Earning the Recommendation: How AI is Reshaping SEO Strategy

Announcer: Welcome to the Ignite Podcast, the only healthcare marketing podcast that digs into the digital strategies and tactics that help you accelerate growth. Each week, Cardinal’s experts explore innovative ways to build your digital presence and attract more patients. Buckle up for another episode of Ignite.

Lauren: Morning, everybody. We have an awesome episode of Ignite Healthcare Marketing Podcast with our VP of Earned and Owned Media, Rob Sauter. Rob, good to have you back. I think it’s been a few months since we recorded one of these together.

Rob: Yes. Sharing the love with the rest of the team.

Lauren: This one is a good one, a hot topic, AI, and what is happening in the world of SEO. We did an early, early episode on what was changing and what is obviously really important. I think now, as everyone’s learning, it’s becoming a lot more what to do with that information. How do I change my strategies to adapt to the landscape?

Today, we want to bring you guys some real tactical insights and opportunities, how we are changing what are maybe traditional SEO programs to look more like AIO plus SEO plus UX, and all the things that really go into still earning the business.

Rob, teeing us up here. I know we’ve talked a lot about ChatGPT, the obvious ones. We’ve got the new Google search generative experience that Google is experimenting with. There’s just so many popping up every day. What we’re noticing is that traditional rankings, traditional, you may go on your SEMrush, and your traffic looks like it’s going down. That’s because all this zero-click experience is really evolving.

What do we do with that information? How do we track that information? How important is that to still winning business? Rob, I think teeing us up here, I would like to start with your thoughts on how the landscape has shifted. I know I called out a few things, but really, I think the last three months have been verbatically different than they were last year.

Rob: Yes, even since we did a webinar, I think in April, we talked a lot about the tips and tricks. Honestly, all the tips and tricks from that webinar certainly hold true. A lot has changed since April, especially in the healthcare space.

For example, we’re obviously seeing not only a continued adoption of AI, but also a lot of changes in Google. If you’ve been paying attention to your SEO in the healthcare space in Google since around March, the SERP features have absolutely ramped up like crazy. We’ve seen this across all of our clients. Effectively, every client that was competing outside of their brand terms in SEO saw their SERP features triple, quadruple. It’s only just started here now in July, following the last Google algorithm update flatline.

We’ll see for now. It’s seemingly found its balance. Obviously, the SERP features include the AI elements, but also other aspects like maps and even the people also ask features. Little things, there’s actually some new social media elements now within Google that’s being surfaced. little things that both incorporate AI, as we tend to know it, like the actual AI tools, but also ways Google is using AI behind the scenes to better crawl and understand information.

A really fun example I saw, this is tested and not yet actually stated by Google. There have been some really interesting tests in the SEO space recently that show actually creating and updating your Google business profile images are actually helping ranking. In fact, some people tested adding words to the images. Saying, best dentist in Atlanta, right in the graphic. That actually–

Lauren: [crosstalk] instead of just a picture.

Rob: The theory is because of some of the elevations and some of those images is that Google is using AI to look at images and get contact from images and then use that as a ranking factor. Again, Google hasn’t owned it yet, but these are examples of how very likely Google is using new AI tools to help impact outside of just the tools that you and I use.

Lauren: You think about all those traditional tips and tricks, we’ve talked about technical crawlability, indexability, these crawlers, whether they’re now the AI crawlers or the traditional ones, still need to be able to access the information. When they get there, the information, the content needs to be helpful, unique, specific. That stuff all holds true.

What’s really changed, I think I heard you make a statement that the AI engines are now recommendation engines. I thought that was really interesting. We used to talk about backlinks. Backlinks are like likes. It’s like a recommendation. Now that concept of a recommendation is really holding true beyond just, can I put the best content out and win my way to the top?

Rob: Yes. Traditional SEO was built on websites, and that’s how it all started. The web, et cetera. It was really focused on brands. You build websites for, generally speaking, products or information. You don’t build it for an answer to a single question, at least not anymore. Now with AI, yes, you’re not really necessarily fighting for that rank, that position. You’re really fighting for that recommendation. That comes from a lot more different sources. We’ve seen it in the AI space a lot.

One statistic I shared with the team this week that I really like is that when you look at the click-through rate from Google’s search position and you look at how the human click-through rate on Google is behaving, that first spot is nearly 30% click-through rate. It plummets down to 15 and then 10, obviously trailed off from there. When you look at the ChatGPT citation and what’s in what position that ChatGPT likes to cite, whether it’s cited or not, it drops all the way down to just below 10% for number one, and then around seven, and then six, and five.

There’s a much more even distribution. By the time you hit rank seven, the actual ChatGPT citation is higher than the human click-through rate. Again, as the human click-through rate trails off to eventually zero, the ChatGPT is much more level. It shows that, yes, being first probably means you have valuable content and you still deserve to be cited. There’s so much other great content out there.

I think that’s a great insight because those of us who work at SEO know that people like to fight over that first spot. It’s less important now. Now it’s about having good, solid content in the top spots. We’re even learning things like the way that Quad came out, and it showed a lot of its back end and how it’s crawling and what it’s doing. It exposed some of its process to us, the professional marketers of the world. One of the things that people deduced there was that it’s using more of a topical ranking platform.

Instead of just looking at a page and ranking it, it’ll look at multiple pages on your site together, all around certain topics, and utilize the weighted score with all of those pages together and position you in rank, which is, again, different from how things had happened before. It helps emphasize some of the things we knew were important, like topical authority, but it gives you a chance to differentiate and not have to just focus on these one or two big pages and have it be a win or lose scenario.

There’s a lot of really good information, lots of really interesting insights. Some of the things I’m most excited about since our last call is I think when I landed the webinar, one of the things I said was, “We really don’t have a measurement platform that we can–” [crosstalk] If there’s some little things here and little things there, but that’s changed significantly since then. Now it’s a bit of a software land race, an arms race on the software front to create platforms. We over here at Cardinal now started tracking prompts and tracking different elements of AI across.

At this point, we’re around seven different engines, tracking your presence, tracking your perception, and even tracking your performance and indexability within the AI platform. It’s not quite obviously as robust as what we can track in SEO, which has had decades of time to fill down platforms. The other thing about SEO that made it easier is for the last how many years, Google’s truly been the dominant space search.

We’re in an environment now, we’re sure, ChatGPT, Google, they’re a bit ahead, but they’re reasonably speaking, six, seven, eight different platforms people use. Imagine if there were seven versions of Google that held a significant enough margin share to care, it would certainly make measurement more tricky. I think that’s where we’re at now, but the insights have been amazing. We’ve made some really cool decisions for clients and drawn some really cool insights from being able to truly track prompts over time, prove time series data to analyze.

Lauren: I think what’s really interesting, you started out by saying the content side of things, what you do on your site, obviously, you can control. That was so much of what earned you top position in SEO. Now that perception component is made up of so many other off-site mentions of your brand. What are you seeing are the big opportunities for organizations to keep an eye on things, look at your perception, try to manage that better?

Rob: Yes, that’s a really good point. We’ve made a few really cool insights lately with clients on this topic, exactly. We’re seeing a lot of importance come back to PR. I think we’re moving a little bit away from the link buying and just frantic link building of previous SEO and looking at more of that value-based, structured-based PR, especially in local businesses. Getting lots of links might help, but getting those local links, showing that you’re connected in that community, it’s making a big difference.

One of my favorite examples that we’ve seen recently becoming very important is, and this is obviously healthcare, where we have professionals, providers working with humans. The metaphor I used is, have you ever checked the last door before going to a restaurant? Not a place you go to check out a restaurant. You’ve got–

Lauren: It’s like working there, yes.

Rob: Yes, you never wonder what it’s like to be a waitress or how they treat their waitresses or chefs. However, now with AI, especially in healthcare, when you search and you ask questions about a particular brand, a particular location, a particular provider, hospital system, et cetera. It will pull data from Reddit. It will pull data from the last door. It will pull data from PR sources. It will pull data from, frankly, everywhere it can get its AI fingers in the Internet, and it will surface that information. It will still tell you upfront what’s most important, things like great care, great reviews for patients.

Now it will also surface direct statistics and information around corporate cultures, it’s a mess, or providers are unhappy, or really high turnover rates. Those are the types of things that can make, especially if you’re looking at health care for a child. As a parent, and Lauren, also as a parent, you hesitate to think like, “geez, like this place is turning over providers way above that industry average, and do I want my child to might need a specific specialty care?”

You have to go through the process of meeting a new stranger or me trusting with my child with a stranger often, or even worse, maybe not even the turnover, but an unhappy person who’s overworked and stressed out and writing about it online. Is that somebody I want to be caring for my, frankly, most important thing in the universe? I don’t know. Again, that’s information you probably wouldn’t have seen by going to websites, et cetera. Again, unless you’re–

Lauren: I would have never planned it into the result before.

Rob: It’s impacting. It’s impacting people, and same with that PR element. It’s double surface, scary PR stories, et cetera. There’s something you can do. Marketing can combat that a bit, but at the end of the day, it’s helping service real business opportunities, and being a part of the marketing team is being part of the business team and being able to talk to some of the leadership that our clients will say, Here’s an issue that you’ve identified–

Lauren: [crosstalk] are going to be more important now than ever.

[background music]

Alex: Healthcare marketers. What’s up? It’s Alex from the future. Guess what? Scaling up the Healthcare Performance Marketing Summit is back. Scaling up is focused entirely, entirely on driving patient acquisition to your group. You’re talking the largest provider group, health system leaders, everything it takes to drive a patient to your practice or health system from media, BI, analytics, performance, creative SEO, AI, because we’ve got to have that acronym in there.

October 28th and 29th, thousands of healthcare marketers are going to be showing up to this. It’s virtual and it is free. That’s the best part. Last two years, we were charging for it this year. I want every healthcare market to come. We need to connect more patients with care. We all do. We all need to do it together. I’ll see you there, scaling up.

[music]

Lauren: You talked about this concept. In some cases, the AI result is maybe additive or complementary to the traditional SEO result, and that may be happening in certain types of searches, and in others, it’s really cannibalizing it. Can you talk a little bit about the types of searches, the search intent funnel, where are you seeing more cannibalization, where are you seeing it be more complementary?

Rob: I keep telling my healthcare, all my clients in the state healthcare clients, but at least we know at the moment, AI is not curing anybody. You’re not logging into ChatGPT and getting rid of your allergies or fixing your broken arms. Generally speaking, the demand hasn’t changed. How people find care has.

I think what we’ve seen, a lot of the declines have come from the, think of like the WebMD type search, the what is, the what are, the when should I, when should I go to the doctor when my child has a fever, what temperatures, a scary woman or a rash, something I can treat topically, or should I go see a dermatologist more urgently? These are things that AI is certainly–

Lauren: A bit funnel-type of searches.

Rob: Yes, they’re certainly helping with. Now I do see, and we have seen in a lot of cases when volume and traffic have come down, especially in those, what is, what are those very early discovery, we have seen traffic maintain, maybe slip just a tiny bit. To be honest, sometimes it’s just shifting to AI. If you’re not tracking your GPT, referral traffic as part of your organic, you’re just not getting the whole measure of the whole. It’s still a smaller percent, generally, but people are still taking those final steps. They’re still searching your brand. They’re still going to your location pages. They’re still going to your Google business profiles, and they’re still experiencing and reading what you have to say.

I think that’s where, and we’ve been encouraging a lot. That’s where you really want to drive that why us in that trust factor home. Build out those case studies, build out those testimonial videos, build out the unique value proposition. Set it on the webinar. Say it every day. I’ll say it here again, being close and having doctors, not enough anymore. Healthcare people are a little more involved, especially, there’s a lot of rhetoric and conversation going around, and healthcare and insurance and all these people care. They want to know, other than just being the most convenient option. What is it about you that differentiates?

Again, think like a hospital system. They all have that answer. Every hospital in America, in my opinion, most of them at least know how they compare to the two or three other hospitals nearby. Are they the most clinical? Are they the best experience? Are they the most compassionate? Are they priced better? Sometimes it’s simple as that, but that’s really important right now, both for AI to be able to glean that and have it backed up by things like reviews and provide that as a recommendation, but also for the people who find out about you and AI and want to figure out a little bit more about you.

I’ve heard a few people talk about how websites are dead and applications that things are taking– it’s definitely not anywhere near there yet. Maybe in some cases, like maybe news and things are a little slowly changed formats, but with the localized healthcare, people [crosstalk]

Lauren: They’re going to end you ultimately on someone’s site. I think that the narrowing of options, the understanding of best perception, that’s its role there. Then you’re going to ultimately, like you said, you’re going to get that bottom funnel click through based on the content that they’ve consumed to help them get to that point.

Rob: You’re right, how often AI is wrong? I have lots of clients where we show them the perception, and it will say, “This brand still operates with this model.” The client will say, “Actually, that’s not true. We’ve updated and changed that model. About a year ago.” You find out that’s what people are being told in the AI space. It might even be an opinion. AI got that information somewhere. It’s either outdated content on the internet, outdated reviews, et cetera.

There’s opportunities there. AI only knows what it can go find. Especially the more niche you are, the more you actually have an opportunity to control that narrative. Think like the Nikes and Adidas’s of the world probably have lost some control, it’s so big and so well talked about, but in niche healthcare, you absolutely– What I see in AI is sometimes 60% of the opinion of AI is driven directly from the website. These websites, again, aren’t in many cases doing a good enough job selling themselves, both to the consumer and to the AI. It will help make that recognition.

Lauren: No, if taking this from what’s going on to tactical, if organizations are experiencing that, what should they be doing? I know you talked about PR, there’s onsite content opportunities. If you feel like AI’s got it wrong, or you feel like you should be seen as the best in these results, and you are not showing up for those things, your clinical care is excellent. Why can’t Google understand that? What do you need to be doing? Give us like three to five strategies.

Rob: the first is, and this is something I learned since the last, since the webinar, get in on the platform, either find a measurement tool, or just open up a whole bunch of tabs with different platforms and find out there are a few different options, a few different prompts, who is showing out, but who isn’t. For example, we found a few times that– I’ll just keep the dentist example right now. We have clients that dominated dentists near me, or dentists in such and such on SEO. Then we went into AI, and we found, oh, don’t even show up. “Why is that?” It turned out certain AI platforms read the term dentist as a human, and not as a dentist’s office or a dentist’s brand.

What we found was, okay, well, we need to start writing some content about our dentists and the dentists in this community, and what they’ve been able to achieve and help. We can appear at all for these terms, or we’re appearing 80% of the time, or better in the top 3%, three positions at SEO on AI, we’re not showing up at all. We’re showing up 3% of the time. It was crazy. You really have to go understand, do you want to rank for AI? What is the logic? Take a look at it, understand how it gathers information, and what it actually is displaying. That can help you understand is there just gaps in my program altogether? I think that’s one of the big ones.

From there, it’s a combination of, I think that the big opportunity now, especially in healthcare, is creating a combination of long-form and strong content. It has been shown that really robust first party is key. Say something interesting, say something unique, have a quote from a provider, have a byline. It really brings something specialized to the table and don’t hesitate to do something a bit more robust and a bit longer. It’s shown that longer form content, well-linked, well-articulated, and first-party content does tend to be better links and better cited in AI. That’s another interesting tip.

I also think having a variety of content in topic is also really important. We’ve seen that blogging has been obviously a huge element, and that’s great. Blogging is not bad about saying you get away from blogs, especially if you’re doing it well and strategically. Bringing things like infographics, bringing things like videos, YouTube’s still one of the top search, maybe the top search engine in the United States right now.

Maybe not necessarily for healthcare, but generally creating that variety of content that is now, you don’t need to just have blogs anymore. AI can understand and see different forms, especially images of content. That variety also really helps because AI likes to provide variety in its results, unless you search something very specific.

Lauren: Let’s say you’re using a tool, and we’ll talk to it, we’ll end what the tools are to give everybody some direction there. You’re using a tool and you’re finding out that the perception of your business, although you may not feel that this is true, is related to billing or wait times or these kinds of things. Online scheduling, availability, cancellation policies, like things that play into perception and people like to go online and talk about, but maybe have nothing to do with the quality of care, but they’re still affecting your ability to return results.

We’re talking things like, get out there, make a process or policy on, and publish how you do your billing. Things like creating a one-sheeter or a YouTube video showing people, this is how to tackle these billing challenges. This is who to call, get out there, understand, and I think this will tie really nicely into walk us through some of the tools, understand the prompts people are searching for, understand what it’s returning, why it’s returning you, why it’s not returning you.

What the top perceptions are across the board, and then tackle those head-on in FAQs, in video, publish them to YouTube, put them on your site. Those things, although your end user may use them periodically, are also really going to help the AI engines understand that you do care about this thing, you recognize that it’s an area of opportunity, and you’re putting effort into it.

Rob: Yes, it’s also great data. I’ve had a lot of the CMOs I work with just say, “This is great.” We knew this was sort of a problem, but this data, this visualization, this insight is something I can take back to the business, help them understand how I can help as a marketer, but also help them understand from a business perspective that we do need to put time and resources against, either fixing, filling, if it’s broken, or making the employee experience better, or at the very least, trying to capture great employee experiences so we can put those on the internet.

I think the only thing I’d add, an obvious trick, and it’s probably published out there, but it’s worth citing again, is when you’re searching on these AI engines, you can ask them directly, “Why did you show these people? Why did you show these brands? Why and how did you pick?” Put that at a table for me and weight it by importance, and it will actually show you its logic table and why it pulled it. You can even ask questions, “Why did you not show my brand?” It will usually give you an explanation. You’re not as localized in this environment. I’ve thought these are more specialized answers, et cetera.

Sometimes, what I really like to do with my clients is try different versions. Let’s say you’re a physical therapy brand, and this is an example I did the other day, is search best general physical therapy, best sports physical therapy in my area, best rehab physical therapy, and you’ll get different results. I think that’s an opportunity to uncover, are we ranking the way we thought we would in these outcomes? Is this an opportunity? Should we lean in?

For example, if you’re ranked number one in one of those categories, but four or five in the others, maybe that’s part of our value prop. Maybe that’s something we should lean into more, as we already are without even maybe on purpose doing it. That’s how the industry, that’s how the human, that’s how our reviewers see us as an expert in a certain area. Sometimes it is better, especially in the world of SEO and AI, to lean in to something advantageous and not try to be the whole thing.

Lauren: When it comes to really discovering these insights, tell us the tools, tell us what you like, what you’ve tried. I know, like you said, it’s an arms race right now when it comes to all these tools coming out and producing their AI ranking features.

Rob: There’s cool tools out there. I think a lot of the big traditional SEO players are working to have their tool participate. The Ahrefs has got some new interesting features that help understand AI and value. They did an interesting podcast recently talking about actually how they actually see the resurgence of brand measurement in this space. As attribution gets trickier, what are some ways to prove that your SEO, your organic, your PR is working, is your brand search for your brand increasing?

That was interesting, and again, their platform’s evolving. You’re seeing some in places like AccuRanker that can pull things in. SEMrush has a really nice enterprise AI platform that does good measurement. Benson, if I’m being totally transparent with you. Yext has something they’re working on called Scout. I don’t know if it’s officially released yet. If this is an agency been able to see on the backend, it’s super cool. I would call it one of the more like next-generation AI tools.

The graphic in the user interface it’s super neat. It’s like a globe. It’s almost like a Google Maps, and it puts off almost looks like a video game. It has territories mapped around your locations with how you rank. It’s this like cool view. That’s going to be interesting tool. The one we like right now that is a really exciting brand to follow. They do a lot of great leadership as well, but we’ve been utilizing the Profound platform, paying for the buck. It’s so good.

In my opinion, it’s a really well-priced platform. It really can let you get in and get started. It’s completely customizable, which is nice. Don’t want to turn this into a commercial for Profound, but really nice tool. Let’s you track your competitors. I joked with you, Lauren, and with our CEO, Alex, once that, man, we could almost start a whole new business unit on competitive intelligence based on some of the elements of this platform which is really exciting.

You can customize your computers, customize your prints, take a keyword set to measure. It tracks across, again, at the moment, I think eight now different AI platforms from ChatGPT, Perplexity, Google, three different versions of Google. The AIMO, the AI Go for Videos, and Gemini, Copenhagen, [unintelligible 00:24:23]. It’s exciting, and it’s a cool tool. That’s the one we’ve been setting up a lot of our clients on and running some really exciting pilots.

Lauren: It’s really easy to stand up. Rob, you gave me a sheet and told me I was building one for a sales prospect who gave me a sheet. It took me 11 minutes to set it up. Then you can take your window once you get up and running, so.

Rob: Yes. On the back end, just even connecting on our side, our dashboards are now connecting directly into referral traffic, which is great. Alex Kemp on our team has helped me set up some dashboards for our clients that can track how many sessions, how many leads, are you getting referred in through AI now. It’s still not as high as you think. I believe this is my own thought, my own insight, but I believe people are still using AI and then Googling and clicking–

Lauren: Clearly, it’s typical navigation and experience.

Rob: Exactly. I’ve talked about this a lot, but AI is still not there yet on user experience. Even though it’s like things, it doesn’t make it super easy always to click. It doesn’t say here are the top five brands, and here’s a big button to click to go to their own page.

Lauren: You got this tiny little gray thing in the bottom that you’re like, “What is that?”

Rob: Yes, so it doesn’t make it obvious. Maybe that’s because it doesn’t want you to leave. Maybe it wants you to stay on AI and work there forever. I think people are still clicking away from their AI tab and searching more traditional ways. That will eventually change, I think. Being able to set that measurement up is nice. There’s hundreds of leads for some of our larger clients coming in through AI that we weren’t taking credit for, in the digital media standpoint.

I’m sure our digital partners are excited to have a couple of hundred leads added to their friend sheet. Obviously, the business is still capturing that growth, but it doesn’t hurt to be able to show marketing is doing good work through their channels.

Lauren: I think that gap in attribution is a little scary for marketers right now. I know it’s coming. My traffic is down, but my leads are not. I just can’t really point to where they’re coming from. Then you start to get questions, and am I making the right investments? Totally get it. Summing us up here for marketers, actions to take right now, get yourself on one of these tools, like data is power, knowledge is power. You need to know what is happening.

I know you talked about the fundamentals, your core web vitals, your crawlability, indexability. That has not changed. Those things are table stakes. Really thinking about your content and using the data that you get from those AI tools around the prompting trends to inform your content strategies. Then you’ve got to work on any perception gaps, take those to your ops team, and start to bring them that data and talk about as an organization how you’re going to tackle those things. Anything I’m missing there, Rob? Summing us up?

Rob: No, it’s still– Alex likes to tease us, too. There’s a lot of the same programs still going on. Don’t abandon your website, don’t abandon content. I think it’s really coming down to, with AI, it’s not as easy. You still can, but it’s not as easy to just brute force your way past your competition by just pumping out more content and pumping out more of a presence in the Google ecosystem.

Now you really have to have something interesting to say. If you haven’t thought about it before, start thinking about it now. Not only just publishing it, you really got to believe it because it’s going to have to be backed up in reviews. It’s going to have to be backed up in PR and partnership, votes of confidence from referrers, and things like that. I think that’s the best opportunity. Really find out what makes you you, the why your patients love you and choose you, and recommend you to their friends. Because that’s the story you want to start telling more aggressively, more creatively in this SEO space. You can win that recommendation.

Then otherwise, do your best to keep up with the exciting changes we’re seeing in this space. We have AI browsers that are being talked about. We’ve got CloudFlare doing funny monetization things in the background, like things are going to continue to change. I think this is a bold statement. We’re not going to be that far off, really. I don’t think from some monetization play in AI, it just makes sense. Google’s making billions like every afternoon on search. There’s just no way these funded back AI organizations aren’t thinking aggressively on how they can monetize this type of thing.

Lauren: Yes, love it. Rob, thank you so much for the insights. This is a topic I have no doubt we will probably be doing a monthly episode on because it is evolving so quickly. Thanks for bringing us up to speed on where things are at and sharing some good tips. Everybody, thank you for listening to another episode of Ignite. Like, share, comment, post it on LinkedIn, tell your friends. If anybody wants to talk about AIO, AI optimization, what’s going on in SEO, feel free to reach out. Rob and I love to chat with you guys about it. Rob, have a good weekend. Yes, thank you.

Announcer: Thanks for listening to this episode of Ignite. Interested in keeping up with the latest trends in healthcare marketing? Subscribe to our podcast and leave a rating and review. For more healthcare marketing tips, visit our blog at cardinaldigitalmarketing.com.