E-E-A-T Fireside Chat With Lily Ray
During a recent fireside chat, Crowd Content’s CEO Carlos Meza talked with Lily Ray, Senior Director of SEO and Head of Organic Research at Amsive Digital, about the new E in E-E-A-T. Lily Ray shared her thoughts on Google’s new search quality guidelines and the repercussions of ChatGPT and other AI-generated content. What follows is a transcript of their conversation.
Carlos Meza: We’ve gotten amazing feedback the other two times we’ve had Lily Ray here. The live SEO clinic was one of our highest-rated webinars based on a survey of attendees. We’re excited to have you here.
Lily Ray: I’m happy to hear that. I have a lot of talks coming up this year. Recently, it feels as if there’s so much new information in the SEO space that it’s throwing us a curveball. Although that’s normal in SEO, lately, there’s a lot of information all at once.
Carlos Meza: It sounds as if there’s a lot going on in our industry. This year will be an interesting year to watch, given all that’s going on, such as AI. I’m sure you get asked this question every single time, and we’ll probably cover a little of that.
Thank you for joining us, Lily. I’m sure our webinar attendees are excited to hear more about the new E in E-E-A-T. I wonder how many new Es they’re going to add, because there’s an implied E around effort, as well. We’ll also talk about that. So, why don’t we get into it and talk about the new E in the acronym. What is it and what are your thoughts about it?
The New E in E-E-A-T: What It Means and Why It Matters
Lily Ray: The new E stands for experience. Google introduced it to the search quality guidelines in December 2022, two months ago. As they started to visualize E-E-A-T differently, they also repositioned it. Now, we have T, which stands for trust, at the center, surrounded by E, E and A. Essentially, the extent to which content has experience, expertise and/or authority lends to how trustworthy it is. So, while trust may be the most important factor, there are several ways you can arrive at being trustworthy.
The introduction of experience was interesting, although not too surprising because of what Google started to do with their product review update. They really started to talk about the importance of firsthand experience, which had previously been reserved primarily for product reviews.
If you read the search quality guidelines, they mention product reviews as the first example of the importance of experience. But, there are other situations, in which firsthand experience should be sufficient to demonstrate that content’s trustworthy.
Carlos Meza: There are a lot of things to unpack in that introduction. Can you mention one case where firsthand experience would be sufficient to demonstrate trustworthiness? What comes to mind when you mention that?
Lily Ray: Interviews are the most obvious example. If you paid attention to what Google said about the product reviews updates and the related documents, such as the recommendations for site owners, they talk about needing evidence that you used the product, such as pictures and videos of you using the product. They don’t ask you to write in the first person, but you can read between the lines. It seems as if they’re really looking for wording, such as: when I did my testing, I experienced this and this thing with the product.
I wrote about this a couple of times when I talked about the product review updates. Last year, I actually did a number of talks and articles about how I think expertise is something Google’s focusing on but that they’re taking it in a different direction, where it’s separate from expertise. It’s experience, because you can be experienced in something without being an expert.
So last year — and maybe the year before — Google started getting a lot of bad press around not having enough results that demonstrated that users had really experienced the thing firsthand. So, people were going directly to Reddit, for example. Late last year, Google launched a new SERP feature, Discussions and Forums, where they’re elevating content from Reddit, Quora and other major forums where people answer questions.
That’s another example they talk about in the quality guidelines. If you’re a cancer survivor, of course you have experience. It doesn’t mean you’re an expert in cancer. It just means you’re an experienced cancer survivor. So, if somebody’s looking for what it’s going to be like to have cancer, a question that can be answered sufficiently by people who have shared that experience, that content can rank.
It’s an interesting nuance, but it’s a way for Google to show that they have relatable information in the search results.
Experience vs. Expertise: How They Affect Search Result Rankings
Carlos Meza: That begs the question: What’s the main criteria to be first? Is it experience or expertise? Maybe there are two parts to this question. Can you have both — and if you didn’t have both, which one would help you rank number 1 or at least higher? If there’s somebody with expertise and somebody with experience, maybe the one with experience ranks higher. Do you know where I’m trying to go with this?
Lily Ray: Yes. That was the first question I had. When is one sufficient to rank over the other? I think it comes down to the query and its intent.
My team and I are looking into this. There are a handful of modifiers that generally return experiential results. So, when you type “best,” you’ll generally get an experiential result. Anything that implies a search for a subjective result — someone’s opinions, experiences or firsthand accounts — will probably result in experience being sufficient. But, when you type “treatments for cancer,” it has to be expertise and authority.
In their quality guidelines, Google says only that quality raters should use their best judgment to understand how much experience, expertise or authoritativeness is required, given the nature of the content. So, it depends.
Carlos Meza: SEO’s favorite response: it depends. I hear you. The first thing that came to mind when you answered the question is this. I know we, as humans, are lazy. We want to type the least and have Google read our minds. Is this going backwards, because to get the best results, we need to be more specific, such as best how?
The way things are evolving, it’s almost like Google can read your mind. You just need to type a few characters and it starts guessing what you’re thinking. So, how can we think about that?
Lily Ray: It’s tricky. We’re working with several clients to try to figure out whether a page needs more firsthand experience to show it’s the best result. We’re working with some clients to pull in more UGC or talk to experts in the field. In general, that’s a good practice, because it shows you have a unique perspective or content.
But, I think the larger conversation, especially with AI content, is that now it’s fair game for everybody to write the exact same thing, including using expert consensus — and not just with ChatGPT. There’s another one, Perplexity.ai, which uses and cites expert sources. That’s the way it generates content. So, we can all write the exact same thing and have it all meet previous understandings of what E-A-T means.
But, AI cannot create firsthand experience, by definition. It’s interesting timing, because Google suddenly introduced new criteria defining what constitutes good-quality content, although that’s not to say it’s necessary for all content. But, that’s something that only human beings can offer. It’s also something that makes pages unique, so it’s an important thing to focus on wherever you can.
Carlos Meza: Perplexity.ai: Ask Anything is an interesting one. I think it’s a Chrome extension. That’s how it’s showing.
Lily Ray: They might have their own AI chat.
Carlos Meza: We’ll dig around AI a bit, because that’s always something people want to hear more about from you, to get your thoughts and experience.
Why E-E-A-T Centers Around Trust
Carlos Meza: You also mentioned trust when we started talking about E-E-A-T. I looked at the Venn diagram and how trust is now in the middle. Help us think more about that. What does having trust in the middle really mean for people or SEOs who are trying to figure this out?
Lily Ray: There are so many elements that Google refers to. But, I think it has a lot to do with understanding:
- Who the brand behind the content is
- Who the content creators are
- The incentive of the brand and the content creators (for example, reputation research)
This is something that’s been tricky for a lot of sites we’ve worked with and that I’ve looked at over the years with Google’s core updates. If you’re a law firm and you engage in an SEO program and your SEO tells you, “We need a lot of great content about what to do if you’re bitten by a dog,” the law firm is going to produce content that aims to get them clients. It’s just going to happen. If they had it their way, they’d probably write content that says you shouldn’t hesitate; you should call a lawyer immediately and here’s the phone number.
Google’s been cracking down on that type of content over the last several years, because it’s not helpful. Maybe you should call a lawyer, but what are all the options? I don’t only want to call this one lawyer. I need to know all the different options. What do I need to know about calling a lawyer?
Google is looking to provide truly objective content, but it’s very tricky for users. Trust is challenging for many sites because it might go against their best interests or they might not see the potential benefits. But, the more you can demonstrate to users why they can trust your content and why you’re objective, the better that content is going to perform for SEO.
Carlos Meza: What I’m hearing is that heavy self-promotion on your content probably won’t gain you trust. On the other hand, if you’re trying to be helpful, you can gain a lot of trust. Is that how we should view that?
Lily Ray: Yes. There are some examples of health sites, such as natural wellness sites or those that promote certain types of diets, that were hit very heavily in 2018 and 2019. They held a lot of market share in the space and did very well for SEO. They seemed to be able to rank for a lot of medical keywords. They lost a lot of visibility.
My team and I have done a lot of work, looking at the Wayback Machine to see what they’re changing, because they haven’t stopped trying to address the problem. They’re clearly making a lot of changes to the content. You can see the subtle ways they’re changing sentence structure. Previously, something may have said: this creates a lot of toxins in your body. Now, it’ll say: according to this and this source, having this thing in your body can cause XYZ concern. Then it links out.
The language is changing. It’s becoming less salesy in many cases. That’s what Google’s looking for. A lot of these examples have trended up in the past few years, but it takes a long time.
It’s a hard conversation to have with clients. You have to be careful about not pushing something on the user that’s not entirely trustworthy.
Building Sustainable Long-Term Growth
Carlos Meza: It’s a difficult balance, because clients are looking to get the fastest ROI, with leads and sales. They probably don’t want to hear that you have to earn that by providing a lot of helpful content. It’s hard to create a direct correlation between helpful content in the short term creating ROI. I believe in the long term it will. It’s hard for clients to understand the need to climb a very steep mountain before they can get the rewards. Is that how clients react sometimes?
Lily Ray: That’s probably the most frequent conversation we have with clients. This approach to SEO is the one that’s the hardest and takes the longest, but it’s the most rewarding. The best way to ensure sustainable long-term growth and minimize any risk of being devalued by Google is to do these things.
Are there other SEO techniques that get you greater results in a shorter amount of time? Yes. They all probably come with some element of risk. After Google’s link spam update last December, a lot of people who’ve never done anything wrong lost tons of traffic overnight. That’s because Google caught on to certain longstanding methods of link building. A lot of people were blindsided by that.
We can’t take those risks with our clients, so we like to do the right thing over the long haul. The problem is, it takes a long time, so certain clients might not see the results of this work until the next core update rolls out, after Google’s had time to reevaluate the site and see that it’s changing its course.
It’s stated in Google’s guidelines. If you read about their core updates, Google states that they need time to reevaluate your site, so you might not see recovery from a broad core update until the next update. In my experience and in the experience of others, it could even take two or three broad core updates, so it’s a long time.
Carlos Meza: I’m almost fearful to ask this, because it might be silly after what you just said. So, new quality rater guidelines came out in December, introducing a new E. It’s just February 2, but have you or your contacts seen any impact? Have you seen any signals in your shop?
Lily Ray: We’re looking for that. I don’t think we’ll necessarily see the new E manifesting in search results very much, very quickly. The search quality raters didn’t have the new criteria to work from until December. In theory, they’re scoring search results right now. Potentially, that data will get factored into Google’s algorithms for the next core update or maybe the one after that. So, it could take months before the work they’re doing with the new E is translated into the algorithms, if at all.
There are some interesting examples. These might be based on Google’s own understanding of intent and the way it ranks results. An example my team uses is CBD oil for sleep. A number of doctors and authorities are talking about the benefits of CBD oil for sleep. That’s what you’d expect to see with most “your money or your life” keywords. However, there are two or three results that say: I tried using CBD oil for sleep for 6 months and here’s what I learned. So, there are expert results and experience results.
Whether it’s due to an algorithm update or it’s because Google’s always testing intent and what people like, more of these firsthand accounts are ranking. Essentially, users like to read that type of content.
Finding the Balance
Carlos Meza: That’s an interesting example. Back to the trust side of things, this might be a difficult question to answer, but do you think one component is more important than the others? Are they all equally important, or does it depend?
Lily Ray: I think it depends on the query. We know that Google cares more about E-E-A-T for topics that can potentially cause harm. That’s what they say. So, the more harm-inducing content is, the more E-E-A-T matters. For content where E-E-A-T matters the most, authority probably carries the most weight. For heart attacks, COVID vaccines, 401(k) investments, terrorism — anything that’s life or death — authority is the most important thing.
In many of these cases, what you’ll see in the search results are Google’s own search features, which highlight Mayo Clinic and Harvard. They started to do some things with the UN, FDA and CDC. For anything COVID-related, you’ll see CDC SERP features. That’s authoritativeness. The top three or four results are generally always going to be those sites.
That’s the A in E-E-A-T. These are the trusted authorities. If you Google “tax keywords,” it’ll be the IRS. Yes, other people can rank, but authority is the most important component.
Then, for things where you need a doctor’s perspective or a financial advisor, expertise comes in. You can have CPAs, doctors and cardiologists — and their content — ranking.
Experience is probably less than “your money or your life.” Again, you’re looking for a firsthand account of something, so it’s entirely query dependent. But, I think the more “your money or your life it is, the more it ventures into that authoritativeness category.
Non-Your Money or Your Life Topics and Their Relationship to E-E-A-T
Carlos Meza: That’s a great point. Going to the other side of the coin, in non-your money or your life topics, how should you think about E-E-A-T? Is there a way to get an edge by focusing more on experience or authoritativeness? How should we think about non-your money or your life topics and their relationship to E-E-A-T?
Lily Ray: I always say this. Even if you’re writing content that doesn’t impact a person’s safety or well-being, or your blog or your website is about a topic on which Google says you don’t need high levels of E-E-A-T, why wouldn’t you include it?
I think the answer to the question is usually: I’m creating a website that I want to just rank for SEO and try to make a lot of money. So, I don’t want to put my name behind it. I don’t want to put the author’s name. I don’t want to talk about who I am. I don’t want to include an address because I’m just doing something for SEO purposes. In these cases, E-E-A-T doesn’t really enter into the equation.
But, for most legitimate businesses, for most people that are doing a personal blog or a personal brand, why not follow all the best practices? I always say this. Even if E-E-A-T has no impact on your SEO — which it will — it will impact your conversion rate. It will impact user experience on the site. It makes you look more legitimate. Other search engines like that. There’s no reason not to.
There are some exceptions, such as if you don’t want to share your name for safety reasons. I’ve worked with editorial teams who don’t want to put their names behind something because it’s a safety concern. Of course, don’t do that. But, generally speaking, everything we’ll talk about for E-E-A-T is a good practice for businesses overall.
Carlos Meza: What I’m hearing is if you can, why not — if it’s going to give you an edge anyway. And, if you’re just putting up a website for the sake of SEO, making some bucks, it’s probably not going to last longer because you’re not building that E-E-A-T. Is that what you’re trying to say?
Lily Ray: Yeah. People love to share examples with me of things that don’t have E-E-A-T in the ranking. Congratulations! You found an anomaly in a search engine with trillions and trillions of results. How is that possible?
Yes, of course it’s possible to not do these things and to rank, especially for Your Money or Your Life content — and a lot of content falls into that category if you think about it. There’s not that much content that doesn’t have an impact on somebody’s well-being, safety, happiness, health or financial security. Most of it falls into that.
I would just do it, even if it’s just for your customers because they’re all good practices overall.
Carlos Meza: That makes sense, because quality content is definitely a more sustainable strategy if you’re in the content business, which I think we all are in some shape or form. If you have any kind of business, you’re in an information content business. But, that’s a different discussion.
How to Scale Experience
Carlos Meza: Let’s talk about media publishers for a second. There are a lot of media publishers. They’re legitimate businesses, of course, but their business is information. They might have their own experts, but they also leverage outside experts because their main business is to produce content.
As they get bigger, and they start branching out into more topics or into the same topic, but with subtopics, how can you scale experience? Let’s say you have a travel blog. How can you possibly go to every single city and every single hotel and every single restaurant? So, how can we think about scaling firsthand experience?
Lily Ray: If Google was answering, they’d probably say don’t write about it unless you have the experience. But we all know that’s not possible. We all know we make money off ad revenue. So, you have to just do your due diligence to speak to people with the experience.
I’ve spoken to more journalists in the last several weeks than at any other point in my career. That’s because suddenly, everybody needs to talk to somebody with experience in SEO about what’s happening.
That’s what you should do if you’re a journalist. Hey, I’ve never been to this country or this city, but I’m looking to speak to people that have and get their firsthand accounts. It’s journalistic. It’s integrity. It’s speaking to people with the experience.
I think the mistake that many people are probably going to make this year and going forward, is thinking that rehashing the same generic content that already exists, using tools that are growing easier to use, more readily available and cheaper to scale content, is sufficient to create content.
But, what I think Google and the other search engines are going to figure out is how to find out who’s said something original. Honestly, more often than not, if you do speak to an expert — and they’re actually an expert — they’re going to say something completely new. Even if they’re saying something that’s already been written about on the web, they’re going to have their own perspective on it.
I think this notion that there’s nothing new to add to the conversation isn’t true when you actually speak to experts. There’s always something new.
Carlos Meza: Well, everybody has an opinion. So, what you’re saying is a great way to go about it is taking a journalistic approach. Maybe you don’t have the experience yourself, but can you find somebody to lend you that experience and put it in the content and bring that opinion to a piece? That’s a good way to put it.
The Intersection of ChatGPT, AI and E-E-A-T
Carlos Meza: I still have a lot of questions. Let’s go back to what you were mentioning about AI. So, ChatGPT came out in November. There’s been a ton of information around it and a lot of buzz. So, what’s your take on ChatGPT and everything that’s going on around the AI trend and machine-learning technology, which is very interesting? There’s now a new app popping up pretty much every day that is AI in some shape or form.
So, what’s your view on the world of content and also the relationship to E-E-A-T? I know you briefly touched on it, but I want to go a little deeper. Probably, some people in the audience want to hear about it.
Lily Ray: Yeah. There’s so much to say about it, but first, I want to be very clear. Nobody knows. Whatever I theorize, I don’t actually know. I don’t think anybody does. I think we’re all in a state of speculation and guessing. I think, including Google, nobody knows how it’s all going to play out. What we can probably safely assume is that over the next however many months, SEOs, content creators and others are going to be using these tools a lot to create content — which I have my own opinions about.
I wouldn’t recommend it right now, for the most part. Certain applications make a lot of sense, such as category description pages. Maybe you start with AI content. You tweak it a bit. That’s harmless.
Things that are boilerplate, that are hard to mass generate at scale, these tools can be helpful for that. Of course, you should review them. I think creating entire articles with these tools is pretty dangerous, even if the content is high quality.
Let’s imagine that Google finds a definitive way to determine that this is AI content and launches an algorithm update in three months that says: we’re devaluing all this content because it’s effortless. The search quality guidelines say that the amount of effort that goes into a piece of content matters for the quality of the content, and if it shows that it doesn’t have a lot of effort or skill it should be rated low quality. So, we were warned as far as that’s concerned.
They didn’t tell us that AI content is bad. They told us auto-generated content and content that demonstrates no effort, skill or originality is bad. If you read between the lines, mass auto-generating content with ChatGPT is essentially that.
That’s not to say it isn’t a good starting point. It’s not to say it’s not great for content ideation and keyword research. Many people are posting information about how to use ChatGPT for keyword research, keyword clustering and content ideas. Of course, do that. But, copying and pasting ChatGPT content onto the page probably isn’t a good idea right now, as it stands. Maybe when GPT-4 is released, my opinion will change.
Also, as somebody who likes to write content, and whose content has a good track record of performing well with SEO and social media, I like putting my name behind my content. I’d never use GPT in my content. The whole point of me writing is that I like to write. I like to share my ideas. I have my own ideas. Even if I’m sourcing content from different places, which AI could do, I prefer to do it myself. Maybe that’s an old-fashioned way of thinking about writing, but generally the content performs pretty well, so I’m going to keep doing that.
So, it’s really tricky. It’ll be interesting to see how Google responds.
I do want to say that I think there are a thousand Incredible use cases for GPT. I have been using it. I use it to create Excel formulas. I use it to create Regex commands, robots.txt directives and tabular data.
That’s awesome. That’s amazing. It’s already improved my workflow quite a lot. But, I would be careful about thinking it’s going to be the solution to your content scale problem, because who knows what’s going to happen.
Could AI Lead to Content Evolution?
Carlos Meza: So, how I’ve been thinking about it is back in the day, accountants used to do everything by hand. Then, calculators came in, and they didn’t replace accounts. They just made them faster and better. Then, Excel and other accounting software came in. If anything, before, accountants could handle a handful of accounts. Now, they can handle 30, 40 or even 50 accounts because they have all this technology.
Is this a case where AI will make content creators more efficient, so they’ll push the boundaries of what can be created, and they’ll create different types of content that includes other types of media? That’s just my guess or my opinion about where things could go. But, if content creators or writers are more efficient in some lower-end tasks by using AI, can they invest or reinvest that efficiency into originality, as you mentioned, or creating better content? Does that make any sense?
Lily Ray: Yeah. I agree. It’s fascinating to think about Microsoft incorporating a tool like this into Microsoft Word. That’s awesome. That would probably speed up the writing process for a lot of different topics that people are writing about.
Already, I saw a funny tweet that said: everybody wants to create AI content but nobody wants to read AI content. Maybe, as it stands currently, it doesn’t read entirely like a human all the time.
But, based on what Google tells us is a best practice for E-E-A-T best practices, we should tell users who created the content and have as much transparency as possible around how the content is created. Essentially, they’re asking us to do what CNET just did, which is to say, “If we’re being honest, this was created using AI content generation.”
That created a whole scandal. It turns out that some of the facts were wrong. It turns out that much of it was plagiarized. Now, that’s a bad mark against AI content, which could potentially influence how some people feel about it for now.
So, there are a lot of ethical questions we have to figure out. But, to your point, there are absolutely ways to use it as a tool to accelerate your process. I will be doing that. I think many people are going to be doing that. But again, what’s going to perform the best over time, once the search engines get a good handle on this, is truly the content that has something different, unique and original in it.
AI’s Potential Threat to Google
Carlos Meza: All of these conversations are around Google and E-E-A-T. Now, with things changing and Microsoft making all these big investments, do you think Google’s monopoly or market share is going to be eaten into by what’s going on around AI and ChatGPT? Some people are theorizing: if you can just ask a question to ChatGPT, why would you go to Google’s SERP to get a bunch of links instead of just getting a succinct answer? Some people are asking that.
So, do you see a world where Google loses some of their market share to some of these technologies? I know we’re just playing crystal ball.
Lily Ray: Generally speaking, I think the answer is probably yes. But, do I think it’s going to deal a death blow to Google overall? No, maybe not for now at least.
Number 1, ChatGPT is a shiny object that everyone is obsessing over. It seems as if Microsoft might be integrating it into Bing and other products, but we don’t know what that’s going to look like. We don’t know if and when this thing leads to some massive PR scandal where they have to turn it off. That’s happened with other AI features in the past.
We don’t know how cost effective it is to continue to have it be free, as it’s been over the last several months. Actually, we know it’s not cost effective. It can’t be free forever. So, everyone’s used to using Google for free and getting answers rapidly. We already know that won’t always be possible with ChatGPT. So, until there’s a ChatGPT that’s as fast, free and widely available as Google, I don’t think it’s going to pose a huge threat to Google.
Now, we also don’t know what Google’s doing on its end. We know Google owns DeepMind, which is part of the Alphabet family, and they do something very similar. We know Google has LaMDA. We know Google’s talking about launching something called Sparrow. Google’s not sitting there twiddling its thumbs. It’s paying attention and it has a good track record of launching products that people like.
Who knows how they’ll monetize it, but maybe in 6 months we’ll all be completely mesmerized by Google’s AI tool and not talking about open AI anymore. We just don’t know.
Carlos Meza: Those are very interesting comments. You have a lot of credibility in the market doing this for so many years, and I think people listen to what you’re saying.
Going back to E-E-A-T and effort, can you give us a good example of something that has really good effort put into it? Writing a blog with AI, your effort is just clicking a button and prompting the AI, but what’s a good example of effort going into the content?
Effort — The Other E in E-E-A-T
Lily Ray: I talked a lot about this last year. Again, I was calling it expertise, but I was really talking about experience in many cases. I shared a lot of examples in which the expert is the same person creating the content or at least putting their name behind it. We don’t know if they’re actually writing it. Bob Vila comes to mind, as does a site called Smith’s Pest Control.
So, I talked about these businesses in which the person who leads the business, the person who’s the face of the business — whether they’re a pest control expert going out there and killing roaches all day or a home contractor who’s doing renovations — is the same person whose name is on the content. When you read the content, there are pictures of them on the job or even the roach they killed.
First of all, they’re taking their own photos. A lot of it is: hey, a lot of contractors will charge you for this and this thing, but you can consider doing these things at home. We’ve talked a lot about these DIY solutions. It’s actual helpful advice that’s not trying to sell something but is legitimate experience and expertise.
You know the difference when you’re creating this content. I’ve created content with clients. I’ve created content with my friends, where you just say the thing that you know about the thing that you do. You can’t look for that stuff in keyword research tools. It’s actual firsthand experience.
So, those sites come to mind. Diet Doctor always comes to mind. That’s a gentleman who talks about the keto diet, but if you look at the content, it’s just beyond. With the amount of effort that goes into it, nobody would question this page. He doesn’t have a lot of content, but the content he has is incredibly robust. My thought is: Wow, you win. I can’t do any more research than that. I can’t have any better photos than that. The UX is incredible. You can tell when a piece of content has a lot of effort behind it.
I was looking at the some of the Bankrate and CNET articles in the news, that were written by AI. I couldn’t tell that it was written by AI, but when you compare it to content that’s written by a human, it doesn’t really go as in-depth. It just talks about the basics but doesn’t add a lot of unique information beyond that.
Carlos Meza: Wow. That’s very clear and very helpful. So, why don’t we take a few questions from the audience now that we’re 15 minutes away from our deadline?
Q & A With Lily Ray
Carlos Meza: Alaina Bergen says: I work at a life insurance company. We currently have a couple of in-house experts we’re using as expert reviewers for content. They have a significant digital presence because they host podcasts and are quoted everywhere. We have a lot of other experts in-house, but they aren’t as visible online. Are they bad expert reviewers to use? What’s your recommendation for vetting expertise and how Google will see a personal expertise for expert reviewers?
Lily Ray: Generally, the more you can do to demonstrate that you’re doing the right things, the better. Google isn’t going to pick and choose between this expert or that expert in most cases. Most of the time, it doesn’t know who people are.
I’ve taken a deep dive lately to confirm that, because a lot of talk in the E-E-A-T conversation has surrounded how Google can’t possibly know who everybody is.
I’m working on some research now. They’re getting better at it. They’re starting to get better at understanding who people are. But, how can you inform what Google knows? Start taking steps. Even if those experts aren’t well known and don’t have a digital footprint, you have to start somewhere. You have to put their name on the site somewhere, assuming they agree to it.
I think we’re getting to a place where it’s in everybody’s best interest to try to have somewhat of a personal brand, whether they’re an expert or they’re trying to be perceived as an expert online. Of course, if you don’t want to participate in SEO, you don’t have to, but if you’re in the business of getting traffic to a company, it looks good.
Yesterday, Google posted pictures of the search advocates and their bios on google.com. So, it looks good for brands to show who works there and for searchers to be able to validate that by searching for the person’s name.
I think, as much as you can, make it a practice within your business to say, “Hey. Even though you’re an expert without a huge digital footprint, can we just include your name? Think about starting a bio page.” You don’t have to do it right now, but try to incorporate that into your strategy over the long term.
Carlos Meza: I hope that answered the question for Alaina. This is a very technical question. What type of changes do SEOs need to make in their Schema markup or structured data, if any, to accommodate the new E?
Lily Ray: I think it’s the same general recommendations that we’ve had around E-E-A-T for a while. Anything you can do to make your organization, authors and experts — I love personal schema for E-E-A-T — build it out as much as possible.
Talk about their credentials. There are so many attributes in a person’s schema that people don’t use, including “knows about,” so you can say knows about this, this and this. Went to school, has credentials — use all that stuff.
Go to schema.org. Look at all the attributes. See which ones you can verify with the person and put that in your person’s schema. It’s the same with the organization. If the organization has been around for X number of years, or they’re owned by this parent company, it shows experience and expertise. There are also attributes for content, such as “reviewed by,” so look into doing that for anything that’s been reviewed.
Generally speaking, I’d suggest starting schema.org, look at the attributes that make sense for that schema property and fill out as many as you can.
Carlos Meza: Andy Gucci asks: how are you using the perplexity.ai tool you mentioned?
Lily Ray: I’m not using it yet. I was using it this week to see whether it is true that it can cite sources properly and how it’s doing that, because ChatGPT has not proven to be the most effective at citing sources. Perplexity claims to be better at doing that.
This is a better example of what it might look like to get the AI content and also have the sources at the bottom. Hopefully, whatever Google and Bing decide to do, citations and sources would be nice. Perplexity is just another visualization of what that might look like.
Carlos Meza: It’s sort of a citation tool, if that makes sense? Will it supply the sources, and you can cite them in your content?
Lily Ray: Yes. It’ll tell you where it got the information from and you can actually click on it and go to the site. So, it’s a blend between featured snippets and ChatGPT.
Carlos Meza: Are there any verticals in which you see the new E being harder to demonstrate than others or where could it be an advantage?
Lily Ray: We’ve worked with numerous product review sites in the last few years, and they’ve said it’s not always possible for us to do what Wirecutter does. It’s not always possible for us to test all the baby strollers before we write about them. It’s not something our company can do.
Yeah, that’s the point. That’s why Wirecutter ranks for everything. Their entire business model is to do that.
It’s not possible for a travel blogger to go everywhere, so maybe you don’t write about the places you haven’t been to. It’s hard.
I think Google is in the business of reducing content that’s created just for the sake of creating content. In a sense, write about what you know. Don’t write about things you don’t know.
As far as where it helps you, there are product reviews or your experiences going through any type of situation. I recently played around with writing content. I’m a digital nomad, so I was writing a blog about some things I’ve learned as a digital nomad. I didn’t have to use any tools for it. I just wrote what I knew. So, I encourage people to write based mostly on what they actually have to say and less on what the tools are telling you to write.
Carlos Meza: That’s a good point, because now all these tools tell you that you need to write about certain topics, but then you have no clue what you’re writing about. I like what you said about how AI cannot have the experience. It’s just a machine, so it can’t have experience.
This question is from me. Do you think AI can write better than some low-quality writers? A lot of people think they can write, but is AI writing better than some humans — leaving experience and expertise aside — in terms of readability, etc.?
Lily Ray: Of course it can. That’s one nice benefit. It’s well written. Asking it to write poetry and songs, that’s so much fun.
It’s a good writer. That’s what’s tricky about it. For some people, it’s actually better than the content they were creating before. But, when a hundred people post content that’s basically the same content, it loses its voice.
Carlos Meza: True. I think that will definitely push out some low quality writing or maybe it will help low-quality writers improve by helping them rephrase and get more succinct. I’m very interested to see what’s going to happen.
Is there a checklist or a way to optimize for E-E-A-T? What are your bullet points for someone who wants to optimize for it?
Lily Ray: There isn’t really a checklist. However, Google has an article that’s called What Site Owners Should Know About Google’s August 2019 Core Update. Google’s been using this article for 3 years, going on 4 years. There’s a list of questions you should ask yourself when you’re evaluating E-E-A-T. Look at this.
Then, there are the search quality guidelines and the product reviews updates. As much as that’s focused on product reviews, they have a similar article for that, that also has questions. You can basically copy those into a spreadsheet and use that as your checklist.
Carlos Meza: That’s a great idea. So, Anna is asking: What’s your advice be for a personal finance website where a team of writers reviews products, such as loans or credit cards, which nobody could realistically have experienced as a user of each product?
Lily Ray: This is where UGC becomes valuable, particularly if a lot of people are saying the same thing. If you’re reviewing it, you can indicate that while you were reviewing consumer feedback about the credit card, you noticed that 64% of the reviews mentioned a particular component. Essentially, the best you can do if you haven’t actually used the product or service is leverage what other people have said about it. But, try to reach out directly to people who have something to say and use that if you can.
Carlos Meza: So, it goes back to the journalistic approach you mentioned before. If you don’t know, talk to someone who does and bring that information.
There are two more AI questions. Stuart says: I’m struggling not to parallel the current AI content plagiarism debacle with the spun article activity prior to 2012, before the Penguin and Panda algorithm rollouts — agree or disagree?
Lily Ray: I love that question. I think that’s why I’ve been wary of the whole thing since it became the hot topic in SEO. I got badly burned by the Penguin update in 2012. Granted, that was different. That was link building. The Panda and Penguin updates really changed how we approached content and SEO.
As somebody who never wants to find themselves in that situation again, I’ve been reluctant to believe it’ll solve the problem of how to scale content very cheaply. I think Google’s going to be one step ahead of us. So, I tend to agree.
What’s different now is that AI-generated content often does sound good, so there will likely be a period of time during which we can trick users into thinking it’s human-generated content. However, I imagine Google will catch up to it soon.
Carlos Meza: I don’t see a world where Google doesn’t come out with something to check on AI or low-effort content.
Daniel wants to know if AI is useful for writing FAQs. I have been playing around with AI-generated FAQs and just spot-checking the answers. For many of these questions, the answers are pretty simple and direct.
Lily Ray: That’s a good question. I would be careful because there are several examples of sites that have done exactly this in the last several months that have seen all their traffic disappear. Granted, they were doing it in an egregious way, so they were scraping people-also-ask questions, using something worse than ChatGPT to create the answers and then creating entire sites based on that.
That’s the worst possible example. It’s different from just adding FAQs to the end of a page. But, that tells me that Google already figured out how to find out that people are doing that. They’re trying to figure this out.
So, just be careful, because the next person who has the same approach of just adding AI-generated FAQs may have the same FAQs as you. Some things may be a little different, but it’s largely the same thing.
Carlos Meza: I would add some perspective from my point of view. I think if you produced content and then asked a tool to summarize it or create an FAQ based on that content, it may be a good use case. Essentially, take the content I produce, feed it into a tool. Can you produce usable bullet points and FAQs out of this?
Lily Ray: I love that. That’s a great idea.
Carlos Meza: I use it to summarize some of my own sentences. It helps me write succinctly. It’s still the same essence, but with a different style. Plus, it checks for grammar. We see a way where you can feed it your own content and data to create things that are useful, but you’re putting in the legwork for the tool. The tool is just something to augment or enhance your process.
We’re now approaching the end. I always walk away with some great takeaways from listening to Lily talk about E-E-A-T. I’m sure our audience does, too. It’s an interesting world ahead of us. We’ll keep bringing great guests to provide value to all of you and help you navigate these interesting SEO and content waters.