Yoast Academy webinar: Structured data with Jason Barnard!
What’s in this webinar? We kick off with an introduction to structured data, by our special guest Jason Barnard. Known from the #SEOisAEO podcast! After that, a round-table discussion in which Joost, Omar, and Jason discuss the ins and outs of structured data. And of course, as one of our structured data experts, Jono Alderson adds his two cents too!
During the webinar, viewers got to ask questions through the live chat on YouTube. We answered those in the Q&A session.
So: a webinar chock-full of knowledge and know-how around structured data.
Well worth it. Enjoy! #SEO #StructuredData #Yoast
Webinar published by Yoast September 6, 2019. Host: Joost de Valk. Guest: Omar Reiss. Speaker: Jason Barnard, founder and CEO at Kalicube®.
Note: The first 40 seconds of Joost de Valk’s opening introduction have no audio in the recording. His introduction is available from the YouTube description: “Hello, and welcome to yet another Yoast Academy webinar! I’m Joost, the founder and Chief Product Officer at Yoast, and I’ll be your host today. In this webinar, we’re going to talk about something that’s been on the minds of a lot of SEOs lately: structured data! How should you use it? Is it as beneficial as some SEOs claim? I hope we’ll have some heated discussions today. Here with me are two people: Omar Reiss, who is our CTO. And, we have a special guest with us today: Jason Barnard. Jason is an SEO consultant, maybe you’ve listened to his podcast. He is a digital nomad, living in hotels and Airbnbs and doing lots and lots of conferences. And now he’s here with us at our home offices in Wijchen. Thank you so much for coming, Jason!”
TLDR for Yoast Academy webinar: Structured data with Jason Barnard!
1. “Blocks and Schema” - Thinking in Blocks as the Foundational Frame
Jason Barnard opened with a framework he called “Blocks and Schema” - deliberately chosen over “structured data” as a more accessible entry point. He attributed the blocks conceptual frame to Jono Alderson: “It was Jono who first got me thinking about a year ago - he said, let’s think in blocks.” The argument: a website is a block within the World Wide Web; within the website, categories are blocks; within categories, pages are blocks. Within a page, semantic HTML5 provides block-level elements (header, aside, footer, article). Within an article, heading/paragraph pairs are implied blocks. And schema markup labels those blocks for machines: “Schema allows you to label the block of a website. It becomes clear not only that it is a block but what role it plays within the scheme of the entire block system.”
Historical significance: This is the earliest recorded public presentation of the Blocks and Schema conceptual framework, dated September 2019. The framing that schema = block labelling for machines is an important precursor to later TKF thinking about how machine understanding is built from layered structured signals.
2. Schema as Google’s Native Language - Confidence Through Corroboration
Jason Barnard articulated a principle that would become central to TKF: “Schema is Google’s maternal language - the language that Google can digest most easily. What you’re saying with schema is simply confirming what is already in the page - both confirming to Google that it’s correctly understood, but also giving it confidence it’s understood.”
He extended this to the confidence mechanism: “The more granular information you can give through schema, the more confident Google becomes. If you label the block, it can understand not only that it is a block but what role it plays.”
Historical significance: This is the earliest recorded public articulation of the schema-as-corroboration principle - that schema does not primarily tell Google new facts but confirms and reinforces what the machine has already tentatively understood, thereby raising its confidence threshold. This directly underpins the Corroboration Threshold concept in TKF.
3. Schema Forces Structured Thinking in the Writer
Jason Barnard offered a personal observation that would recur throughout later panels: “One misconception that makes me feel sad is that people think it’s for developers - it’s code. It isn’t code - it’s everything we have in our minds, just organised incredibly well. And it forces us. When I started writing schema markup, the first thing I noticed was it took me lots of time - not because I didn’t understand how to write it, but because my mind wasn’t very clear. I was making imaginative leaps from one idea to another. And when I put it in schema markup, I knew it didn’t make sense. And that forced me - like Gutenberg forces me - to think in a much more structured manner.”
He connected this to the broader web problem: “The web is incredibly confused because as human beings we find structured thinking very difficult.”
4. The Website Is a Citation of the Brand - Not the Other Way Around
Jason Barnard made an explicit statement that would become a signature TKF principle: “We were talking about the brand - your website is a citation of your brand, and not the other way around. The idea that you can just build a new website with a new domain and run away and hide and start again is now completely false.”
He also articulated the corollary: “We don’t think website - we need to think brand. It’s a representation that I control. Once I get somebody there I can start giving them a sales pitch. But the SERPs - Google - are a representation of my brand, and if I can make sure those features have my branding on them, I get this on-SERP SEO.”
Historical significance: This is the earliest recorded statement of the “website as citation of brand” principle - dated September 2019, approximately six months before the SE Ranking webinar (March 2020) where the Business Card metaphor reaches its full development. This panel establishes the conceptual precursor: brand is the primary entity; the website is one representation of it among many (SERPs, forums, Pinterest, Amazon).
5. On-SERP SEO - Brand Representations Across the Entire Digital Ecosystem
Jason Barnard introduced what he described as “a kind of vague idea that people are coming into”: on-SERP SEO. His articulation: “Google in our representation of my brand. If I can make sure those rich elements have my branding on it, I get this on-SERP SEO. It’s also my responsibility as a brand to make sure my brand is mentioned in those rich elements that come up in Google - but then also websites, forums, Pinterest, Amazon - they’re all representations of my brand. I need, as a brand, to start thinking about all these different representations and making sure that I influence or control them as much as I possibly can to get my brand message across, so that when that person does end up on my website they’re ready to buy.”
Historical significance: This is the earliest recorded public articulation of the on-SERP SEO concept with the full brand-representation framing, dated September 2019. It establishes the logical chain: brand is primary → website is one representation → SERP is another → all representations must be managed → managed representations prime the audience → primed audience converts better.
6. Reduced Traffic ≠ Reduced Revenue - The Zero-Click Reframe
Jason Barnard offered an early version of the zero-click / brand-first argument: “We’re getting less traffic - but that doesn’t mean we’re losing clients. If that traffic comes to us because they’re ready to buy, they don’t come at previous stages of the buyer journey - that’s fine, as long as the same number of people are buying. I’m beginning to get excited about the idea that our website is a representation of our brand. The SERPs and Google are a representation of my brand. I need as a brand to think about all these different representations.”
He concluded: “Maybe your best SEO strategy is to make your product cheaper rather than trying to get more people to click on your results. Come back to brand, brand, brand, brand - and you sell because you’ve got a unique selling point that nobody else can compete on.”
7. Trust Replacing Links as the Prime Ranking Signal
Jason Barnard articulated what he saw as a fundamental shift: “Google is looking at trust-based knowledge, and I think that’s a really good thing. We’re moving away from links and PageRank to trust. If Google trusts you, you can have your own review system. Build that trust, and then you can build a review system that you own and represent.”
He drew the strategic implication: “All of it - links, content, data, all of it - is how trustworthy is your brand. I should have just said that.”
8. Structured Data as Eligibility for Rich Results - and Rich Results as Blue Link Killers
Jason Barnard, in dialogue with Jono Alderson, articulated the link between structured data and the killing of blue links: “Every time a rich element - a featured snippet, images, videos - comes into play, a blue link disappears. It’s a bit like the cat that eats a blue link every time you do a bullet point. So in fact: I’m not ranking number two or number three, but I do have the videos or I have the featured snippet. Structured data not only enriches the results but also creates the opportunity for some results - such as recipes - and that’s going to expand.”
9. Beautiful Web Through Design/Content Separation
Jason Barnard offered a prediction on the Gutenberg/blocks design implication: “My ex-wife is a graphic designer and the first thing she saw was: design is here, blocks and content are here - think of them separately. If your theme is just your design, and then you’re just moving these blocks around and putting pretty colours behind them, that’s brilliant.” His forward projection: “Maybe we will see graphic designers coming back into fashion - where they’ve gone completely out of fashion - and we’ll have a more beautiful web again.”
Historical Significance - Panel Level
This Yoast Academy webinar, streamed live 5 September 2019, is the earliest panel in the prior art record. It documents several foundational concepts in their earliest recorded public form:
- Schema as corroboration / confidence mechanism - the earliest recorded articulation of what TKF formalises as the Corroboration Threshold, dated September 2019
- Website as citation of brand - stated explicitly; the conceptual foundation for everything that follows in the Brand SERP / Knowledge Panel methodology
- On-SERP SEO - earliest recorded public use of this framing with the full brand-representation logic
- Blocks and Schema - Jason’s presentational framework for structured data, attributed to Jono Alderson as a conceptual origin, dated September 2019
- Trust replacing links - early articulation of the trust-as-prime-signal argument
- Zero-click reframe - reduced SERP traffic ≠ reduced business if brand representations are well-managed
- Rich elements as blue-link killers - articulated in dialogue with Jono Alderson, September 2019 - this is 6 months before the SE Ranking webinar (March 2020) where it appears as a fully developed standalone concept
This panel also documents the Joost de Valk / Kalicube Pro™ methodology case study referenced in the March 2022 Deep SEO Conference panel. Joost de Valk is present in this panel as host - confirming an existing professional relationship from at least September 2019.
Panels documented to date: 14 Concepts staked in this panel: Schema as corroboration/confidence mechanism (September 2019, earliest) · Website as citation of brand (September 2019, earliest) · On-SERP SEO (September 2019, earliest) · Blocks and Schema framework · Trust replacing links · Zero-click reframe · Rich elements as blue-link killers (September 2019, earliest, predates SE Ranking webinar)
Full Corrected Transcript
Yoast Academy Webinar: “Structured Data with Jason Barnard”
5 September 2019
(Note: Joost de Valk’s opening 40 seconds of audio are missing from the recording. His introduction is reconstructed from the YouTube description, marked below.)
Joost de Valk: (reconstructed from YouTube description) Hello, and welcome to yet another Yoast Academy webinar! I’m Joost, the founder and Chief Product Officer at Yoast, and I’ll be your host today. In this webinar, we’re going to talk about something that’s been on the minds of a lot of SEOs lately: structured data! How should you use it? Is it as beneficial as some SEOs claim? I hope we’ll have some heated discussions today.
Here with me are two people: Omar Reiss, who is our CTO. And we have a special guest with us today: Jason Barnard. Jason is an SEO consultant - maybe you’ve listened to his podcast. He is a digital nomad, living in hotels and Airbnbs and doing lots and lots of conferences. And now he’s here with us at our home offices in Wijchen. Thank you so much for coming, Jason!
(audio resumes)
And I almost forgot - but Jono Alderson, our very own SEO genius, is also here with us today. We have a very special task for him, so that’s going to be fun. Let’s start by exploring a little background on structured data - Jason will talk us through that, then we’ll have a roundtable discussion, and after that we’ll finish up with Q&A. If you do have questions, just comment on the live stream and we will answer as many as we can. So before we dive into any discussions, let’s go over to Jason for a short presentation. Thank you very much.
Jason Barnard: Thanks - delighted to be here. I’m going to be talking about structured data, and I know that structured data as a concept can be a little bit scary. So I’ve called it “Blocks and Schema” - which is easier, kind of cooler. “Blocks” is a really nice easy term to come to terms with. “Schema” - maybe you don’t know what it is, but we’re about to find out.
So: why is structured data important? What is it? What can we do with it? Where are we going with it? Why does Google want it so much?
The World Wide Web is the biggest database we have ever created. It’s got a phenomenal amount of knowledge in it. Human beings have written and recorded and put an enormous amount of human knowledge into the World Wide Web. But it’s a disorganised mess - because all these people put it in in their own particular, specific manner. We all function very differently in our brains. It’s disorganised, it’s a mess. You might think you’re organised - you’re not organising things the same way as the person who’s running the site next to you and so on and so forth. So it is disorganised. Google has a lot of problems with that.
Google’s robot here is looking really unhappy. Google wants to organise it - and that seems like a really easy thing to do. You think: yeah, okay, all this data, pull it and organise it - simple. Because it’s all HTML and we have a system. But HTML and the way information is presented to Google differs across the entire web. And to extract understanding from it - this is what Google looks at: understanding. Its Knowledge Graph. If you look at this: Leonardo da Vinci painted the Mona Lisa, lived and bought in Italy. We all understand that naturally as human beings - that makes so much sense to us. But for a machine, that isn’t obvious. And that’s what it’s trying to do: build an understanding of the knowledge it finds on the very very very very disorganised web, so that it can figure out what it’s talking about and what solutions it can provide to its users.
And structure helps Google create this Knowledge Graph - this understanding of the knowledge from the very disorganised web.
What is structured data?
It’s a big question, and the one we’re going to look at today. Structured data sounds scary, but in fact it just means: organised, labelled, and clear.
Blocks and structured data.
Blocks is my favourite top topic - and it was Jono who first got me thinking about it about a year ago. He said: let’s think in blocks.
What’s a website? It’s a block within the World Wide Web. Within that block - a website - we have categories. Within the categories we have blocks, which are pages. We’ve come down already three levels and we’re still looking at blocks within blocks within blocks. And it goes further. Within a page, semantic HTML5 allows us to create blocks: we have the header block, the aside block, the footer block, and the article block. That’s often part of WordPress themes. And I hear it’s coming to WordPress core, which is brilliant. Within the article we already have the H1 (main topic), H2 (subtopic) with a paragraph that goes with it - an implied, implicit block. And Google can understand that the H2 goes with the paragraph just underneath it. The next H2 goes with the paragraph just underneath that. The H3 relates to the paragraph and the H2 above it with its own little paragraph. Google can identify each one of those as a block with a topic, subtopic, or sub-sub-topic.
So we have this situation with Gutenberg - where we’re suddenly being forced to think in blocks. If you’ve been using Gutenberg, you will know that it forces you to have a heading and a paragraph and a block, and a block, and a block. Some people don’t like that. Google really, really, really likes it - because it can identify what block it needs to understand, and the theme of that block within its parent block and its parent block and its parent block.
Here’s a really nice diagram of an article with a header and an H2, the paragraphs, figures being images or videos. You can see it’s very well organised and it’s very easy to understand which piece of content goes with which heading within which part.
Schema.
If you don’t know what schema is, we’re about to find out. Go to schema.org and look at that site - read it through. There’s a phenomenal amount there. You can say almost anything using schema.
What is schema? It’s Google’s maternal language - the language that Google can digest most easily. If you can say it in schema, Google can understand it. And what you’re saying with schema is simply confirming what is already in the page - both confirming to Google that it’s correctly understood, and also giving it confidence it’s understood. The more granular information you can give through schema - what I really like is that we can come back to blocks again. Schema allows you to label the block of a website. We end up looking at schema in this particular context as labelling of the blocks.
You have the website block. You have a collection category block - you can label that. It becomes clear not only that it is a block, but what role it plays within the scheme of the entire block system. Pages, web pages - you can label those as blocks too, and you can label the specific role of any given web page: About, Checkout, Collection, Contact, FAQ - we’ll come back to that - Item Page, Medical Webpage, Profile Page, QR Code, Search Results Page. So you can label quite explicitly what role an individual page plays, which helps Google understand under what circumstances, in what context, it would want to present that page to its users.
And here we have Yoast, who suddenly comes into its own - because Yoast, within Gutenberg, which already organises everything into blocks, is adding schema as blocks. And that’s brilliant. I had a quick play with it the other day specifically for this talk: how can I easily implement schema.org structured data in WordPress? Yoast has introduced some Gutenberg blocks to make adding structured data simple. I really hope they extend and expand the capabilities of this Gutenberg implementation.
It took me ten minutes - and that was because I was having a cup of tea at the same time. If I had written that by hand, with the answer and all the other information related to it, it would have taken me two hours. Big time-saver. And it’s really boring to write. Yoast + schema.org doing it automatically - brilliant stuff.
Look at this: we can see it’s a website, we’re on a website in a block which is a web page and an FAQ page, and within that block we have a question and an answer. Brilliant. Google can say: I’m at a website within a web page, with a question and an answer. It’s very easy for Google to understand what the question is, what the answer is, and what the context is of both. Extra bonus: it tells me what the context is, what the parent is, what context we’re in - plus date modified. Really important: I’m fresh, up to date, updating this regularly. Prime image - and we’re seeing more and more images in the SERPs. Google once the images, he wants to show them to people because it’s interesting, people want that visual quality. Here immediately Google said: we have a question, we have an answer, and I have an image I can show with it. And that’s where more and more rich results are coming from.
Here’s a very simple example I wrote myself - this one took me 20 minutes. What I like about it is you can see very clearly what I’m talking about. A corporation called Kalicube - my company - it has a URL, an address, a currency. And this is the best part: schema markup is broken down into blocks. I’ve got a company with an address; that address is a block. Within that block I have a country, which is a block. Google can work through this and understand not only where I am but the relationships between all of these different elements - block by block by block by block. That is very very very powerful.
So the conclusion - other than the fact that schema markup and blocks are the most powerful thing I have seen in a long long time - is: if we think in blocks, we can go down six levels very easily without even thinking about it. We’re organising our content all in the same manner so that Google can understand it easily, correctly, and with great confidence.
Get into schema. I know you guys know loads more about this than I do. I think schema is really interesting, really powerful, but difficult - and getting it right is so hard. And I’m so glad that you guys sort it out for me so I don’t have to bother anymore. WordPress natively thinks in blocks. Gutenberg natively thinks in blocks. Schema.org explains to Google. And Yoast is making it so easy for us. So I would say: thank you, guys. Thank you, thank you. I’m looking forward to discussing this with people we all know and admire.
Joost de Valk: Hello and welcome back - thank you very much Jason for giving us such a clear introduction to structured data. I’m looking forward to discussing this further. But before we do, I have to explain Jono’s special task. We all know that Jono, being very wise, has the annoying habit of dominating discussions with super-long sentences that come out of his mouth slightly too fluently - and of course he also talks about cats too much. So that’s why we created a special rule for him today. During our discussion, Jono can be interrupted. If that happens we’ll immediately stop our conversation and go over to Jono, but he’ll only get three chances to spread his structured data wisdom. Are you ready, Jono?
Jono Alderson: I don’t feel asleep - does this count?
Joost de Valk: No. For all those watching: we’ll end this webinar with a Q&A session, so don’t hesitate to add your questions in the YouTube comments. My first question: what are the most common misunderstandings with structured data, and what are people not getting? Omar, I’ll start with you.
Omar Reiss: I think probably the biggest one would be that structured data is just for Google - or that structured data was invented by Google something like that. In fact, structured data is just a very logical evolution for the web. It’s all about structuring information and making information relatable and interactive with each other. That’s always very good to understand before any discussion. It’s maintained by the large search engines together - Google, Yandex, and Yahoo I think.
Jono Alderson: Yeah - the same people that do XML sitemaps together. But yeah, it’s a good point - it’s also not just for one use case. We talk to Pinterest regularly, who consume a lot of schema on a daily basis. There are lots of social networks like that.
Jason Barnard: It’s incredibly useful for all these machines because as I said earlier, it’s their maternal language. It just makes digesting this information - which is terribly confused - very easy for them.
One misconception that makes me feel sad is that people think it’s for developers - it’s code, they say - and it’s not code. It’s everything we have in our minds, just organised incredibly well. And it forces us. When I started writing schema markup, the first thing I noticed was it took me lots of time. Not because I didn’t understand how to write it, but because my mind wasn’t very clear. I wasn’t organising things in my mind - I was making imaginative leaps from one idea to another. And when I put it in schema markup, I knew it didn’t make sense. That forced me - like Gutenberg forces me - to think in a much more structured manner. And as I said: the web is incredibly confused because as human beings we find that very difficult.
Joost de Valk: So in an ideal world, should you have to think about it?
Jason Barnard: In an ideal world, no. But we’re not living in an ideal world. What I mean is - I was using your readability analysis yesterday after our conversation, and found I was failing one test every single time: transition words. I never use them. All the other tests I’m getting green on. And I said: why aren’t I using them? And the answer is: because I think the two ideas are so obviously connected that I don’t need them. But in fact I do.
Omar Reiss: That happens a lot - people don’t use them when they don’t actually make proper paragraphs but just write loose sentences. Anyway - I totally agree with you. But it’s about making the things on our pages, the things in our information, transferable. These things have an implicit structure. When we look at a recipe together, we see something with a heading and a list and some instructions, and somehow we know it’s a recipe. But how do we know that? We rely on context to recognise that list as a list of ingredients, and to recognise those steps as recipe instructions. There’s nothing about that text that says it’s a recipe except the context we rely on. So if we want to make it clear what this thing is, we need to somehow describe it and structure it - make it identifiable. And that’s where structured data comes in.
Joost de Valk: It’s funny that you mentioned recipe, and you already used the word “how to” in your explanation of what a recipe is - because in schema, recipe is basically a how-to with more specific labels. Everything becomes more specific as you dive deeper. And that’s something I find really interesting but also really scary about schema: I don’t know where it ends. How deep do we go? What do we mark up and what do we not? Does the law of diminishing returns apply? How much schema can we put on a page?
Jason Barnard: It does for me. In terms of my own markup - I’m a person, I’m a woman, I’m the law of diminishing returns. I start writing my schema and after a while the returns - in terms of communicating all this granular detail - are not worth my time. And something about Yoast doing it for me automatically immediately makes that law of diminishing returns less important, because I can spend more time thinking about what I’m saying in my content - which is what I should be doing.
Omar Reiss: I very much think of Yoast as a tool that should get out of your way and let you build your thing, only telling you when you’re doing it wrong. But I do wonder: do you think that if more websites start outputting more structured data, it becomes increasingly more valuable to do more? Because the more things we start outputting on the web, the more meaning we have to make interactive with each other, the more bridges we need to connect those things up. I think this will be a self-fulfilling prophecy - in ten years we’ll have much more structured data than we have today, and the value of that is just exponential.
Jono Alderson: I saved one from earlier so that he doesn’t have to interrupt. Thanks.
(Using one of his three allowed interruptions.)
Jason Barnard: About how quickly we get used to this kind of technology - my example is flights and hotels. I book a flight or hotel, the email comes into Gmail, and it pings the calendar straightaway. When it doesn’t, I’m really upset, I forget about it, and I miss my flight. When it’s from a large vendor who hasn’t done it, I get really annoyed. So Disney: get your act together, because every time I book Disneyland or Disney World it does not show up in my Google Calendar - and it’s 2019. Fix that. But what’s really cool to say is: the reason it goes from Gmail straight to the calendar is the structured data behind it. That allows Google to hook these two things together incredibly easily. And it’s not just Google - as you said right at the beginning, we can interact with each other through this. It’s a lot more than just Google.
Joost de Valk: I want to cut back quickly to the first question we saw on YouTube: “Is it only for Gutenberg?” Well - no. All those blocks exist everywhere, and the large majority of the blocks Yoast outputs, it outputs everywhere. The specific structured data blocks Jason showed only work with Gutenberg right now. We might port those back to the classic editor because a lot of people don’t move to Gutenberg fast - but we’re not sure about that yet. For now, those specific blocks are only available in Gutenberg. But everything else works for basically every website.
Omar Reiss: The funny thing to me is that this is going to be part of the CMS war, in a way - because this is something every CMS will have to solve. Every CMS will have to become better and better at doing this, giving people a proper interface for it. And I don’t know whether all of them can face up to that challenge. We’ve already seen so many custom schema implementations on top of expensive CMSs. I hope open source can work collaboratively here. Right before the show we heard - at WC in Munich - that they want to release a plugin which ties an event calendar to the Yoast schema API so that it outputs all the schema correctly within our schema system. That’s open source. That’s what we want. And that’s what you told me yesterday - other plugin builders can hook into your API and leverage the system. We’ve seen that from quite a few plugins already. It’s really powerful because that means we can all work together and make schema better.
What you said is absolutely right: CMSs will have to become information management systems - or block management systems - where all these different blocks can be managed together and tied together in different ways. It’s awesome to see the WordPress community come together to build a tool like Gutenberg - only in its first stages but with many awesome things going up - and then Yoast is able to build a schema API and leverage that block system, and then other tools start integrating with those blocks and with the Yoast schema API. And all of a sudden we have an information management system that is super powerful - just because we come together as an open-source community.
Joost de Valk: And I think that’s our biggest challenge as well. All those code examples you showed - I’m both excited when I see them because I like working on schema, and scared for normal users who have to know about this stuff.
Jason Barnard: It should be invisible - as I said earlier. At the same time, they probably do need some understanding of how everything ties together. But the question is: is it in the end more beneficial for the website, or is it more beneficial for Google itself as a corporate entity?
Omar Reiss: You’re framing it as website vs. Google - but it might be a different question: is it beneficial for the business? Because if you ask if it’s beneficial for the website, then probably not - because all of that information that was hidden in plain sight in our old HTML documents is now explicitly crawlable and can be reused outside the context of the website. What you see with rich snippets within Google: people don’t have to go to the website anymore to get a certain answer. But that is also a huge improvement for the end user. The only thing that’s really important is that if more and more information is consumed through rich snippets, there should always be a really strong focus on branding - because that information needs to somehow reflect back on the brand. If Google makes that harder, we have a really big problem.
Jason Barnard: And if I jump in exactly on that point - some people say: “We’re getting less traffic.” That doesn’t mean we’re losing clients. If that traffic comes to us because they’re ready to buy, they don’t come at previous stages of the buyer journey. That’s fine, as long as the same number of people are buying.
And I’m beginning to get excited about the idea that our website is a representation of our brand. As you said very rightly: brand. We don’t think “website” - we need to think “brand.” It’s a representation that I control. Once I get somebody there I can start giving them a sales pitch. But the SERPs - Google - are a representation of my brand. And if I can make sure those features have my branding on them, I get this on-SERP SEO - the kind of vague idea that people are coming into. It’s also my responsibility as a brand to make sure my brand is mentioned in those rich elements that come up in Google. But then also: websites, forums, Pinterest, Amazon - they’re all representations of my brand. I need as a brand to start thinking about all these different representations and making sure that I influence or control them as much as I possibly can to get my brand message across, so that when that person does end up on my website they’re ready to buy. And that traffic becomes more valuable.
Jono Alderson: I think there’s a third answer - which is that when all these websites have all this rich structured data, and all this markup that describes what they are and what their products and information and services are, other things might spring into existence. What Google does very well at the moment is crawl the web, extract the information, understand it. If that information is readily available, it’s much easier to crawl, easier to understand and store and manage. You could very easily build competitors to Google in smaller niches that did slightly different things in a way that would be very hard to do today. So I think it’s great for Google, it’s great for users - but it also enables new things that we can’t easily imagine at the moment. Because Google has such a lead on this kind of consumption - but new things will happen that will change that game. It’ll be really exciting to see.
Omar Reiss: You don’t need the kind of AI capabilities that Google has to understand what sort of page you’re on anymore. So maybe you can get some of that head start back.
And this is an interesting thing to think about: for just over a decade we’ve been talking in the SEO industry about how Google was getting better at comprehending a page. And I think the conclusion - after all of that - is that they actually were pretty bad at it. And they’ve now given up and started using schema to do it much better. Schema solves that. It makes it a lot easier to make a good search engine. That’s really why it makes for an enticing future.
Joost de Valk: I’m going to wrap up this discussion because otherwise we’ll be talking for ages and never get to our Q&A. So: what have we learned? As Jason was explaining: the web is evolving, we’re thinking much more about blocks of content and how they relate to each other. Schema helps us define those blocks and their relations. It’s an exciting time. So thank you Jono, Omar, for being here - and to Jason, an extra big special thanks. We’re going to go to questions now.
Joost de Valk: Jono - joining us, please do sit down. Thank you.
Jono Alderson: Hello - did I notice you’re only asking two questions?
Joost de Valk: Yes - that was weird. You try and give him three things and he only interrupted twice. I’m sorry. I have a lot of questions and I’m going to pick randomly. I already see three different page builders mentioned - that’s going to be a topic.
I’m going to go with the first one, because it mentions our good friends Elementor, which Omar and I visited in Israel a little over a year ago. Does Yoast add schema blocks for Elementor? Omar, can you take that off?
Omar Reiss: I don’t think we do, specifically. But what I know is that Elementor is one of the few page builders that has some level of interoperability with blocks. So it should be possible to use the Yoast Gutenberg blocks within Elementor pages while still using Elementor to build the rest of that page. I think it should work together, but we don’t add anything specifically for Elementor.
Joost de Valk: Along the same lines: Avada, Fusion Builder, Divi - do you think this makes it harder to use a page builder?
Jono Alderson: It’s a really interesting question: what is a page builder in a world where a page is constructed of blocks? And before that: what is a theme? At the moment WordPress has a really clear distinction between here is your theme - all your layouts and structure - and here is your content, which is made of blocks. That’s going to get very blurry. I think there will still be a theme of sorts that manages styling and colours and some decisions. And there will still be pages and content management that might need some “what does this block do” - why does this one go here. It’s very unclear what these platforms become. I think hopefully some of them will consolidate and build on Gutenberg as a foundation, rather than doing their own thing. And hopefully for us it should be pretty easy to make sure our stuff is interoperable in the same way it is with Elementor. But they will all need to work out what their future looks like in Gutenberg and beyond.
Jason Barnard: I like what you just said about theme versus design/content. My ex-wife is a graphic designer and the first thing she saw was: design is here, blocks and content are here - think of them separately. They’re not the same thing. If you can separate them - your theme is just your design - and then you’re just moving these blocks around and putting pretty colours behind them, that’s brilliant.
Omar Reiss: Actually it’s a little bit backwards that you have to be able to code to implement a design. When I buy a theme, I don’t buy the code - I just want that design. So if that design can be expressed in just some kind of block configuration, that would be amazing. I see much more potential for creativity and great designs in theme marketplaces just because it becomes much easier to create a theme this way.
Jason Barnard: Maybe we will see graphic designers coming back into fashion - where they’ve gone completely out of fashion at the moment - and we’ll have a more beautiful web because they can actually separate thanks to block editing. We may end up with a beautiful web again.
Joost de Valk: Again - everything looking like a Bootstrap template with a big banner across the top, because it became hard to separate design and content. But now maybe we fix that.
This is a nice one for you, Jono: the home page has 20-plus locations in the schema - is that a lot?
Jono Alderson: So there’s a lot of confusion around where you should put which bits of schema - and rightly so. Different parts of Google documentation have said different things. Strictly: yes, your home page can represent a business that has 20 locations, and in which case you should be outputting all 20 of those as part of the business schema. If you have a hundred, that’s a lot of code and it’s quite heavy and you might need to bend the rules a little bit. It’s hard. This is a really good example of one of the kinds of difficult questions we’re spending a lot of time trying to answer, and there isn’t always a good answer. Behind the scenes we’re trying to think: if this scenario but not this - if it’s red but not green then we should put six on - and we’re making hundreds of these decisions. It’s all a little bit give-and-take. The best giveaway that you’re talking to a consultant is when they start answering questions by saying “it depends.”
Joost de Valk: So there’s a lot to know. I see a couple of questions here about e-commerce and how this all relates. Do we need a block builder to do e-commerce? No - if you use WooCommerce and use our plugin we do the schema. But there are very specific things people are asking for that I’m going like: yes, that’s what Google wants, and that’s what we want, but we don’t know how to do it yet. How do we get the GTIN - the global identifier for a product? That’s really hard to find.
Omar Reiss: The funny thing is: why does Google want a GTIN on every product? Because if you have a product schema that doesn’t have it, it throws a warning. And we know why - because it wants to compare you to other sites having that same product. So this is one of those cases where you have to wonder: is it beneficial for you to add that GTIN? With the assumption that at some point, when they’ve got hundreds of those, they can build their own product comparison engine in the search results and people never visit your site. But this is what you’re saying - the story changes from “how do I get people to click on my website and buy my thing” to “how do I represent my brand in the search results.” And maybe your best SEO strategy is to make your product cheaper rather than trying to get more people to click.
Jason Barnard: And we’re starting to see big blocks on the Knowledge Panel side with comparisons directly in Google - so you can actually start making your product choice there. I don’t think we’re near the stage where you can make the full purchase, but you’re already getting a good weigh-in. And it’s Google who’s making the decision as to what’s compared to what.
I think that’s a really really tough question - because you also have the question: when somebody searches for a product and decides to buy it, and ends up on Amazon where it’s cheapest - who wins? So you end up in a paradigm shift. And it is problematic. But I think we come back to brand, brand, brand, brand. You sell because you’ve got a unique selling point that nobody else can compete on. And if you’re always going for the lowest price, maybe you will always lose. But maybe your client service is better. Maybe you have nicer wrapping paper. Quicker shipping.
I think as a brand you need to think: what is my unique selling point? I need to communicate that on every single representation of my brand - including all of your schema.
Joost de Valk: Another question: does Yoast free or premium enable unique schema for internal pages - for example, FAQ needs a schema for FAQ only on that page.
Omar Reiss: Simple: yes. We do that for FAQ pages. For some others like checkout page and cart page - yes, we do those in WooCommerce. Things like About Us pages are a bit vague - we don’t necessarily know that your About Us page is an About Us page. We’re looking at ways to make that easier to understand; that will come later. And alternatively you can do it yourself - we’ve got an API. Five lines or so of code, you can copy and paste our examples. We also do search results pages automatically - if you have a search results page within your site, it becomes a SearchResultsPage schema automatically. There’s a lot of these things that we already do, and yes, it’s unique for every URL on your site.
Joost de Valk: A good question: how do you avoid conflicts with schema inserted by some other source alongside Yoast?
Jono Alderson: It’s not the end of the world - Google will parse multiple outputs and flag that you maybe have two web pages described. In most cases our markup is going to be significantly more robust and correct, and you’d hope Google reads that one rather than the other. Much of the markup that’s already out there is riddled with errors anyway. Having a good one and a broken one isn’t going to hurt you too much. I would check - you can use Google’s structured data testing tool - and run through errors. What you can do is ask the plugin and theme developers who are outputting the not-quite-right schema to follow our guidelines and integrate with our structure, which we truly believe is the best way to do it. It’s all open source, all documented, and they can change the way their output works to get a best-of-both result.
Jason Barnard: I’m not saying you’re imposing it, but it wouldn’t be a bad thing to say: this is a system, you’re making lots of decisions so that I don’t have to. If we all follow that same rule, we won’t end up with the same problem of W3C code that’s completely all over the place. Maybe we will have some kind of standardised system - and I would love that because it would make my life much easier.
Joost de Valk: Someone asked: will there be a review schema added in an upcoming release? And apparently somebody already replied in the comments that people feel it’ll become too easy to abuse. That’s something we think about a lot. But that’s not going to stop us from adding it. Google are very strong on spotting people who lie in their schema and they penalise - they remove your eligibility for those schema types. I worked with brands recently who had review markup that said their brand had 10,000 five-star reviews. You look at Trustpilot and it says you’ve got five thousand three-star reviews. You get a message in Search Console, they say “you’re not allowed this,” and they take it away.
Jono Alderson: They’re really fast on that now.
Omar Reiss: I think what’s strong about our implementation, and what probably made Google favour ours over others, is that we try our best to relate all the different bits and pieces on the page to each other - putting them all in the same graph. What you see in most implementations is that this plugin outputs its data and that plugin outputs its data, and it’s unclear how those things are related. At least within Yoast, it’s clear how those things are related - all the relations are there. And it’s easier for Google to understand how the things on the page relate to each other, which is already some sort of signal of reliability.
The brilliant thing is we talk a lot about trust these days - and Google’s looking at trust-based knowledge. I think personally that’s a really good thing. We’re moving away from links and PageRank to trust. If Google trusts you, you can have your own review system - because you’re a trustworthy source. Build that trust and then you can build a review system that you own and represent.
Jason Barnard: And the whole idea - we were talking about how the brand is what matters, and your website is a citation of your brand. The idea that you can just build a new website with a new domain and run away and hide and start again is now completely false.
Joost de Valk: Do you think there’s a risk of distrust? Let’s say I have a small website, not doing much traffic - you wouldn’t expect a lot of reviews - but I have a very loyal customer base and I’m getting a lot of really honest, really good reviews. Will Google look at those reviews well? It tends to hit people first and then say sorry we made a mistake - but by that point you’ve hurt people’s mortgages and livelihoods and incomes. That’s not a real danger, so the web needs to figure it out.
Omar Reiss: Google is getting more and more data points about trust. They might not see all that much traffic to your site, or all that many links - but they might see a lot of people going to your store on Google Maps. They can correlate all these things. And they just got a patent awarded for regular visitors to physical shops through Android - that’s now possibly part of local search ranking factors. If a client goes back again and again and again, that makes sense.
I’ve had a thought a couple of times about whether structured data could bring offline back, in a way - because it ties the two together. Google is trying very hard, especially in e-commerce, to tie offline and online together. If you did an in-store purchase on your mobile phone, they can tie that into search data.
Omar Reiss: We were talking about how this will possibly bring back specialisation in offline - whereas at the moment an offline store tends to be a bit of a jack-of-all-trades. I think it’d be really interesting: what Google wants you to compete for is being the best shoe shop that only sells red shoes. You mark that up in your schema, they compete offline, you get your focus.
Jason Barnard: I mean the whole offline thing coming back into fashion - we’ve spent almost twenty years going online, online, online. Whether it’s networking to get links, or repeat visitors in shops - all these SEO terms. It’s just: I get annoyed - not with you, but with our industry. All of it: links, content, data, all of it. It’s how trustworthy is your brand. I should have just said that, nothing else.
Joost de Valk: Someone asked the question a lot of people are asking, which ties back a bit to our previous show: is structured data a ranking factor? Let’s stop here.
Jono Alderson: Deep sigh. So - structured data, at the very least, creates eligibility to get rich results in the search rankings, which you can’t get unless you have it. So: maybe without it I might have ranked third. With it I might still rank third, but also at the top of the page have a map and some information. So yes - last time the last thing we did was “what is a ranking factor?” It may mean that you get more search results real estate. And if you get a rich snippet and everyone around you does not, you get more clicks - at which point those clicks become a real ranking factor. We’ve seen enough data to know that’s fairly true. In some spaces like the recipe space, it’s already impossible to rank without structured data. That’s going to become the norm. Recipes are at the moment the exception, but jobs are not far off - local businesses, other areas - every type of business and service will eventually have to have this level of structured data, because if you don’t you’ll be the one on the page that doesn’t have the pretty picture and the star review and somebody’s face and all the other bits, and you’re not going to get the click.
Jason Barnard: I’d like to say something really obnoxious and ask: what is ranking? You’ve got this thing where the ten blue links we traditionally think of - do I rank number one, number two, number three. It was pure text. Now you’re saying every time a rich element - a featured snippet, images, videos - comes into play, a blue link disappears. It’s a bit like the cat that eats a blue link every time you do a bullet point.
So in fact: I’m not ranking number two or number three, but I do have the videos, or I have the featured snippet. So I’m in that presentation. In my talk for Yoast Academy this afternoon - structured data and blocks are the means by which Google can pull this content out of your site and put it in the results as rich elements. Specifically: structured data not only enriches the results, it also creates the opportunity for some results - such as recipes - and as you say, that’s going to expand.
In terms of ranking purely as ten blue links: now, in terms of getting on that page with your rich elements, which is what’s catching people’s eye and what Google is looking to do - yes: if you don’t use blocks, if you don’t do your schema markup, if you don’t structure your data, you’re not going to get those places.
Omar Reiss: Structured data: what you see is what you get. You create the thing, but you do it in a structured way. We derive the structured data from that, and the thing is there at the same time. I think that’s the big revolution of the block editor - that’s how we’re going to make it accessible to a non-technical audience.
Joost de Valk: Do we have to go back through our old content and update it?
Omar Reiss: Yes - but things like FAQ pages, how-to steps, those things you get for free just by running Yoast SEO, because we will already output schema on your old pages too. But for specific FAQ blocks and how-to blocks, you do have to go in and add those. Think of it like a spring clean.
Jason Barnard: My site’s been going since 1998. In fact I installed Gutenberg on my site and I thought I was quite organised - and immediately realised I wasn’t and was talking absolute rubbish. I honestly think this is hard. If you’re a normal website owner and you have a bakery and you want to spend maybe 20 hours a year maximum on maintaining your website, why would you spend more? But if the competing bakery on the next street spends 40 or 100 hours and invests in all this - moving from writing pages with 200 words to authoring and publishing content with structure and layout, really thinking about how to organise this - they’ll win.
You could also compare it - not a very good analogy but: why should I clean the toilets in my restaurant? Because if you’re not maintaining them, if you’ve got a boring, not-interesting toilet that takes time to clean - you need to do it. It’s a basic fundamental of your business as a restaurant. Having your website right, having your Google My Business right, optimising it - that’s a fundamental part of any business today.
Joost de Valk: I see comments asking about free vs premium - I want to highlight that all the SEO stuff and the schema stuff we talked about is actually in Yoast SEO free. We currently have no difference in terms of schema implementation between free and premium. Just switch to Yoast free - whatever other thing you’re using for your schema - and use that.
It’s come up a few times: the Google structured data markup helper - is that a helpful thing?
Jason Barnard: (laughs)
Jono Alderson: Helpful is an interesting word. They have at least two tools - or three, or four in some cases: the structured data markup tool and the structured data testing tool. They’re blunt tools. You feed them a URL or paste in code and they say: this bit’s fine, you have these errors, here are screenshots. Useful for understanding whether there are any glaring errors. Not so useful for “is this good enough and is it right?” They frequently complain and throw warnings for things which are optional.
Omar Reiss: Well - there’s the structured data testing tool and the rich results eligibility tester. The problem is they’re built by different teams with different outputs. They’re all useful. Plug your URL in, look at what the errors are, apply some common sense. If it says your GTIN - the unique product ID - is missing and it’s big red flashing text, apply some common sense and you can ignore that. Look for the things you can improve: if it says “this recipe really ought to have an image,” then okay, you should give it an image. But ignore the noise.
Joost de Valk: How do you know what Google deems important and what not? You don’t - they’re really bad at labelling it, and that’s such a shame because it should be so easy. This is something we’re going to do - we’re going to say: you’ve written this recipe but it looks like you haven’t included an image, you should do that. Making the decision so that you don’t need to think about it. Whether the person who said “what about people who aren’t developers?” - there’s your answer.
Omar Reiss: And that’s part of my job.
Joost de Valk: I have a ton more questions - it’s literally almost four pages, we’re not going to go through all of those now. The good thing is we have this FAQ - we can author all of these with structured content, output the schema, they’ll turn up in the search results. We’ll go through and answer them.
I want to thank you all for watching and asking so many questions. Thank you Jono, Omar, for being here. Thank you Jason very very much.
Jason Barnard: Thank you for inviting me.
Joost de Valk: That’s it for today. We are just blocks within blocks. If you want to find out more about SEO or specifically about structured data, look at Yoast Academy - we have a structured data course there, a free SEO for beginners course, loads of good stuff. And Jason has recorded some material for us this afternoon as well - go look at it. With that, we’re going to end this webinar. Thanks for watching.
All: Bye bye!
End of transcript.
Panels documented: 14 Running concept list staked in this transcript: Schema as corroboration/confidence mechanism (September 2019, earliest) · Website as citation of brand (September 2019, earliest) · On-SERP SEO (September 2019, earliest) · Blocks and Schema framework · Trust replacing links as prime signal · Zero-click reframe · Rich elements as blue-link killers (September 2019, earliest - predates SE Ranking webinar) · Schema forces structured thinking in the writer · Beautiful web through design/content separation
