Episode 119: Knowledge Graph and authors, new link reclamation tactics and how GSC reports impressions

Play this episode:

Or get it on:

• Apple iTunes
• Anchor
• Breaker
• Google Podcasts
• Overcast
• Pocket Casts
• RadioPublic
• Spotify
• Sticher

What's in this episode?

In this episode, you will hear Mark Williams-Cook talking about:

Google Knowledge Graph authors: How Google is taking further steps in combining author and article data within its Knowledge Graph and showing rich results.

New link reclamation tactics: Good this bold new tactic work for you?

How GSC reports impressions: Interesting tests showing how you don't need your site to appear in the search results to get impressions.

Show notes

Google expanding KG to connect articles to authors and sentiment https://blog.google/products/news/learn-more-about-person-behind-news-article/

Episode 85 with Lily Ray https://withcandour.co.uk/blog/episode-85-all-about-googles-eat-ymyl-and-patents-with-lily-ray

SE round table: Again, Google Does Not Use Sentiment For Ranking Purposes https://www.seroundtable.com/google-sentiment-not-for-ranking-purposes-31632.html

Arnout Hellemans tweet https://twitter.com/hellemans/status/1410550321928101889

Tweet regarding new outreach technique... https://twitter.com/Dominic2306/status/1412053718900555781

Transcription

MC: Welcome to Episode 119 of the Search with Candour podcast recorded on Friday, the 9th of July, 2021. My name is Mark Williams-Cook and today I'm going to be talking to you about some creative link reclamation tactics that we've seen in the wild. We're going to be talking about some interesting ways Google Search Console is collecting impressions data and as well having a quick discussion around how Google is making another effort to link up authors and articles and what this might mean for search.

Before we kick off I want to tell you, this podcast is still very kindly sponsored by the wonderful people at Sitebulb. If you haven't heard of them, Sitebulb is the SEO-auditing piece of software. It works on Windows, it works on Mac. I've used it alongside all of our other SEO tools now for years. We use it at Candour Agency hand on heart, it's my favourite piece of auditing software for many, many, many reasons that if you listen to any other of these episodes I literally go through each of the features that I like about it.

But we just had some new people start with us at Candour and some of them hadn't used Sitebulb before and it was a pleasure getting them to use it because unlike other auditing tools that I previously used, Sitebulb has it's crawler but then it has this whole analysis engine where it's taking that data and actually trying to diagnose issues for you, and they have this huge built-in repository of information that they've built on the web about all of these issues and they make a really good first stab at actually prioritising the issues for you. So obviously like with any tool you're going to need to go through your own validation and prioritisation process in context to your business and the resources you've got. But this does take out a lot of the legwork and it does allow people who maybe have weaker areas of knowledge in some places to get a massive head start.

If you're listening to this podcast, it means that you can get a special extended trial of Sitebulb. So there is a free trial normally you don't need a credit card, you can just go and download it. But if you go to sitebulb.com/swc, that's sitebulb.com/swc, you will automatically then be able to get a 60-day trial, that's two months of trying out their software. There's no obligation, as I said, no payment details needed, so if you don't like it you don't have to have it, but I'm sure you will so do give it a try.

I want to kick off with something that I just found really funny this week. It's incredibly niche humour so I can't really tell it to anyone who doesn't work in SEO or digital PR, so I thought I would share it with everyone that listens to this podcast. As we all know, link acquisition, digital PR, link-building, link-begging, whatever you want to call it is still a big part of SEO. A lot of the stuff we do around content is around trying to get people to link to us because we still know that most major search engines use link graphs to work out which sites are good, which sites are useful, whatever kind of term you want to use. It helps us rank, we know this.

And one of the trickiest aspects and one that requires a lot of experience or tact is dealing with journalists. And not because they're difficult people to deal with but because we want something very specific from them and normally because of time and budget restraints we're sometimes limited in what we can offer them of requisite value to get what we want, which is essentially coverage and, hopefully, a link. So we've probably all been in a situation where we've done some marketing, some outreach, and we've secured some coverage, which is brilliant. And then you have that realisation, when you look at the article that I'll know, it doesn't have a link and normally, well, you should be trying to, if possible, get someone to actually add that link in for you.

Now, this is where it gets really difficult because I really feel for journalists because they are absolutely hammered by SEOs and the majority of them are producing very poor emails just demanding links and it must get incredibly tiresome. So much so to the point I've seen journalists complaining on Twitter about SEOs and digital PRs, starting emails with things like hope you're well, because that's the pretty standard opening line to someone that you don't know a lot about personally, "Hope you're well. By the way, here's the thing I want." And this created some discussion about, well, what should we say and how can we make this work for both parties? So that is well-known, that's an incredibly sensitive topic.

Now, what I found so funny this week was there was a tweet by Dominic Cummings. For those that maybe aren't in the UK and not sure who Dominic Cummings is, he was essentially a political strategist who was an advisor to our prime minister, a little bit disgraced during the coronavirus pandemic for breaking quarantine and going for a drive and he was in all the newspapers, ended up resigning, yada yada.

Anyway, what happened was a BBC reporter had written a story and referenced something that Dominic Cummings had said, and this is what Dominic Cummings tweeted back to him. "Oi," so he has the attention with an ‘oi’. "There's no link to my blog in your online piece, please insert or I'll cancel my imminent interview with BBC. Quoting is okay but must link to the original..." Which I imagine if you're anyone but a well-known public figure making that statement publicly is going to get you a very short reply from most journalists.

And he replied to this, "Fair point, people might want more than edited highlights," and added the link. And I just found this absolutely here. There is a link to it in the show notes at search.withcandour.co.uk. But Dominic Cummings essentially, yeah, doing his link reclamation in an incredibly abrupt and rude way and it worked. So maybe that's it, maybe we're being too polite and we just need to start calling out journos and saying, "Oi, add our link in on Twitter." Now, if any of you want to try that, give it a go and please let me know how it goes and I will feature it in another episode.

Okay, on to some real news now. This one was flagged up to me by Carl Hendy that Google appears to be making another concerted effort to link authors together with articles, the things they wrote through their knowledge graph. And they've done a blog post on the keyword blog about this and, of course, as usual search.withcandour.co.uk for the show notes and I'll link you to it. I'm just going to talk you through it and then actually it does relate to a previous episode we did.

This blog post was by Jen Granito who's the Senior Product Manager of Search at Google and it's titled ‘Learn More About the Person Behind the News Article.’ "Individual voices play an important role in the news and information we consume. With so many complex important stories unfolding daily, people not only rely on specific publishers for the latest news but also increasingly turn to trusted individual journalists, authors, and experts. We're committed to helping people access timely, authoritative news, and information from a variety of different sources, as well as investing to help sustain a quality news ecosystem.

"Earlier this year, we launched full coverage in Search to make it easier to explore all aspects of a story from different perspectives. We also launched the fact-checking open fund to help journalists around the world correct misconceptions that circulate online and offline. Today we're rolling out a beta feature in Search to help people learn more about an individual journalist or author by more prominently highlighting their recent work. Currently, when you search for a certain journalist or author you might see an information box, also known as a knowledge panel, at the top of your results that provides a quick snapshot of information about that person.

"Starting today for a small subset of these journalists we'll begin including a carousel of their published articles on the search results page. This carousel will help you quickly understand what subjects that person has recently covered and make it easy for you to find some of their latest work. To start, this new feature is only available for a limited number of U.S. English language journalists and can be accessed on mobile devices. We're looking to expand the feature over time to more journalists, devices and languages, but we're going to be testing out the different ways of organising the content to ensure we optimise for the best experience.

"To that end we're actively looking for ways to improve the experience and ensure it consistently shows the most useful information. Anyone can submit feedback by clicking the feedback button in the bottom right-hand corner of the articles carousel, publishers and content creators can help make their content more accessible and enhance their appearance in Google Search results by adding structured metadata to the article pages, including the journalists or authors name and bio pages for their journalist describing their expertise."

I found this particularly interesting because way back in episode 85, we had an interview with Lily Ray and we talked all about expertise, authority, and trust, E-A-T, and we widely speculated on the future of Google. And if my memory serves me correctly, we talked about various paintings Google had for things such as identifying individuals' voices online in podcasts like this. So they might know through not only a text writing style but through a voiceprint that this is me, Mark Williams-Cook talking about SEO as usual.

And we were talking about how these things might become ranking factors, how machines or Google Search are essentially trying to emulate the things humans use to make decisions about trust. And this is what we're talking about here. While websites, domains, and URLs can hold a certain amount of trust in that as a human, if I go to a website that I know is trustworthy and well-respected, anyone writing on that website for me has a certain amount of trust inherited from that because that brand has decided to let them write.

However, the reverse is true in that if someone who I trust or who is a particular expert writes on a website, this kind of nobody, if you like in the online space, that's not particularly well-reflected by search engines at the moment, but it should be. Because if I went and wrote an article for search engine land, why should that be any more or less important than an article I write and put the same amount or more effort in for somebody's blog that doesn't get a lot of visitors?

I think it's super interesting now that not only are we seeing this data being actively surfaced, so this is being added to the list of types of structured data that can give us enhanced search results but I think it is signaling towards, hey, look, this whole tying up authors articles and becoming known. And again, we did another podcast with Jason Barnard about knowledge graphs and knowledge panels of individual users. It's all becoming very apparent that this is very important so we should be taking steps to do this.

To give you an example, we had a client that we work with that deals in a medical-sensitive area and they have lots of blog posts on their site that are dealing with this subject matter and the author for a lot of them was just an admin. And I was saying to them, again, not even thinking about search engines, if a user lands here and they're trying to decide whether this is good information or not, it's not really doing us any favours that the author is an admin. We ideally want to see the name, qualifications, credentials, and experience of the person that's writing, so that we can make a decision on how much we're going to trust the information we're seeing.

That in itself is super interesting. And kind of tied to this, I saw an earlier... Well, not early this week, actually at the end of last month it was, that there was a headline on search engine round table that says, again, Google does not use sentiment for ranking purposes. So this is related to me in terms of the discussion of knowledge graphs and taking into account factors that search engines use to make judgments about what to rank. And this article was pointing out in 2018, Google's Danny Sullivan, said that Google does not recognise sentiment in Google Search. And he was asked again in 2021, so it was a tweet by Mark Sampson that says, "Google does not recognise sentiment, is this still true today? With the predatory site algorithm and sentiment analysis capabilities in natural language processing has this changed? Does Google now consider sentiment in content about people, reputations, slander, et cetera?"

And Danny Sullivan has given a great breakdown in his answer, he's written, no. While that might be seen as a clear answer, there is something I want to point out before we just take this as, no, Google does not use sentiment, which is that we've spoken about links quite a few times now on this podcast and this episode, in fact, and what are links if not an expression of sentiment? Meaning the reason search engines like Google pay attention to links is that there is such a strong sentiment that something is relevant or something is helpful or content is good that someone links to it, that's what they're listening to.

Now, the question does specifically say things like consider people reputation or slander, and Danny's answering no to that and I think we can take that as right, as correct, but I would keep an open mind when it comes to what are we defining as sentiment and in relevance to what environments to websites in relevance to individuals and how are we measuring that? I think it's an incredibly hard thing to quantify and obviously a lot of the time in sentiment there isn't really an objective answer. You could have different answers, even culturally or across regions, so it's a very difficult puzzle for an algorithm to tackle, but I would consider the different ways that it could be measured and the ways we see it expressed in things like links before we just write it off as a no.

I wanted to end the episode on something particularly interesting I saw at the end of last month tweeted by Arnold Hellmans. And it's about Google Search Console and how impressions are counted. Now, maybe I'm late to the party and everyone else knew this before me, but I wanted to talk about it because I think there's a fair chance at least some of you don't know this. That the summary is that Google Search Console will count impressions for your domain, wherever your URL appears in that SERP, such as if I mentioned your domain in a tweet and that tweet ranks for a specific keyword that your actual website/ your domain does not rank for, Google Search Console will record that as an impression because your domain has appeared for that search term on the Google Search results page.

I'll just repeat that because it is a little bit hard to follow. Essentially, Google Search Console will count impressions whenever your domain appears on the search result, even if it's not a listed result. So the example that Arnold gave was if someone tweets your URL and that URL appears in a search result, you'll get impressions counted in Google Search Console. Now, some people apparently already knew this but he did... Annette went one other step to do some other testing on this because there were some questions around, well, what happens if there's things like URL shorteners, do the impressions get canonicalised?

So if there is... We know from the initial experiment if someone tweets my domain name or my URL and that appears in a search it will get counted as an impression. But what if someone tweets a shortened version of my URL that maybe they've used a shortening service for? Will Google see that URL, follow it, and then canonicalise the impression back to my domain, even though it's a URL shortener version appearing in the SERP I'm going to see the impressions in my search console, or does Google just record what it sees?

And Arnold had a nice test for this because he had a domain that didn't have any pages life that he was actually using as a URL shortener service so he tested tweeting some of those shortened URLs which were triggered by a keyword. And he found that the impressions were appearing in the search console for the shortened URL domain, so for the actual URL shortener domain. This means you would not see those impressions counted in your actual website that people are being directed to. It's just a little bit of insight into how Google Search Console is counting impressions. So it's pretty much a very, very basic approach of if it sees your domain, your URL appearing anywhere in that document object model for that search channel you're getting an impression, it doesn't matter if it's URL shortened or not.

I thought this might be helpful because I didn't know this as I said, and sometimes there's been some weird key phrases that I've spotted in search console that I know the site doesn't rank for, it doesn't have a page ranking for it and I couldn't track down where it was so I assume it's going to be situations like this that's triggered those impressions. I thought you'd enjoy that if you didn't know about it. If anyone else has got any more data on it I'd love to hear it. It's really cool seeing these little experiments coming from the community and just learning a little bit more about how Google Search and their various tools do work.

That's everything I've got for you in this episode. If you'd like to join me on our e-commerce live event I'll be with you on Wednesday on LinkedIn, so you can just add me - Mark Williams-Cook on LinkedIn. And on Wednesday at 9:00 AM BST, we will be doing another session with Quickfire Digital about e-commerce SEO. Until then, we will be back, of course, in one week's time which would be Monday, the 19th of July, and I hope until then everyone has a lovely week.