Everett Sizemore

Editor's Notes

1. MOVE ON. Nothing to say here.
2. Here’s what we’re going to discuss today. Hopefully there will be a few takeaways even for those of you who have successfully edited your knowledge graph listings already. 1) All of this linking and defining of data is pretty cool. But what does it mean for your bottom line? Will it help you get more traffic and make more money? In other words IS IT WORTH YOUR TIME RIGHT NOW? 2) How can you mark up your own website to help Google understand your brand? We’ll start with the basics: What does the code look like and where do you put it? 3) And finally we’ll take it a step further and make sure your brand is listed as an entity on the Data Sources most likely to be used by Google to populate their Knowledge Graph boxes.
3. JSON stands for Javascript Object Notation and LD stands for Linked Data. So we're placing very simple Javascript into the HTML Header of your page so we can provide meaning and context to machines like Googlebot.
4. It’s important to note here that it’s not guaranteed that you’ll get a knowledge graph listing - even after doing everything in this presentation. But doing these things WILL improve your chances. Maybe Google already displays a Knowledge Graph listing for your brand, but you would like to Correct it or ENHANCE it in some way. For example…
5. Sur La Table’s Knowledge Graph does not contain any brand-owned content. The RELATED SEARCHES may be taking traffic AWAY from Sur La Table. But what can they do? Google will show the Knowledge Graph whether Sur La Table wants them to or not. So the only thing you can do is to take some CONTROL of the situation by enhancing that box. The further you go to the right of this image, the more enhanced and customized the Knowledge Graph boxes become. Williams-Sonoma makes full use of its Knowledge Graph box with Social Profiles, Posts from Google+ and other information. Notice how far down the Related Searches are, and how much more real estate Williams Sonoma will have in the search results for branded searches.
6. There is a company called Inflow Inventory that develops Inventory Management Software. They’re not our direct competitor because we’re in different industries, but they have been around longer and rank #1 for the word “Inflow”. However, notice which company has the knowledge graph box on this search, which invites users to search for our brand instead, while also dissambiguating which is which by using the phrase “Internet Marketing Company” or “Online Marketing Company”. Having a knowledge graph listing allows us to regain more of that potential lost traffic from the other company. It gives us an advantage.
7. It is no accident that Google uses a phrase like "online Marketing Company" to describe our brand. SICcode.com makes it easy to get BOTH your NAICS code for North America AND the SIC code for the rest of the word. These are used to define which industry your brand is associated with in a machine-readable, database-friendly, easily searchable format.
8. This Answer Box, derived from data found in the Knowledge Graph, proves that Google knows the nature of the relationship between two entities: Mike (The CEO) and Inflow (His Internet Marketing Company). So far - ALL these screenshots were taken before we added the Wikipedia and WikiData pages for Inflow. The information in them comes from two sources: Freebase – Which is not in operation anymore. And The JSON-LD script on our home page.
9. And This Is Kind of Cool.
10. Hopefully by now everyone is onboard with the WHY. Let’s move on to the HOW.
11. This is the Person Schema that I’ve put on my Author page. When I first put it up I had a Knowledge Graph listing for my name before I had a WikiData page. It was coming from Freebase. I migrated it over to WikiData, but then an asshole WikiData editor took down my page and the Knowledge Graph went away, even though the Freebase listing was still up. So don’t expect to go put JSON-LD script on your site today and have a new Knowledge Graph listing next week – unless you have a Wikipedia page. Because, as we all know, Google is FINICKY.
12. So don’t expect to go put JSON-LD script on your site today and have a new Knowledge Graph listing next week – unless you have a Wikipedia page. Because, as we all know, Google is FINICKY and Open Data Editors take their work VERY Seriously.
13. This SCRIPT is placed in the HTML <HEAD> area of our home page. JSON-LD allows us to put the Markup in the HTML Head area so our human visitors don’t have to look at all of that junk, which is there just to provide context to a machine.
14. In that way, it is very much like the old days, back when we used to focus mostly on Meta Keywords and Meta Descriptions. Things that most human visitors would never see. Back then Meta Tags were the primary way of explaining a page to search engines. Now we are using Schema Markup and Linked-Data. We have come full circle
15. Some of you may remember when writing Meta Tags all day was part of the SEO Job Description. Many of us, myself included, don't come from a CODING or development background. That’s OK. This stuff is pretty easy when you break it down into its component parts.
16. The highlighted code here starts off with THREE very important pieces of information: #1 – This is a script. Specifically, it is Linked Data in JSON format. (JavaScript Object Notation) #2 – Our Linked Data is using the Schema.org Vocabulary #3 – The Schema type we’re using from that vocabulary is Organization
17. Now Google knows that our JSON-LD script is following Schema.org’s Organization Schema. And this opens up many different related Properties that help you explain your business to Google. For example, you can tell Googlebot in a machine-readable language the names of your employees, and their official job titles. You can define your official social profiles, and even connect SUB-Organizations with PARENT-Organizations.
18. In addition to the street address, you will want to include the GeoCoordinates in the form of Latitude and Longitude. You can get these coordinates from a Google Maps listing of your business.
19. Here’s a better view of how to get those coordinates.
20. Here is where you start connecting this entity to the other places on the web that represent “Official pages” about the organization. We’ll talk more about some of these sites later, but even if you don’t have a WIkiData or Wikipedia page you can still get a Knowledge Graph listing (sometimes) by linking to your official social profiles and by providing the rest of the information in the script. The SameAs tag says, these are all "official" pages about this Organization.
21. Once you add the code to your site, head over to Google’s Structured Data Testing Tool to make sure there are no errors.
22. Now you know what to put on your site to get the BASICSs out of the way. You have told Google and the rest of the Internet a little more about Your Brand. It’s time to start looking for more ways to grow and connect your entity”.
23. Anyone can provide feedback on any Knowledge Graph listing.
24. Here’s what you see after clicking the “Feedback” link.
25. I used the Feedback feature to ask why our social profiles weren’t showing up, even though I had the SameAs tags in our HTML header. Within a few days our listing was displaying all of our social profiles.
26. Barry is listed as a “Technologist” on Wikipedia, which is how he has this listing. I took this screenshot from a Marketing Land blog post. I accessed this for Inflow using Mike’s account and was able to update the URL being presented for our site. It had been using a URL-Shortened version. Sorting out the official Google account associated with your brand can be a pain in the ass, just like anything involving Google accounts.
27. This is a good point, and one we should take a moment to think about. [PAUSE] Is it OK that Google doesn’t think you’re important? Wouldn’t being considered important by Google probably also be good for SEO? So the question question isn't – “WHY should we do this?” It's the more actionable one – “HOW do we get Google to think we’re MORE important?”
28. The answer is to grow your entity and connect it to other entities. It’s the same concept as linkbuilding. The more entities pointing to you the better. And the more entities pointing to THEM the better for you too. BUT DON’T BE SPAMMY. It won’t work. Everything you do will be undone by editors, and you will have wasted your time, along with everyone else’s. There are lots of sources of Open Data on the web. Any reputable open data source is a potential opportunity to Define and Connect your Entity. If you can help Google sort through these connections and understand your brand you will be more likely to get the information you want to appear in the Knowledge Graph. So let’s start looking into some of these Data Sources... ----- Meeting Notes (5/9/16 10:16) ----- Trust me, I speak from experience. Wiki Data editors are Nazi's, just like Wikipedia editors - maybe worse.
29. WikiMedia Commons, Like WikiPedia, is owned by the WikiMedia Foundation. The purpose of WikiMedia Commons is to provide a repository for Public Domain and Creative Commons media. A good place to start would be to upload your company logo and headshots for the executive team, or anyone in the company who speaks at events, writes or gets interviewed. Other things to add would be photographs and video of events, and sound files like a radio or podcast interview with the founder.
30. WikiData will allow you to provide a few VERY important pieces of information in a way that machines can understand. Things like Company Name and Description, Industry, Website and Logo are just a few things you can put on here. Google used to use a similar site called Freebase, but it has since been deprecated. If you already have a Freebase page a Volunteer Editor may eventually move it over to WikiData. But don’t wait for that to happen. You can add it yourself, or hire someone to do it.
31. If you or anyone at your company has published a book, the Open Library Project is a great way to get information about them into a Machine-Readable format. They’ll even accept eBooks, which is what I have added here. The author is our Conversion Optimization expert Keith Hagen, who now has his own Author Page on the Open Library Project. You’ll see Inflow is listed as the PUBLISHER, which means the brand gets its own Open Library Page too.
32. I took some quotes from Keith Hagen’s eBook seen in the previous slide, and added them to WikiQuote. They stayed up for about 48 hours before an editor took them down. The reason given: Keith was, apparently, not Notable enough in their eyes. I disagree, but can appreciate the need to maintain these standards.
33. DBPedia is a community effort to extract structured information from Wikipedia, and to make this information available on the Web. Basically, Wikipedia is built for Humans, and DBPedia is built for Machines. This slide has a lot of text, but DBPedia is a critical component of what we’re talking about today. So I’ll give you a moment to read. [PAUSE] That is an enormous amount of data-points, all connected to each other to form the structural skeleton of THE WEB OF THINGS. ----- Meeting Notes (5/9/16 10:16) ----- Change "data-points" to "data about entities,..." and "the web of things" to "THE STRUCTURED WEB".
34. About Us is an easy profile to get, and includes all of the basic information about your company. I don’t think Google uses them directly, but they may provide data to other Open Data Sources.
35. Wikipedia is a must-have for any brand. But they can be picky about which brands and people they include. The trick is to have press. The more you can get national and global publications to write about you or your brand, the easier it will be to get a Wikipedia page to STICK. So if you don’t have enough Press Coverage yet, that is a sign that you should do some PR work. Get someone from your organization interviewed, offer quotes to the media about your industry, send out interesting press releases, and do the paperwork necessary to be included in lists like the Inc. 5000 or Internet Retailer Top 1,000.
36. Marking up your HTML Head with JSON-LD allows you to add the SAME-AS tag. This is your opportunity to provide a BI-DIRECTIONAL link to the other sources we just discussed. Essentially, it says “WE AGREE THAT THESE ARE OFFICIAL PAGES ABOUT US”. ----- Meeting Notes (5/9/16 10:16) ----- Remember Authorship Markup? It's sort of like that.
37. Anyone who has ever tried to create a new Wikipedia page, or to use Schema.org markup, can tell you that it’s not as easy as you might think. Wikipedia has a lot of rules to keep people from spamming it, or uploading false information. Sites like these require all entries to follow very specific formats and rules that are different for each data-source. In other words, it can be confusing to anyone who doesn’t specialize in this sort of thing. The Solution? [NEXT SLIDE]
38. HIRE SOMEONE! For a few hundred Euros you can have someone do this for you. Also, once you get all of these pages up, make sure you add them to the Links section on your Google Plus “About Us” page. ----- Meeting Notes (5/9/16 10:16) ----- Change Euros to Bucks.
39. It’s easiest for us to comprehend this stuff if we think of ENTITIES as a “HUBs and SPOKEs”. And if we ZOOM IN like this it makes sense. BUT… the further BACK we step, the more we see…
40. THAT IT’S A MESS What you see here is just a list of entities that Rely On or Get Data From -DBpedia. And DBPedia gets their data from Wikipedia" which CITES its "Information" from lots of different Sources. In other words, this is a picture of only ONE HUB and its MANY spokes – which are themselves major hubs. And if we take another step back to View a more Complete Picture…
41. [PAUSE] They Call It STRUCTURED DATA. But It ends up looking MORE ORGANIC than STRUCTURED. The thing about ORGANIC SYSTEMS is: THEY ARE DIFFICULT TO MANIPULATE. That is why THE BEST WAY TO MAKE GOOGLE THINK YOU’RE MORE IMPORTANT ---> Is to actually BECOME more important. In other words, GO BUILD YOUR BRAND.
42. Here are a few more resources for you to check out when you get home. THANK YOU.