the page title text you have written, as the ad-title (I think there is such a thing). Then you can see which gets better CTR from ads. Since PPC ads and SERP snippets have a broadly similar format (at least the text ones do, for the most part) it could be a good testbed

That looks valid to me. It's possible you may not need "*" at the end of each rule but I can't see it doing any harm either

I might go more like:

User-agent: ia_archiver
Disallow: /*/search/

User-agent: rogerbot
Disallow: /*/search/

^ this would stop all search URLs being indexed, so even if you introduced new search facilities later in other directories - they would 'probably' be caught too (assuming that is your intention, assuming they were still in /search/ subdirs)

Don't think what you have done is wrong though.

Always check using Google's robots.txt tester to be safe. Just put your rules into the tester (altering them to be used for all user-agents), and try out some different URL patterns. When it works as you like, update your real robots.txt file (remembering of course, to restore your rogerbot / alexa UA targeting - if you don't want the rules to also apply to Google!)

I think it could be something that someone would buy, but obviously that would need a more detailed investigation. It's great if Moz's metrics are high, but is that universally agreed by other tools? For example is the Ahrefs Domain Rating similar to Moz's Domain Authority score? Does SEMRush see the site as a link toxicity risk? Does Majestic SEO give reasonable Trust Flow and Citation Flow read-outs?

It also depends what the backlinks are like. Certainly it's great if there are a lot of linking domains (and the number of overall links isn't disproportionately different to the number of linking domains). But some of those links might be great, whilst others might be garbage.

You can get three really nice articles (1,000+ words written by a proper PR person, not a link builder!) posted across three really nice sites (with DAs ranging from 65 to 80+) for about £700 GBP / $890 USD. You have to figure that, since those are one-time posts, the value of each containing site must be significantly higher (in the thousands or tens of thousands of dollars)

When prospecting websites as potential link opportunities, I also look into tools like SEMRush or Ahrefs which give estimated traffic figures for those websites (although the traffic estimates are usually only estimates of what those sites take in from Google). I would certainly do the same thing in a lot more detail if I was going to be buying a site rather than a one-off placement. Some tools give even better traffic data for 3rd-party sites which they buy from ISPs and build a front-end around that

If your site could assure a buyer across different platforms (even if some readouts were variable, if they were all relatively high...) then I don't think you'd have too much trouble selling it

Of course some shady black-hat cats might wait for you to get sick of it and delete your own site. They may then buy your expired domain and use something like an advanced scraper to pull the (historic) site files off the 'WayBack Machine' (webarchive.org) and resurrect your website for $20-$40 USD

Be wary of the necromancers of the web

You can use Screaming Frog to do this

Crawl the whole site, then once it's done save the crawl file (File->Save) so you don't lose the project (important, in-case you accidentally export the wrong thing - you can load it back up again!)

Then go: Bulk Export->All Outlinks

That will get you the data you need. It will list all the links on the site, including links to resources (as long as your crawl settings are right, it can be somewhat of a learning curve - if for example you neglect to even crawl pages blocked by Robots.txt, if you neglect to crawl links inside of CSS or if you don't use the JS crawling method correctly you might miss stuff)

The spreadsheet will contain loads of columns containing data on the links. One will show you the referring page / source URL, the other will show you the link target / destination. On the link destination column, filter by your desired URL (which in this case would contain /image-name.jpg or similar)

Once that's done, on the left you'll see all the referring URLs. Sorted!

If the data which is created by Screaming Frog is too large to be analysed in Excel you can export to CSV instead of XLSX. You can then proceed to:

a) Upload it to a local MySQL database for analysis with MySQL queries

b) Use a big-data CSV manipulation tool, like Delimit instead to narrow the data down. Once it's narrow, re-save and then analyse in Excel

---

Problem solved!

Ah man I love stuff like this!

Ok so here's your post: https://www.hackerearth.com/blog/artificial-intelligence/artificial-intelligence-101-how-to-get-started/

Now I can see you think this post only has four inbound links and you have drawn that conclusion from Moz Link Explorer. By the way I think Moz's tool is great. My agency wouldn't have a subscription otherwise! BUT - MLE (Moz Link Explorer) doesn't have a complete picture of all links on the web. Even Google doesn't have this (yet - their MO is to 'index the web' though).

There are several platforms which aim to supply 'backlink' (external links) data. Moz is one platform but there are others. All of these platforms or link-explorer tools, are coded to crawl the web in different ways. They all have their own unique philosophy on how to index the the 'important' links on the web more quickly. I'm not going to get into whether certain tools are better or worse, really you need **a few of them **so that you can aggregate and de-dupe your own data!

From Moz we can see that these URLs are linking to your post:

• http://blog.crescenttechnologyconsultants.com/artificial-intelligence-101-how-to-get-started/
• https://blog.steppingblocks.com/the-tools-that-saved-me-thousands
• http://analyticstraining.com/the-best-way-to-learn-ai-and-ml/
• https://analyticstraining.com/the-best-way-to-learn-ai-and-ml/

Really that's only three pages as the last two entries there are the same page being served via different protocols (HTTP vs HTTPS).

What do other tools have to say about your post?

Well I can tell you that another tools (Ahrefs) claims to have found more live links for this particular URL. That's not to say that it always will find more links overall, but in this instance it has.

Ahrefs believes that you have 10 live links (more than 4) from 8 domains.

I'll list the domains for your reference:

• refind.com
• steppingblocks.com
• analyticstraining.com
• innovify.com
• snyxius.com
• techgenies.com
• grokbase.com
• malware.xyz

In particular "refind.com" and "grokbase.com" have pretty high SEO authority metrics! These two were not picked up by MLE but will invariably have been factored by Google.

This is the link from "refind.com": https://refind.com/TinaTran_au/collections/ai - it looks pretty weird! If you go to the base-level homepage (https://refind.com/) you can see that the site is an effective content curation network with social aspects. It's likely that when your post was added by users or editors; at some point it featured on the homepage. If so that would have driven a pretty hefty whack of SEO-juice to your post. It may be worth uncovering how this network operates and seeing if you can get in with them, leverage the platform further. Turn fluke into process...

Here's the link from GrokBase: https://grokbase.com/t/gg/django-users/13a438tgw5/inconsistency-in-model-field-with-file-storage-on-amazon-s3 - seems a bit more 'on-point'. But what the heck is GrockBase? Let's go back to the home URL (https://grokbase.com/). Here's what they have to say about themselves: "Grokbase is a growing mailing list archive that facilitates __discovery of discussions and users within and across groups." - it doesn't mean huge amounts to me, maybe it will mean more to you. What ever it is, even though it looks a bit rough and simplistic - it seems to be popular. It **looks **a bit like some kind of Reddit knock-off. Maybe I'm just being ignorant and it was well established on Web 1.0. Do some digging, see what you can find. I wanted to take this further for you by looking in SEMRush (another tool) to get an estimate of the domain's traffic. Sadly it seems to be down or having problems right now.

Another popular link-indexing tool (Majestic) has **no information **on your post, so clearly both Moz and Ahrefs index your site more efficiently in terms of backlinks and the particular sites which you network with.

This was another interesting link found by Ahrefs: https://www.snyxius.com/where-to-begin-with-ai-development/ - Ctrl + F for the anchor text of "different industries and processes" and you'll find the link to your own article. It seems like a contextual link from a well established app / software development company in Austin, Tx. Guess what you wrote resonated.

It's not about volume of links, it's about the **quality **of backlinks. Either your links are high in quality for this post, or you have beaten the system somehow. Either way - congratulations

Hard to tell what to do without seeing specifically where this error is coming up, but it usually has something to do with font sizes being too small on mobile devices. Mobiles have much more densely packed pixels on their screens (images are sharper, the pixels are smaller and more numerous). This means that 12 pixels (font-size) on desktop can look like 5 or 6 pixels on mobile devices (or even smaller)

This is why, with responsive design, many people don't specify a pixel-based font-size. They'll make the font size relative (rather than absolute) somehow, be that by calculating pixel width against screen width as some percentage, or by working with newer font-sizing specifications (there are many, and many ways to use them). It's all about inflating font-sizes on devices with more densely packed pixel screens, so that the fonts don't come out looking minuscule

Sometimes you can get errors where, even though the site's design is responsive, as someone was writing text in the CMS text editor - it appended styling info including the px font-size, which is then inlined (overruling all your perfectly thought out font-sizing CSS rules)

If it's not mobile related at all, it's likely that the font is just generally too small

That's pretty good. One big red flag is when sites have like, 60+ CF and like 13 TF or in the single digits. If you want you can send your site to the email on my profile and I'll have more of a look. It still won't be full depth but should show if you have a dud or not. Up to you though

For broader, larger keywords it's better to have specifically designed content on stand-alone pages that does 'the best job possible' of answering Google's queries. Look up the 10x content initiative

For smaller keywords, they can often survive on blog posts as competition is lower and blog posts are lower-brow content. Lower-tier content can win keywords which no one is really competing over, but the second there's competition - you need to be building pages, resources, tools and content pieces which are 10x better than what Google's current search results supplies

https://moz.com/blog/how-to-create-10x-content-whiteboard-friday

Sure, no problem I'll take a look now!

So **this is your post URL: **<a>https://www.hackerearth.com/recruit/resources/e-books/a-complete-guide-to-talent-assessment-software/</a>

This time Moz Link Explorer has only detected one backlink to the post.

The link is from: https://www.yasteq.com/US/Fremont/123161387835128/HackerEarth

The site describes itself thusly: "One of the largest resources on the Internet for finding computer & electronics services and related businesses, worldwide." - so basically it's a local-themed business directory with a tech focus and content curation elements. You can tell that more easily by looking at the homepage (https://www.yasteq.com/). It only has 18 domain authority so that's pretty weak to be honest. The listing is relevant with a reasonable write-up - but it's from a place no one really cares about.

Majestic SEO (another tool) found this link: https://www.eventshigh.com/detail/Bangalore/48e3b2743f568bd46a83bb182fe91ce8-djangothon?src=ecbox whilst **Ahrefs found a separate link from the same site here: **https://www.eventshigh.com/detail/bangalore/63c53417b3eadb328fa3d568d970624c-travel-tech-meetup-with-goibibo

At a glance these just seem to be simple events listings. In most instances, links from these listings are either image-based or have no-follow set to true. Contrary to popular opinion, 'no-following' links does not prevent or discourage users or search engines from following a link (probably one of the worst named attribute values ever). What it does is it prevents Google from transferring any link equity (or 'SEO juice') through the link. The linking page (which hosts the hyperlink) will still lose some SEO authority, but at the receiving end you won't be gaining any (it just gets vented into cyberspace).

Seeing as the page hosting a hyperlink loses SEO equity either way (a measure taken by Google to stop PageRank sculpting from actually 'being a thing') - the only reason to no-follow hyperlinks these days is to avoid outbound Penguin penalties. This is where you are selling hyperlinks but want to tell Google "this link is somehow commercial or advertorial in nature, so don't weight it into your SEO ranking algorithm(s) for Google's results". Google will take that on board and it will prevent the link-host from garnering a penalty (for selling links to others with the intention of manipulating rankings, kind of like being labelled a drug dealer)

The domain actually has very strong metrics in Ahrefs, earning an Ahrefs domain rating of 70. They are probably aware of their site's SEO-clout and either don't want to be giving you a free ride (they expect you to pay for the equity) or they're very cautious about how they link, not wanting to lose the established authority they have built up via penalty / manual action. In any case, though the site is strong and you have links from it - those links are not sending you much / any SEO equity - so there's no party here amigo. Image-based links tend to transfer less equity anyway (even forgetting about 'nofollow') and they have no (user-visible) anchor-text to explain their relevance. Search engines do use image links as signals, but they're not such strong signals (as they're more ambiguous in general)

Ahrefs also found two links from "Mail Pug":

• https://mailpug.com/?p=454
• https://mailpug.com/?paged=2&m=201808

Both have a Domain Rating and URL Rating of zero in Ahrefs. Content doesn't actually seem that bad but... it's from a site with no equity to give you.

In basic terms, your really decent contextual (content-based) links are from sites no one cares about. On the one site with high metrics, they're refusing to transfer SEO equity to you for one reason or another. As such, no rankings for this post.

The reason why it's not ranking well is that although it may be detailed and well written, unlike your other post it didn't attract links. As such it's not deemed very shareable and it has very little SEO authority. You might say it's a vehicle with no fuel.

The reasons why it attracted no 'significant' buzz or attention online could be variable. Was it promoted as well as the other post? Like you suggest - were there CRO or UX defects in the design or copy of the content? I'll leave that side to you.

I wouldn't say there would be massive chances of a penalty here, that being said it's an area where you could be 'adding value' and uniqueness to your pages and you're not doing it. So your pages may be 'less competitive' and you may be missing out on an opportunity. It's more of a competitive missed opportunity than an 'error' per-se

In reality you should have one product page for each product and then just have 'product variants' for stuff like quantity, size, colour etc. On the modern web people find this easier to navigate and since many sites do offer that, they might seem like more competitive places to shop for paint cans than your site. Price does matter, but it's not the sole arbiter of how products are ranked on Google's search engine - other stuff matters too. Unless you have a virtual monopoly on the product (only you can sell it, or only you can sell it at a greatly discounted price due to a special relationship with the supplier) then I would consider the UX and design of your site. No one wants an 'arse-ache' of a browsing experience

Many tools will flag what you are about to do as duplicate content and they're technically right. But instead of going on some crazy copy-writing crusade, think about the architecture of your site. You can still have separate URLs for different product variations if you want, even via parameter-variables (though that's a bit of a 'basic' implementation). If you make it clear to Google through new, more streamlined architecture that they're all actually the same product, the duplicate description(s) won't matter 'as much' (though they'll still be a missed opportunity for more diverse rankings IMO)

You can make it even more apparent to Google that all the different variations are actually the 'same product' by utilising Product schema and some of the deeper stuff like ProductModel which will bind it all together. Whatever you implement, test it here. If this tool throws errors and warnings, keep working away until they're all fixed

Canonical tags are another option but they will decrease your ranking 'footprint' and in this case I wouldn't recommend them, despite 'slight' content duplication risk (which in reality, are mostly negligible)

Final note: you say you have 'unique' descriptions, but remember if they're used elsewhere online they're not unique. If they're unique internally that's great, but if you got them all from a supplier then... obviously loads of other sites are probably using them, which could easily be a big issue for you

So you have! According to Ahrefs the site averages around 1,193 visits (all countries) per month from SEO (Google) specifically. These estimates can be inflated or deflated depending upon how well their keyword index covers your particular website. From Ahrefs you have a 'Domain Rating' of 82 which is really strong.

Looking in SEMRush it's less favourable with the tool suggesting you see between 150 and 300 visits a month from Google in the USA. SEMRush's keyword index (and general handling of data) isn't as good as the stuff you get out of Ahrefs so that's pretty much what I was expecting.

The Ahrefs estimated SEO traffic chart stretches back to 2015 so it seems as though traffic from Google was ascertained in a relatively sustainable way.

I think that you could find a buyer for your site. What they'd be willing to spend would depend upon their current resources and ability to 'keep it going'. Someone who could only just afford the purchase would see diminishing returns as they might lack the ability to invest in future content and community management. To someone with deeper pockets, maybe they could afford to spend over ten grand on it.

Certainly the metrics are pretty decent. Obviously I haven't dissected those metrics in a granular way by conducting some kind of forensic analyses, but the top-line signals are pretty good (in my opinion)

Probably not very, as long as you stuck to their theme and didn't change to the product-related theme, which as you can see (evidently, from Google's results) ranks lower

Some keywords have very obvious intent, e.g: "buy SEAT Ibiza Mk IV Facelift 2016"

Some query-spaces are messy. The keyword can mean multiple things, so multiple audiences (by performing the same search, yet clicking upon radically different sites) are competing over a query-space and making it a bit... Schizophrenic (polarised)

Even in a polarised query-space, Google usually has a preferred thematic relevance for a keyword. Since some programming site is beating Amazon, Google is obviously trying to coerce searchers to accept the programming definition (not the product definition). So you probably could displace that top site, but if you're going to jump from programming relevance to product-based relevance, you may just end up 'another' site underneath (boo!)

Hope that makes sense

That's a really good idea! Cut down what you have to manage to the essentials and then spend more time on those pages. Make sure you do some kind of ranking or traffic audit against all the pages though. You don't want to delete the versions of each page which have some (even if it is small) SEO power. You want to target the ones which Google isn't using

Can you explain the difference between the 'products listings' and the 'actual products themselves'?

You say you still want products and product categories to rank, but not product listings. But to most readers, a product listing is usually a product category or product page (so the info seems to contradict itself, which actually it may not do - just needs more explaining)

This is a really technical, really interesting question. Something to clarify is, what is the exact nature of the chat plugin? When I think of chat plugins, I think of stuff like ZenDesk where, users click a little bubble at the bottom and chat via text with an operator. But you mention phones, right? How is that fitting into it?

I think what you are saying is, you want to know:

When someone converts on your site, shortly after they have 'chatted' (phone / text?) to an advisor, but only when they originally hit your website through Google's Ads (PPC, display etc)

Is that right?

The difficulty would be binding the chat to the actual conversion. What I need to know is, which of these formats is correct:

a) The user clicks a Google ad, the user comes to your site, the user 'chats', the user goes off the chat - back to the site, and then converts

b) The user clicks a Google ad, the user comes to your site, the user 'chats', the user stays on the chat and converts with (via) the chat agent

Depending upon your answer, tracking could be implemented in radically different ways

I don't know of a simple plugin, off-the top of my head that could do what you want. I may not have the answer you need either, but I'm giving it a go!

This means that Screaming Frog is not 'waiting long enough' before returning the time-out error.

Just do this:

1. https://d.pr/i/D7Cj5M.png (screenshot)
2. https://d.pr/i/N6xdLA.png (screenshot)

Raise that number up, until you don't get 0 / Time Out any more. Note that if it does fail a lot on moderate crawl settings, there are likely to be underlying page-speed issues (either that or the machine you are crawling from has bad bandwidth)

It could also be that your crawl ' frequency' is too high, go Configuration->Speed and lower the thread count (to 2-3) and the URI/s (to one or two)

Finally it might be that the SF user-agent is blocked so go Configuration->User Agent and switch it to Chrome

**To help you **- I did the crawl for you, here is your crawl data: https://d.pr/f/a1ux4b.zip (archive of crawl file and some exports)

You can actually use my crawl file as a starting point (by double clicking it) when you want to re-crawl in future. Should be useful to you

It all depends upon how intensely these domains and sub-domains are networked together. If they're very strongly interlinked, Google will likely consider them to be a single website. That brings benefits such as the pooling of SEO authority, but also weaknesses such as the transference or spread of Google penalties. That being said, if one sub domain really doesn't have their sh*t together but other departments have created genuinely useful resources, Google will be loathe to impact useful pages just because 'someone else in the same family' doesn't really know what they're doing.

The largest problem will be lack of vision and operating within the confines of an organisation where, people all have different ideas instead of letting data do the talking. All decisions should be data-led. If the data says, this area of your site is the most important and it also has the biggest opportunities - all efforts should be focused there. If you fundamentally lack the ability to direct all of the things which impact marketing (content, UX, CRO, SEO, design) across all sub-domains with one unified vision, you'll always be limited in terms of the gains you can make. On the other hand within such an organisation, that may be well understood and respected (thus alleviating some tensions)

Domain authority does not come from you, your resources or anything that your organisation does. Authority is a reflection of how popular and useful, the rest of the web finds your websites, pages, resources and content. Editorial (not advertorial or ad-based) links to your live pages from recent, authoritative sources garner your domain (and pages) additional authority. What you're struggling against is the shell of academic culture. It's a sector where, if people do X and teach Y they earn Z. They are happy to be institutionalised (sorry if that offends anyone but it's true) so they seldom consider the impact which the views of others have upon their success

I remember I had a debate with a uni tutor once who got on my case because I complained about their attendance (in terms of being a lecturer). They said academia is not a business and that I shouldn't hold it as a service I was paying for. Further down the line I was unable to finish an essay within the specified time limit and was told "well in the future, when you are employed - your employer will expect you to be able to cut to the chase. Knowledge for the sake of knowledge isn't worth anything" - **small wonder **I went ballistic at the double-standards which had been put across to me

The point is, your organisation's authority isn't determined (in SEO or real-life terms) by their arbitrary academic choices. It's determined by performance, and how useful people find the specified website or organisation. Everyone in life asks "what's the ROI on that?" - especially since tuition fees are pretty high in the USA and creeping up in the UK too

When you get good, non-manipulative, editorial links from external sources (of high relevance and repute), which Google doesn't think have been produced 'for SEO alone' - your SEO authority grows larger. With more authority in a given area, you have more power to rank (as long as errors and crawl barrier don't inhibit that authority from cascading down to the relevant sub-pages). Removing roadblocks like errors, could make you rank much better. But if your authority is low, it'll do jack diddly squat

You need to engage your organisation from a more PR-esque perspective, then when you have built up some good links - tackle the errors much more aggressively. That's my 2pence anyway

If it's a small snippet of text (like you get for boilerplate product descriptions on eCommerce sites), then the risk isn't very high. In fact Google know that often, many sites will end up with the same product descriptions which they take directly from the manufacturer(s), e.g: hardware stores selling parts

If it's a significant piece of content like an article or custom designed content piece, then there could be some risk involved. In that case you might tell the person who is using your content, to add a canonical tag to their page pointing to yours

Unless it's a study, article or something pretty large, I wouldn't be overly worried. That doesn't mean that what you are all doing together is optimum, it would be better to have unique descriptive snippets (so that each page has a different value-add). That being said, sometimes budget constraints don't allow for that

Of course one thing you could do, is wait until everyone has copied your snippet of text, and then change your own snippet to be different and better than everyone else's