Authoritas Tea Time SEO: "What Are Knowledge Panels and How to Get Into Knowledge Graphs"

Webinar | Authoritas Tea Time SEO Series Date: Spring 2020 (exact date not confirmed from transcript; COVID lockdown context suggests April-May 2020) Session Host: Jason Barnard (Kalicube) - guest-hosting this session Authoritas Hosts: SEO Joe Bloggs, Kerry Shepherd (Authoritas) Co-presenters: Dixon Jones (InLinks), Andrea Volpini (WordLift) Source: https://www.authoritas.com/blog/what-are-knowledge-panels-and-how-to-get-into-knowledge-graphs Recording: https://www.youtube.com/watch?v=2sBA74rS_uI

Note: This is an unusual format - Jason Barnard was guest-hosting the session rather than being interviewed as a guest. Each of the three presenters (Barnard, Jones, Volpini) contributed a set of practical tips followed by a roundtable discussion. Andrea Volpini explicitly states “now we are in 2020.”

TLDR for Authoritas Tea Time SEO: “What Are Knowledge Panels and How to Get Into Knowledge Graphs”

1. Wikipedia Is Not Necessary for a Knowledge Panel - 391 Non-Wikipedia Sources Documented

Jason Barnard opened with his most direct statement to date on Wikipedia independence: “Wikipedia is not necessary. Up until last year perhaps it was - I don’t think it was, but certainly since last summer - what I call the Budapest update - Wikipedia is definitely not necessary. It’s easy to get into the Knowledge Graph without Wikipedia if you know how.”

He cited Kalicube Pro™ data: “In the Knowledge Panel, Wikipedia is only cited 60% of the time. 30% of the time there is no citation at all - and behind that there is more often than not no Wikipedia page. About 40% of Knowledge Panels aren’t from Wikipedia.” He reported 391 non-Wikipedia Knowledge Panel sources collected via the Authoritas API, up from 140 one month earlier.

Historical significance: This is the earliest recorded panel in which the 60%/40% Wikipedia/non-Wikipedia split is stated as a specific data point with a collection methodology cited. It also establishes the Budapest update reference as “last summer” (summer 2019) from the vantage point of spring 2020. The 391 non-Wikipedia sources figure provides a dated baseline for the growth of Kalicube Pro’s source database.

2. Knowledge Panels Are Multi-Source and Multifaceted

Jason Barnard described a live experiment conducted with Andrea Volpini - mutual brand SERP interference: “We’ve been doing some experiments, pinging backwards and forwards with silly games about analysing SERPs and brand SERPs, seeing how much we can mess with each other’s Knowledge Panels - and Andrea got himself onto mine.”

He used this to illustrate the multi-source principle: his own Knowledge Panel entry came through Wikipedia (the “easy way in”). Andrea Volpini’s came through WordLift and Crunchbase (“the hard way in”). Both achieved the same result through different source paths. He noted that YouTube is also a cited source for songs and videos within a Knowledge Panel.

3. Nurture Your Sprouts - Building Sub-Entity Knowledge Panels

Jason Barnard introduced the “sprouts” concept: “Nurture your sprouts. We’ve been working on their Wikipedia page and their Knowledge Panel - we got that - and now we’ve got sprouts. That’s the two bosses of the company, Jack Bonnie Faye and Betty L’Amour, and we’re building those up.”

He also noted a timing observation: “My sister - I created a site for her and added some information to some sites, and she got a Knowledge Panel yesterday, and the API hasn’t yet thrown back a result for her. I think that’s a delay in the time the Knowledge Graph itself updates and the Knowledge Panel updates and the API actually updates.”

Historical significance: This is the earliest recorded use of “sprouts” as a named concept for building related-entity Knowledge Panels - the idea that the primary entity’s Knowledge Panel becomes a launchpad for building connected entity panels. The timing observation about the API delay versus the Knowledge Graph and Knowledge Panel updates also appears here for the first time as a documented observation.

4. Google My Business vs Knowledge Panel - The Yellow Pages Distinction

Jason Barnard drew a clear conceptual distinction that would become a recurring TKF principle: “Google My Business is a business listing - it’s like the yellow pages. The Knowledge Panel is something completely different. It’s saying: these are facts about a company that you don’t control. And that’s the point of it. Whereas Google My Business - you have control, it’s a business listing.”

He also described how Kalicube Pro’s international tracking reveals the switching behaviour: “Google will appear with Google My Business when we’re local and with the Knowledge Graph when we’re not.”

Andrea Volpini added the tactical application: the schema:hasMap property inside an Organization schema type can explicitly connect the entity to its Google My Business listing, giving Google a structured signal about the relationship between the two.

5. The Scale Argument - Why PageRank Cannot Scale and Entity Search Will Win

In roundtable discussion, Jason Barnard built on Dixon Jones’s visualisation of the Knowledge Graph as a single-access direct storage for a single point of truth: “The PageRank link system can’t scale far enough or fast enough. One of the reasons perhaps we’re advancing so quickly - or Google moving forward so quickly - is because they’re now much more entity-based in their search.”

Dixon Jones confirmed from InLinks data: “It turns out there’s only tens of millions, hundreds of millions, billions of topics - not trillions. Therefore most of the internet is duplicate content in the sense that it’s talking about the same thing. The Knowledge Graph is a single point of truth for Google - web pages can inform that single point of truth.”

Andrea Volpini added observational evidence: “The crawlers that arrive at our triple store - where we publish the data connecting to Wikidata or DBpedia - are not the same classes of IP as the crawlers that arrive on our HTML pages. That’s a kind of proof that these are two separate systems.”

6. Schema as Content Model, Not Just Rich Snippet Trigger

Andrea Volpini articulated a principle Jason Barnard would echo and develop: “My advice is to make sure you understand schema not only as a way to get the rich snippet, but as a way to reorganise and remodel your content. If you’re not thinking this way, get on board - because that’s really the most important aspect of it.”

Jason Barnard responded with his own experience from the WordLift podcast experiment: “The first thing was I had to sit down and think: how do I organise this? So how should I organise it? And that made me think about the podcast slightly differently. It forced me to think about existing content and also what I’m now going to do in terms of my content strategy. And forcing you to think about how you’re structuring things shows you often how badly structured your ideas are in the first place.”

Historical significance: This is an early recorded instance of the schema-as-thinking-tool argument - the idea that schema markup forces conceptual clarity on the content creator, not just on the machine. It predates the Yoast Academy panel (September 2019)? No - this is spring 2020, so it follows Yoast. It represents a continuation and deepening of the argument first made in that panel.

7. Entity-Based Engagement as the Real Challenge

Jason Barnard identified the next strategic frontier: “Currently we’re trying to get all this data into the Knowledge Graph, into the Knowledge Panels. But the challenge now is: how do you get the engagement of the user back to your own brand, back to your own business? The challenge is helping the Knowledge Graph to help your business - to get the Knowledge Graph to work for you.”

Andrea Volpini confirmed: “I think we have already passed the phase in which we are focusing on just building the entity data. The real game changer is to create engagement with the user by structuring content around a larger graph. Structured data is a way to get to the search engine at a faster pace - but the actual value comes from creating an interaction with the human behind the search.”

Historical Significance - Panel Level

This Authoritas Tea Time SEO session, spring 2020, documents:

60%/40% Wikipedia split - earliest stated as a specific data point from the Kalicube Pro database
391 non-Wikipedia Knowledge Panel sources - dated baseline for the source database growth
Budapest update identified as “last summer” (summer 2019) - earliest recorded reference to this update by name
Sprouts concept - earliest recorded use as a named strategy for building sub-entity Knowledge Panels
Google My Business / Knowledge Panel distinction - the yellow pages metaphor stated for the first time
hasMap schema property for connecting organization to GMB - documented here as a practical tactic
Two separate Google crawlers - Andrea Volpini’s IP-range observation confirming HTML crawl vs triple-store crawl separation
Entity-based engagement as the next challenge - transition from “build entity data” to “make entity data work commercially”

Panels documented to date: 17 Concepts staked in this panel: 60%/40% Wikipedia split (data point, spring 2020) · 391 non-Wikipedia sources baseline · Budapest update named (“last summer”) · Sprouts concept (earliest recorded) · GMB = yellow pages / Knowledge Panel = fact record distinction · hasMap for GMB connection · Two separate Google crawlers (Volpini IP observation) · Entity engagement as next challenge

Transcript for Authoritas Tea Time SEO: “What Are Knowledge Panels and How to Get Into Knowledge Graphs”

SEO Joe Bloggs: Hello everyone, and welcome to Authoritas Tea Time SEO - where we encourage everyone to take a break at 4 p.m., the traditional tea time, to connect with others and learn something new. I’m SEO Joe Bloggs from Authoritas, and I’m also joined by Kerry Shepherd here from Authoritas. Hi guys. Today we have something new in store for you - we have a host for Tea Time SEO this Friday. We have Jason Barnard, and he’ll be hosting today’s session with Dixon and Andrea. So I’m going to pass over to Jason to introduce himself, share his tips, and leave it to you guys.

Jason Barnard: Thank you very much - and thank you for inviting me and allowing me to present. Thank you Dixon for turning up, and Andrea - I know you’ve got some great tips. I’ve actually had a sneak peek and I’m pretty keen to hear what you’re going to say. I’m going to start, all about Knowledge Graphs and Knowledge Panels. Then Dixon second, and then Andrea third.

I’m Jason Barnard. I call myself The Brand SERP Guy®, and I’m hoping other people start calling me that soon as well. I collect Brand SERPs for a living - a very small living, but a living nonetheless. I use the Authoritas API, which is brilliant. I’m a digital nomad, temporarily on hold in the centre of France. And I’ve got a Brand SERP course which is just starting to sell. I’m not making a big living out of Brand SERPs just yet, so I still have to do some digital marketing consulting on the side.

I’m going to go right into my quick tips.

My first tip: Wikipedia is not necessary.

Up until last year perhaps it was - I don’t think it was - but certainly since last summer, what I call the Budapest update, Wikipedia is definitely not necessary. It’s easy to get into the Knowledge Graph without Wikipedia, if you know how.

I’ve actually started a collection, through the Authoritas API, of sources from Knowledge Panels that are not Wikipedia. 391 of them so far, and that’s growing all the time. A month ago I was only at 140 - that’s because I’ve enlarged the database, got a bigger dataset going.

As you can see in the Knowledge Panel, Wikipedia is only cited 60% of the time. 30% of the time there is no citation at all - and behind that there is more often than not no Wikipedia page. The rest is from other sources. So about 40% are from neither Wikipedia nor no-citation. And on the right you can see some of the sources that you can put your information into so that Google can crawl it, find it, and bit by bit build up an opinion about who you are and what you do. That’s the essential thing for having a Knowledge Panel - that Google understands who you are and what you do, who you’re related to, what your relationships are. And then it will start showing that as fact.

My second tip: Know that Knowledge Panels are multi-source and multifaceted.

If we look at this - it’s Andrea and myself. We’ve been doing some experiments, pinging backwards and forwards with silly games about analysing SERPs and brand SERPs, seeing how much we can mess with each other’s Knowledge Panels. And Andrea got himself onto mine, so I got myself onto his - as you can see down the bottom there. I got in through Wikipedia - the easy way in. And Andrea took the hard way in and got in through WordLift and Crunchbase and various other sources. WordLift is a very powerful tool for that, and I’m also using it for my podcast.

You can also see that the sources for the songs - that’s YouTube. YouTube is a source. And you can see it’s multifaceted. I’ve written some songs. Andrea hasn’t written any songs, but he gives great business advice through those Google video things - what are they called, Andrea?

Andrea Volpini: Cameos - these are the Google Cameos. That’s an experiment of Google. I think it hasn’t rolled out massively, but interesting enough. I think it’s fun.

Jason Barnard: And that shows that Google thinks you’re a worthy person to give Cameo appearances to - to advise people about things. I’m just there to sing songs. And it also has the organisations he’s founded - that’s obvious information he’s placed around the web, that Google’s got a grip on and can put into the panel without a Wikipedia page.

And you can see it also knows who my mother is, that I created Boowa and Kwala - they say “movie” but it’s a TV series - I’ve released albums, when I was born, and which university I went to.

My third tip: Nurture your sprouts.

This is a client of mine. We’ve been working on their Wikipedia page and their Knowledge Panel - we got that - and now we’ve got sprouts. That’s the two bosses of the company, and we’re building those up. What we’ve got is this little sprout linked to the Knowledge Panel in the Knowledge Graph. And that’s another interesting point: to get a Knowledge Panel, sometimes you can get one without being in the Knowledge Graph. That may well be a delay between the moment the Knowledge Graph updates and the API throws back a result.

(Dixon Jones noted the correction: more commonly one can be in the Knowledge Graph without having a Knowledge Panel appear.)

Yes - you can have it both ways around. Thank you. And my sister - I found this earlier on. I created a site for her, added some information to some sites, and she got a Knowledge Panel yesterday. And the API hasn’t yet thrown back a result for her. So I think there is a delay in the time the Knowledge Graph itself updates, and the Knowledge Panel updates, and the API actually updates.

That’s it for me. I’ve done my sprouts, I’ve done my Knowledge Graph, I’ve done my Knowledge Panel. Oh - almost it for me. I’ve got a free tool where I track Brand SERPs, Knowledge Panels, and Knowledge Graph presence. I use the Authoritas API, which is brilliant, and I’m going to be using WordLift tools for analysing the SERPs. It’s called Kalicube Pro. If you want to add your brand, please do - because it’s actually a research project, and the more brands I get in there, the more honest that dataset is going to be. Please do add yourself, your competitors, your events, your product, and tell your friends to do it. The more I can get, the better the data gets, and the more we can learn as a community.

Off you go, Dixon - please introduce yourself.

Dixon Jones: I’ve been around doing SEO since about the same sort of time as Jason, and I’m CEO of InLinks.net - I also know the brand ambassador of Majestic. I used to be CMO there, and I once had an agency as well back in 1999.

The first thing I wanted to talk about is visualising the Knowledge Graph, because it’s hard to explain. If you look at the top line of this chart, that’s a bunch of web pages - that’s how the internet used to be. Google used to go and look at all those web pages, try and find the content on them, try and send back the best bit of information for your search. The problem is: as web pages grew from billions to trillions of records, that becomes a problematic system. And it’s much better to start organising the world’s information by topics than by web pages - because it turns out there are only tens of millions, hundreds of millions, billions of topics - not trillions. Therefore most of the internet is duplicate content in the sense that it’s talking about the same thing.

I visualise Google’s Knowledge Graph as a single-access, direct-access storage for a single point of truth for Google. Then the web pages can inform that single point of truth, and they can store everything based around the idea of topics, things, or entities - rather than just web pages. That also gives the advantage of being able to feed in other data sources: lyrics, manually entered data, internal data, IMDB, and the other data sources Jason was talking about.

This is quite a good way of visualising it, at least from my point of view.

(Dixon notes technical connection issues and talks around a missing slide.)

My last tip is about topic trends. What we’re using now - hopefully coming out next week - is a trend graph, a market trend intelligence system, looking at how topics and ideas are trending over weeks and months. We’ve got some ideas: one is “antibacterial soap,” which had a big spike over the last month and is coming back down - still demand. And there’s “hydroxychloroquine” - something that Trump thinks is really cool - which has suddenly had some big spikes and is also coming back down. We’ve taken tens of thousands of topics and you can go and look at the trends over days or weeks. The idea is discovering new ideas: as a new idea pops up, it’ll pop into the system. Good for content writers, for SEOs, and eventually for businesses - identifying topics and being able to look at trends. That’s my last tip.

Andrea Volpini: I’m CEO of WordLift. I’ve been focusing on automating SEO and bringing some semantics to it. I built my first website in 1996, I’ve been doing digital marketing for a long time, and I’m happy to be here.

Andrea Volpini’s first tip: Same-as links or SameAs is important for getting your presence into the Knowledge Graph and helping Google and other machines properly disambiguate your entity.

In the beginning I thought this was going to be the new link-building - it would completely replace traditional HREFs. And in a way it’s happening - we can see the impact on the interaction of my SERP with Jason’s SERP, built entirely on these links that are referencing data.

So the first step for you is to use structured data and get to the level of understanding where you’re not just linking content, but linking data with other data. This is our tool - available for WordPress and any other website - and you can use it for creating a knowledge graph that describes yourself, your companies, and your products.

Andrea Volpini’s second tip: You should link your data with other people’s data.

That’s really a core concept of linked data as introduced by Tim Berners-Lee. Here we see my entity within Wikidata - a publicly available giant graph. I am creating links - much as I would from one page to another - from my entity to its equivalent entity on another website that’s using WordLift. In this particular example, we’re using a WordLift ID, which is a property inside the Wikidata ontology that allows people on Wikidata to create backlinks to data published with WordLift. You can go into Wikidata and ask: “Who is doing SEO that is also using WordLift?” and get a response that collects data from both us and Wikidata. When you create this structure, you can transfer knowledge about yourself and your product at wider scale - and it’s not just Google using this data.

Roundtable - Data links vs traditional links:

Dixon Jones: They’ve got a bit more structure to them. But there’s nothing to stop Google from deciding that a typical old-fashioned hyperlink is a relationship in the same way that schema-based links connect things. What happened was that they set up a new team for developing the Knowledge Graph, and the new team thought about things in slightly different ways and didn’t necessarily use traditional links as a significant signal for that kind of work. I think they kept the PageRank-based approach and the entity-based approach reasonably separate within Google, and they were fighting for meritocracy in search results over the years. But in the end, entity approaches are going to win - because you cannot have a database with trillions of web pages all indexed all the way through. Scale becomes an issue. When you find 200,000 pages on how to tie a bowtie - really, one is enough.

Jason Barnard: That’s a brilliant point - the scaling issue. The PageRank link system can’t scale far enough or fast enough. One of the reasons Google is moving forward so quickly is because they’re now much more entity-based in their search.

Andrea Volpini: I totally agree. And as proof: the crawlers that arrive at our triple store - where we publish the data connecting to Wikidata or DBpedia - are not the same classes of IP as the crawlers that arrive on our HTML pages. That was a kind of proof that these are two separate systems. One is a typed link - a link where you have the structure to understand what type the relationship is. The other is an untyped link, which is by nature very messy.

With a typed relationship you can infer new facts just by creating queries across multiple datasets. You can go into Wikidata and ask who is doing SEO and also using WordLift, and get a response that federates data from both sources. When you create this structure you can transfer knowledge about yourself and your product at wider scale.

Roundtable - The real challenge:

Jason Barnard: The interesting challenge for SEO now is: currently we’re trying to get all this data into the Knowledge Graph, into the Knowledge Panels. But the challenge now is getting the engagement of the user back to your own brand, back to your own business. The challenge is helping the Knowledge Graph to help your business again - getting the Knowledge Graph to work for you.

Andrea Volpini: I think we have already passed the phase in which we are focusing on just building the entity data. That’s been my focus since 2014, when things were coming up around the Knowledge Graph. For me now I don’t see that as a challenge - I see that there is a lot of interest around structured data, which is good. But the real game changer, as you say, is creating engagement with the user by structuring content around a larger graph. Structured data is just a way to get to the search engine at a faster pace - it’s an immediate result. But the actual value comes from creating an interaction with the human behind the search.

Andrea Volpini’s third tip: Think about schema not just for snippets, but as a content model.

A lot of the focus in the industry right now is: what type of snippet should I get? What type of visibility can I get? That’s really part of the immediate result. I started selling structured data in 2011, when schema.org was just starting. Now we are in 2020 and we should go way beyond just thinking about what type of snippet you get.

Here we see the work that we’re doing with Jason on his website - working on a content model for providing visibility to his podcast using schema and the Knowledge Graph to wrap the relationships inside the content for users, not only for machines. One page becomes a nice navigation that moves the user from the podcast page to the contributor page to the event page where the podcast was recorded. My advice: understand schema not only as a way to get rich snippets, but as a way to reorganise and remodel your content. If you’re not thinking this way, get on board - because that’s really the most important aspect of it.

Jason Barnard: What struck me about this experiment we’re doing: the first thing was I had to sit down and think - how do I organise this? How should I organise it? And that made me think about the podcast slightly differently. I had to think about the structure: a podcast series with podcast episodes, recorded events with different contributors. But secondly, it forced me to think about existing content and also what I’m going to do moving forwards in terms of my content strategy. Forcing you to think about how you’re structuring things shows you often how badly structured your ideas are in the first place.

Q&A - GMB vs Knowledge Panel:

Dixon Jones: (reading an audience question) “I’ve got an unclaimed Knowledge Panel for a client brand search. However, I’d rather the company’s GMB listing appear instead, as it’s much stronger in appearance, and I can also track links from my GMB. Can I request my Knowledge Panel to be removed?”

I would say you can - but I don’t see any reason why you can’t have both. One of the reasons I screwed up early on is I tried to mix my GMB listing and my personal branding and they heavily overlapped. I think there’s no reason why you can’t have both. If you’ve got an unclaimed KP, I’d still claim it - because otherwise somebody else will try, or it won’t get claimed. But you can still get your GMB listing to appear when it’s supposed to.

Jason Barnard: This brings in a boatload of questions and problems. What’s Google doing? It’s showing what it thinks is most relevant and valuable to the user searching your brand name. If it’s showing the Knowledge Panel instead of the Google My Business, it thinks that’s the most valuable to that person. What I’ve found is: Google often shows the Google My Business at the top and the Knowledge Panel underneath for local businesses. But increasingly, when you’re not in the local area, it will show the Knowledge Graph.

And I was looking at my sister’s Knowledge Panel today and also trying to figure out how many Knowledge Panels appear with Knowledge Graph presence, and how many Knowledge Graph presences don’t result in a Knowledge Panel appearing. The second is much much bigger. And what I noticed is that a lot of local businesses are being shifted over to the Knowledge Graph as well. As a local business you’re going to want to manage which one’s being shown, but you have no direct control - you can’t just say to Google “show the Knowledge Panel.” All you can do is try and influence it. But you’d better claim it because it’s the only control you’re gonna get.

Google My Business is a business listing - it’s like the yellow pages. The Knowledge Panel is something completely different. It’s saying: these are facts about a company, and you don’t control them. That’s the point of it. Whereas a Google My Business - you have control. It’s a business listing.

And the fact that Google is shifting local businesses over to the Knowledge Graph is an interesting development. Kalicube tracks around the world across several countries - you do see the Knowledge Graph appear when you’re not local and not appear when you are.

Andrea Volpini: A very tactical suggestion: there is a very important property in schema that allows you to do exactly what you need - schema:hasMap. The hasMap property, used inside an Organisation entity type, allows you to connect the entity of your organisation with its Google My Business. The value you would place is the Google Maps link you get for your GMB. That’s a fantastic way to tell Google: this organisation is linked to this Google My Business. It gives you control - Google My Business will appear for local searches only. If you look at WordLift as an example, we do both: the Bing Places and the Google My Business. When you are in Rome or around us, the GMB is more prominent. Otherwise you get the Knowledge Graph presentation of our organisation as a software company. I have this extra level of control by linking data inside my schema. That’s super powerful.

Q&A - Cameos:

Jason Barnard: Somebody asked about the Cameos and how did Andrea get them.

Andrea Volpini: It’s invite-only - it’s really a search experiment. You have to get to some level of clout. I don’t have that clout - I don’t know, probably they got me wrong. My challenge when I started was really: what can people do if they’re not notable enough to end up in Wikipedia? That’s really when we started with the idea of publishing a personal knowledge graph that enables everyone to become their own Wikipedia. I think that’s fundamental.

Cameos is still invite-only, but it’s super interesting because it will create questions for you based on the facets that the Knowledge Graph holds about your entity. The requirements: first, your entity needs to be in the Google Knowledge Graph. Second, you need to have multiple facets - they need to know at least your occupation. For me, it only asked me about how many companies I’ve founded and things about startups - because that’s what I do. Once you’re in, they build up questions based on what they think - through direct links - which is completely logical. They’re trying to get you to provide content to their users on that SERP.

Jason Barnard: I have one last question I’d like to throw out: the idea of schema markup - it doesn’t have an immediate payback except maybe the pretty things in the SERP. How do you sell it to your boss? How do you sell it if it’s not just the pretty search features?

Andrea Volpini: I sold it in 2011 - so imagine how easy it is now. Right now for schema it’s so easy to demonstrate. With some enterprise clients we actually look at the number of clicks that WordLift can generate for them in the first few months. Usually we see a clear uplift on average after the first month of using WordLift in the proper way. You measure it in clicks, apply the conversion rate, calculate the business value behind those conversions, and then calculate if it’s worth the effort. Of course you don’t need to structure everything - you focus where the core of your business lies. But right now, structured data ROI is not an issue.

Dixon Jones: On certain types of schema it’s very easy to show. If you’re something and you start being an FAQ schema and you’re the only one answering the question, that’s a pretty good sign. We’re doing experiments right now with WebPage sameAs schema - we’ve got around thirty different SEOs playing with whether putting in that schema changes rankings - and we’ll hopefully have a decent case study to demonstrate that at least for WebPage schema, schema has a value for the core algorithm. And that’s why I try and describe it in this “single point of truth for Google” concept - because then a CEO or CMO says: “Oh - it’s not just my website that’s important. It’s everything that counts.”

Jason Barnard: So we’re going to be looking for case studies to come out to actually prove the ROI of this stuff, to get bosses and clients on board when they’re reticent today. Thank you very much, Joe.

SEO Joe Bloggs: Thank you very much to Jason for hosting us today, and thank you very much to our speakers for coming on board for Tea Time SEO. I’ll pass over to Kerry for some concluding remarks.

Kerry Shepherd: Just remember that we have a fuller blog summary from last week and this week going up, and we will see you next week. Have a good weekend.

Jason Barnard / Others: Thank you - have a great weekend. No goodbyes - I like that!

End of transcript.

Panels documented: 17 Running concept list staked in this transcript: 60%/40% Wikipedia split (data point, spring 2020) · 391 non-Wikipedia sources baseline · Budapest update named as “last summer” (summer 2019) · Sprouts concept (earliest recorded) · GMB = yellow pages / Knowledge Panel = fact record distinction · hasMap for GMB connection · Two separate Google crawlers (Volpini IP observation) · Entity engagement as next frontier · Schema forces content model clarity (continuation of Yoast thread)