Welcome to the Q&A Forum

I have an acquaintance/customer, who is a politician in Italy, he had a website under his name as domain firstnamelastname.it (where firstname and lastname are the firstname and lastname of the guy).

He used it during his campaign, then he forgot about the existance and didn't pay the bill to the isp.

I don't know what happen exactly, I assume someone registered it after it expired and started using it for some fraudulent purpose, because checking the backlink profile I can see the first backlinks from 2013 are legit and coming from relevant sources (local newspapers), but all the following 300+ links are spam, using keywords like "louis vuitton", "ralph lauren", and so on...

Right now the domain is in status "deleted", so I could just register it again, disavow the spammy links and had recycle it. But I am scared as hell.

I never had to disavow a single link in all my life, and I don't know if it is going to really clean up the domain from any possible google penalization. Is it?

That's because you are using RDFA, if you were using microdata markup you would see itemscope and itemprop.

https://developers.google.com/structured-data/breadcrumbs

https://schema.org/docs/faq.html

http://stackoverflow.com/questions/14307792/what-is-the-relationship-between-rdf-rdfa-microformats-and-microdata

I agree with you.

But let me just add something.

New Delhi to New York is 12.000 km, that's around 120ms of latency. While within India it's probably in the 3 to 15 ms range.

1. TTFB should be below 500ms to be in a safe zone, that 120ms leaves 380ms for everything else, is definitely doable,

2. each http request made from the browser is going to be affected by the latency, but for a long list of good reasons the website should be optimized reducing the number of request necessary to render pages,

3. to improve performance reducing latency a CDN is the thing to use, duplicating content around the world on different physical or virtual machines with the purpose of improving performance, still make sense, but only if your traffic is epic.

Standard wordpress installation with zero plugin can do what you described, I suggest you read docs and search for info directly on https://wordpress.org/support/

I suggest using Google Analytics Experiments, with multi-armed bandit.

Doing so you can set the significance level you want and the experiment will stop when that value is reached. By default the p-value there is 95%, I usually left it set to 95%.

And I usually repeat the experiment a second time to confirm results.

99% of internet users are aware of only two TLD, their homeland ccTLD and .com

If you use .eu most of people who actually type the domain will either try their ccTLD extension or .com, so if you own the .com too you should redirect the .com to the .eu, but why do so? If you own the .com just use the .com, if you don't, change domain and use one with .com

Test using the appropriate tool: https://developers.google.com/structured-data/testing-tool/

I tried with the first url you posted and breadcrumbs are there.

I would merge copy and 301 the weak one to the the strong one.

Finally I understand your point, but... Have you measured it? You have some cases where you moved the images to a third-level domain and after re-index you observed a ranking drop?

Because I did it once, for an e-commerce, and I didn't observe that. It's just one case, and doesn't mean much, but just searching on google for this topic I can find many who claim to have witnessed an increase in images ranking thanks to the adoption of a cdn.

Domain of an image url is for sure a ranking factor, and you probably are right. But I would like others to jump into the discussion and help to clarify.

Because I think google images algo is more keen on evaluating other (well known) factors. And it's logic. If you have some html on site.com with one image coming from flickr.com and one coming from worthlessdomain.com, which one would you favour in images SERP? Flickr? Why? Doesn't say anything about the relevancy of that image for that keyword.

Of course if you have an html page about flowers on site.com, and you have one image coming from worthlessdomain.com and one image from asitefullofflowersimages.com it may have some relevancy, but is it for the domain authority/trust? Or because all the images of asitefullofflowersimages.com are of flowers, with flowers alt text and so on? I bet the second.

Plus, google guidelines suggest to serve images from a cookieless domain to improve performance, of course they mean to improve performance to improve ranking of the page, and in that context they are probably disregarding SEO for images.

I don't have the minimum idea if domain authority/trust is such a strong ranking factor for images, and I have the impression you have more experience than me on this subject, but I am still very skeptical.

Of course it does reindex them on the third-level domain, but it's not going to affect ranking of tld. And it should not affect images ranking either, you just have to give them a short while to regain their positions as soon as reindexing is done.

Is it a problem to have the images indexed on a different domain? Why?

What kind of issues are you experiencing? I do have a couple of websites serving images from a third level domain with a different SSL certificate and I have never experienced any issue, images are indexed and well positioned.

Plus, I have never heard of such a penalization and it doesn't make sense, a SSL connection is secure no matter who issue the certificate, and html img src is supposed to be used to host images from other domain, it doesn't make sense to penalize a website for that.

Google is meticulously punishing cloaking, I would stay away from it.

Just all content must be served from SSL http connections. It doesn't matter if the certificates are not the same.

This one may help: https://moz.com/blog/cant-do-keyword-research-like-its-2010-whiteboard-friday

Point 2 of Rand is specifically:

It used to be the case that we built separate pages for every single term and phrase that was in there, because we wanted to have the maximum keyword targeting that we could. So it didn't matter to us that college scholarship and university scholarships were essentially people looking for exactly the same thing, just using different terminology. We would make one page for one and one page for the other. That's not the case anymore.

Thank you for your answer Gary.

Analyzing a competitor backlink profile I noticed they have the habit of adding multiple links to the guest post/advertorial they get, one link to their site, plus some links to authority site like wikipedia and institutional websites.

Is there anyone with some data, testing, and so on, supporting this?

I can see their SERP positioning is very good, but they spent a lot of money on quite expensive links, so I don't know if it's the DA of the linking domains or the multiple links making the difference.

Anyone has an idea?

I am using PayPal as a payment option on a couple of website from day 1, paypal have been preferred by an average 35% since then, slightly behind credit cards (the two other options are cash-on-delivery and wire-trasfer).

I have been shocked by these numbers too.

We use Website Payment Pro, the customer UX is subpar when compared to the two credit cards getway we use but customers doesn't seems to care, the merchant management user interface (to manage refund and phone payment solution) is incredibly sluggish.

In terms of conversion rate changes I cannot judge because we never A/B tested with or without.

In terms of fraud, since it's not asking Visa 3DSecure and MasterCard SecureCode it's positively more prone to frauds. But in my experience the merchant protection they offer is good, we had 90% of our fraudulent orders covered, even in months where frauds were higher than the limit of 4% of total payments theoretically covered by the contracts.

Thanks for the long answer EGOL.

What I intentionally didn't mention in my question (because I was curious to learn if the fact by itself was revealing of anything special going on) is that those results are from one of my sites, the one with the lowest PR/DA/mT, where I write freely, without a schedule and not caring much about keywords, density, and all the usual suspects.

It is about a niche subject, and I think you are totally right.

I thought this one was carved into stone, max number of results from the same domain in SERP is... two. Or... three?!

I was searching for some familiar keywords and found three results from the same domain, isn't that... unusual?

That's a common cheating technique (Product schema in place of Organization), so far I never saw google spotting it.

I would expect product pages to have AggregateRating markup for that specific product. Not for the organization. Google is not going to show that rating anyway, so why put it there?

Keep in mind, you want star rating to appear in SERP because will improve CTR.

To have star rating appear in SERP you need markup and a user query matching the target of the rating.

Let's assume you sell vegetables and your organization name is "Veggie Inc".

Would you expect the product page for "Early Girl Tomato" to show up in SERP when people search for "Veggie Inc."? Unlikely. So if you add organization aggregate rating markup on that page no one will see that.

Do you think the star rating will appear on "Early Girl Tomato" page for other queries? No way.

Best practice is Organization AggregateRating on 1 page, usually the homepage, which people will likely search for using your brand as query, a query which will match your Organization name...

Hope I was clear.