• Vicky

    This might be a daft question (but I’ll ask anyhow) does Google definitely support breadcrumb schema markup?
    I recently stumbled across this post: https://plus.google.com/115106448444522478339/posts/8v1yDUUgTnS#115106448444522478339/posts/8v1yDUUgTnS which says they don’t. Breadcrumb schema markup is new to me so it would be good to know.

  • http://www.igorware.com/ Igor

    It appears to support it, my website has schema.org markup for breadcrumbs and they are shown in search resutls. And if you test your website with Google’s rich snippet test tool it will detect breadcrumbs with schema markup.

  • http://www.seoskeptic.com/ Aaron Bradley

    Thanks for this Chris – good overview! However, a slight correction/clarification here. You say:

    “In the past, people were unsure as to whether to just stick with using the older Microformat markup, RDFa or to move to Schema (Micro Data), or to use both simultaneously. It seems clear that Schema is now sufficiently well-supported that people do not need to worry about whether it’s finally safe to fully transition. It is.”

    You’re mixing syntaxes and vocabularies here.

    RDFa and microdata (not “micro data”, which just means “small data”) are syntaxes for marking up structured data in HTML.

    schema.org (not “schema”, which is a generic word that can applied loosely to any vocabulary) is a vocabulary.

    A microformat is an approach to markup that provides both a markup framework (using HTML classes) and controlled metadata for a small range of specific types: in other words, it is both a syntax and a vocabulary.

    schema.org officially supports markup of the vocabulary in HTML using either RDFa (including RDFa Lite) or microdata (as well as direct encoding with the data-interchange format JSON-LD).

    So the choice is not between “Microformat markup, RDFa or … Schema (Micro Data)”. It is between microformats and schema.org, the latter of which can be marked up using either RDFa or microdata. And yes, it is safe (and for many reasons otherwise advisable) to use schema.org rather than microformats. :)

  • http://www.seoskeptic.com/ Aaron Bradley

    It’s not a daft question and there are long-standing problems with schema.org breadcrumb markup that have not yet been resolved (http://bit.ly/17FqXP7).

    To make a long story short your best bet at this point for structured data markup of breadcrumbs continues to be data-vocabulary.org in microdata (though RDFa is probably safe too) as described on Google’s page on breadcrumbs and rich snippets (http://bit.ly/1bS5mQV).

    @IgorWare:disqus Google has been able to parse some schema.org breadcrumbs, but the results are uneven. The presence of breadcrumbs in the SERPs doesn’t require structured data markup, and Google displays all sorts of things in the Structured Data Testing Tool that don’t actually make their way to the SERPs. The most reliable choice at this point remains data-vocabulary.org-based breadcrumbs.

  • Renee Nicole Girard

    What microdata does Google show without markup? For example, I’m now seeing pagination included in the SERPs and breadcrumbs on pages that don’t use rich snippets.

  • Chris Silver Smith

    Vicky, Aaron’s guidance is accurate. As he alluded-to, there are instances where Google does not choose to display the breadcrumbs, even if the markup is on the page. The algorithm determines whether it will display rich snippet treatments of all kinds, depending upon factors such as if the site is sufficiently trusted, the query, and the number and type of rich snippets already displayed for other listings on the same search results page. However, I think that using the markup that Google’s help page for Breadcrumbs provides as an example is the best bet at this point, along with Igor’s suggestion that you use the Structured Data Testing Tool to make sure there are no mistakes and that Google can successfully test it.

  • Chris Silver Smith

    Thanks, Aaron — for the most part, I don’t think it matters whether it’s a syntax or a vocabulary — I treat them all as protocols for structuring data. Throughout the conference and increasingly on the net, people refer to the protocol defined at Schema.org as just “Schema”, which appears to’ve come into common colloquial usage. That may not be very precise — indeed, I’m well aware that a “schema” is a much broader term, but in the context of articles about search optimization, I consider it commonly known enough that the colloquial use is fine for making my points. I believe Schema protocols are a specific set of markups whereas RDFa is a broader method for structuring data — one could use RDFa beyond what is supported by the search engines, and one could use some older versions of RDFa markup that are not the newer Schema markup, so I prefer to differentiate by stating that the choice is indeed between the older Microformats and RDFa versus the newer Schema protocol. I think probably most were not confused by the informal, colloquial use here, but I appreciate your desire for more precise usage.

  • Chris Silver Smith

    Renee, you’re correct that Google is automatically identifying particular types of information on pages and using it in various Rich Snippet markup treatments. This has been going on for a while — such as identifying the dates on news stories or blog posts, displaying data from bulleted lists or from tables, as well as for the breadcrumbs on some pages. As Google becomes more certain that they can successfully identify and call out discrete pieces of data from pages, they’ll continue to do this and broaden the types of data that they can feature in SERPs.

    However, to increase their algorithm’s confidence about items and thereby increase your listings’ chances of having a Rich Snippet treatment, using structured data is ideal.

  • http://www.seoskeptic.com/ Aaron Bradley

    Thanks for your response Chris.

    While I get your point about the colloquial usage of “schema” to mean “schema.org” I think the distinction between vocabularies and the syntaxes employed to encode those vocabularies is an important one. And the continual confusion I see between them, I believe, makes it more difficult for webmasters and SEOs to get a grasp of structured data markup.

    Sorry, but whether or not you mean it colloquially or not, to say that “the choice is indeed between the older Microformats and RDFa versus the newer Schema protocol” makes no sense. If by “newer Schema protocol” you mean schema.org, then schema.org needs to be marked up with something, and that something is either RDFa or microdata; what you say suggests that RDFa is an older mechanism that has been superseded by something else … that can be marked up in RDFa.

    Insofar as professional, forward-thinking search marketers (or, the very least, technical SEOs) now require a solid handle on structured data, I don’t think anyone’s interests are served by imprecise language on the subject. The distinction between a vocabulary and a markup syntax is important, as there are already other vocabularies aside from schema.org that can be employed to mark up web pages – so this is more than mere “semantics” (no pun intended).

  • http://www.kyleeggleston.com/ Kyle Eggleston

    I think this is professional nonsense.

    I’ve tested structured data extensively with a variety of different clients in different industries. I have not seen ONE shred of evidence that indicates that having structured markup will improve your rank.

    It is an absolute false statement to say that it does improve rank.

    The only reason they are claiming it helps is so webmasters adapt to the new, more efficient markup code. There is practically NO correlation between site markup and your Google rank.

  • Vicky

    Thanks Igor, Aaron and Chris, lots of really useful information in your comments I defintitely have some reading up to do on the topic.

  • Renee Nicole Girard

    Thanks for the quick response!

  • Beverly Mapes

    When one reads this article carefully, it focuses on the advantages of structured data being better for visibility and click through rates from SERPs. While ranking is mentioned, one instance adds statistics of 3 placement positions, and the other is very general. At least to me, the point of the article implies improvement in the clickability factor far more than better ranking. @Chris “to’ve??” LOL

  • Mik Dunne

    Hi, you mention adding rich snippets onto to category pages “On category pages (based on aggregate ratings)”. Can you please expand on this or how to do it, as from what I read on Google WMT guidelines, thy can only be used for individual product pages or an aggregated offer page (a page that lists a single product, along with information about different sellers offering that product). Can you please expand or explain how to implement this? Thanks

  • Chris Silver Smith

    Beverly is correct — and I should’ve reported some of Marcus’s statements around how correlation of statistics on ranking does not equate with causation.

    Kyle, I believe you’re correct in asserting that structured data necessarily immediately have any affect on rankings. However, I believe (and the session speakers’ data appears to support) that structured data can have a rapid and dramatic effect upon clickthrough rates, and I also believe that clickthroughs from search results is a factor that Google uses to determine rankings over time. As such, structured data may not have a direct affect upon rankings, but over the longterm it could indeed positively impact rank.

    It’s always possible that there could be some other explanation for the apparent advantage in rankings that pages containing structured data appear to enjoy. For instance, perhaps site designers that employ structured data are generally more proficient at all the elements that positively influence rankings, and perhaps those designers also create sites with better usability and better attractiveness — all resulting in better rankings over time.

    But, until the structured data impact might be isolated from all the other myriad factors, I’ve seen enough indications that it can positively influence ranking over time to believe that the structured data conveys a significant advantage, in of itself.

  • Chris Silver Smith

    Mik, I suggest that you take a look at the HostelBookers.com page that Stephan used as an example. On their page, it lists a number of Hostels in Madrid, and each of the percentage ratings of those individual hostels. However, for the structured data within the page, the site developers combined all of these individual rating values into one aggregate value to display in the search results. You can take that page’s URL and submit it in Google’s Structured Data Testing Tool to see what structured data Google can identify within the page. Then, look at their page’s code to see how they implemented it. http://www.hostelbookers.com/hostels/spain/madrid/

  • Mik Dunne

    Hi Chris,

    Thanks for the info. I will have a look at the code and see hows it done. Thanks for your help.



  • http://bloggingtipstricks.com/ Avnish Gautam

    According to my own Experience, It does not increase rank, but It increases click through ration with the same SEO work & same ranking in search engines.

    I have observed personally on my blog. Thanks.

  • http://www.computershowto.pro/ Attila Szabo

    The greatest error on the part of those who support “implementing” these “breadcrumbs” – namely google and yahoo, is to presume that all webmasters should understand by default all the useless, repetitive, and sometimes contradictory terms used in RDFa, og tags, and such.

    I have struggled for a few months to “do the right” thing with several of my blogs and sites, and after seeing that not only the tags described or pointed to in google’s help articles do NOT do what they ‘re supposed to, but have CONTRARY effects – sites get rated down because of implementing these tags, functionality that previously had worked well stops working, etc. – I have given up on them. Instead of creating a zillion different ways for each giant out there – google, yahoo, facebook, etc – why don’t these good folks sit down at a roundtable and SIMPLIFY AND REDUCE the already way too large number of possible tags and their options..
    These things are unfortunately, at the present time, NOT in the favor or benefit of webmasters, on the contrary, they make one’s work harder and much less rewarding.
    My 2 cents :(

  • http://www.computershowto.pro/ Attila Szabo

    my point exactly. sometimes, no matter how well you’ve implemented these “snippet codes”, snippets don’t show, while in the case of some sites… even though they haven’t implemented anything… snippets show.

    So the problem is, that IT’S NOT CLEAR and transparent at the present moment, implementing what has which results

  • http://www.lacksokning.com/ Energiläckage

    My authorship markup isn’t showing in the SERPs either, only in the rich snippet testing tool. Not sure why the testing tool wouldn’t alert you of any issues like the questions mentioned above?