1

Is it considered blackhat if the JSON-LD structured data is only shown to search engines? The content is still present on the web-site for users, but it's only marked up as structured data when we detect that the visitor is a bot. Can we get penalized because of this?

This is to add a layer of protection against people trying to scrape the site.

(I changed the title of the question to re-iterate my point. There are instances when showing one type of content to search engines and another type of content to humans is permissible, example is sites that use Flash. I'm wondering if an exception exists for structured data.)

Example of the data

Here's one example of structured data that's inside the source code that the bots are supposed to read. This same content has to be present on the page in visible format as well so the humans can read it.

{
 "@context": "http://schema.org",
"@type": "Person",
"address": {
"@type": "PostalAddress",
"addressLocality": "Good city",
"addressRegion": "Great State",
"postalCode": "47918",
"streetAddress": "701 N Nice St"
},
"name": "Firstname LastName",
"telephone": "765-764-1111" 
}

1 Answer 1

1

Cloaking is when you show content to a search engine that is different than what you show to a user.

Google tests web pages from outside of it's own network and you will never know. If there is a difference between what you show googlebot and users, Google will spot-check more pages for differences. If enough pages appear to have significant enough differences, the penalty is applied.

It is as simple as that.

To answer your question, it can be cloaking if the content change is significant enough. Never show different content to search engines than to users. Just keep it all simple.

[Update]

Thank you for the example of what you are specifically asking.

Cloaking is what I have defined earlier, however, there is a bit of tolerance especially in light of desktop versus mobile. In the early days, cloaking could simply be determined by capturing the page twice, once via the crawler, and once external to the crawler often from another network, and comparing the checksum for each page. However, these days, it is not so simple with desktop versus mobile.

We know that Google can fetch a series of pages and determine templated content versus page content rather easily. In light of the state of the web these days, I would have to assume that some level of analysis takes place to compare the content portion of the page and possibly the template portion of the page separately. How pages are analyzed for cloaking these days will likely remain a mystery. However, it is reasonable to assume that some minor differences in the non-content portion of the page is to be expected in some cases.

The next question is, Is it wise to present JSON data to crawlers only?

No one can say specifically if a search engine, Google in particular since Bing seems to be rather tolerant, will see the omission of JSON as being deceptive. It has to be recognized as a risk even if it appears to be small and a reasonable thing to do. As a recommendation, I would say to include the JSON data to both users and crawlers to avoid any issues. Why? Because cloaking is not a small violation at least in Google's eyes. If cloaking is detected, Google will spot check the site and then apply the penalty. This is an automated process. Once the penalty is applied, it can take quite a while to remove the penalty and likely is a knock on the sites trust metrics effecting search even after the penalty is lifted.

11
  • Thank you for the reply, but it still doesn't answer my question. The users never see the 'structured data' it's not rendered on the site for the users to see. It's data that only the search engines use. So if I'm not showing a snippet to humans, which they cannot see anyway (unless they open the source code), would that penalize the web-site? Jul 13, 2016 at 15:49
  • To paraphrase: it is technically cloaking, but it is a type of cloaking that isn't sure to get you penalized. Jul 13, 2016 at 16:34
  • You're loading different HTML to search engines than to "humans", which is essentially the same concept. Jul 13, 2016 at 19:12
  • @RobertSinclair I am not sure what exactly you mean. Structured data is often content that is seen by a user, though not always all of it as in reviews. It would help to see an example of what you mean. As it stands, there was only one answer I could give. A generic answer. If you add an example, we can get more specific.
    – closetnoc
    Jul 13, 2016 at 21:59
  • Ok I added it to the question to keep the formatting Jul 13, 2016 at 22:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.