AI Visibility Starts With SEO Basics

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.

Ashley Pedrochenko: Hey, everybody. Welcome to the Ignite: Healthcare Marketing Podcast. My name’s Ashley Pedrochenko. I’m the VP of brand marketing here at Cardinal. Joining me today on this episode is Caroline Adamec. She’s the Senior SEO Data Manager at BetterHelp. Welcome to Ignite, Caroline.

Caroline Adamec: Hi. Thank you so much for having me. I’m super stoked on this.

Ashley: Thanks. Today, we’re going to dig into how AI search is evolving and what that means for how do you to stay visible along the patient journey. Before we dig into that, Caroline, why don’t you tell everybody a little bit about your background and the work that you do at BetterHelp?

Caroline: I have been with BetterHelp for about four years on the SEO team. I started in an entry-level, very execution-focused role. In the last year or so, I’ve taken on more responsibilities leading our strategy for both AI search and traditional SEO search.

Ashley: We’re at about a year mark. It feels like AEO and GEO and all that has just been at the top of everybody’s mind for the past year. Now, it feels like we actually have enough data, enough experience to understand how things are evolving and what that means for patient acquisition. Looking back the last year, what SEO fundamentals are you finding are the most predictive of AI visibility across the Google AI overviews and LLMs?

Caroline: This is a great question. I think there’s a lot of overlap between good traditional SEO and AI search, specifically because AI search acts more like a human than an algorithm. It’s really looking for those same trust signals that humans are looking for when they’re trying to find genuinely valuable content. I would say number one is PR, third-party backlinks, and building out your brand through third-party website collaborators. Having a really strong profile on the internet from multiple sources is going to signal to AI that you are a trustworthy brand and that you should be cited when people are asking queries related to your brand. That’s step number one.

Obviously, we’ve seen that in traditional search with backlinks. Backlinks have been around forever, but I think the difference is that domain authority and trusted sources matters way more than it did in traditional search with AI search. Really making sure that you have high-quality sources that are linking to your site and you’re not just putting site URLs on backlink firms is really important for showing up in AI search. I would also say genuinely helpful content. Obviously, there’s a lot of abstract terminology about this. [chuckles]

Genuinely helpful content is just content that answers the user’s question, provides them with all possible relevant information related to their query, and approaches it from multiple user search intent points. That goes into meeting the user where they’re at, making sure that you understand what they’re looking for, and that you’re providing that information in a clear, transparent, and easy-to-understand format and way on the page.

Ashley: I think a lot of these being these RSEO principles that have been true for a long time, but people could get by with bare minimum effort, like really low-quality backlinks. Maybe it helped a little bit, and it was enough, but you really need those trusted sources to be endorsing your brand. It’s almost like the resurgence of PR in some way. It’s like you have to control this narrative that you’re putting out there and really be trusted. Is there certain advice or tips that you can share with people on how are they bringing those stories to life and getting that momentum back?

Caroline: That is a great question. I would say for smaller brands specifically, one of the things I’ve been seeing is user-generated content is really important. If you’re a larger brand, you probably have a partnership or PR team that can work with your SEO team to build out those stories with third-party sources. You’re going after Forbes, Wall Street Journal, you’re trying to get that high visibility opportunities for your brand. I would say if you’re a smaller brand, UGC has the same effect, making sure that you’re on Reddit and posting in threads where people are trying to find your product or complaining about problems that your product solves.

I think that’s an easy, low-hanging fruit way to get your product out there and increase your brand awareness with this new AI search strategy.

Ashley: How do you think about local versus national, too? Is that something that you guys are exploring in terms of there’s PR from those really national publications, but what about local?

Caroline: Yes, absolutely. I think with AI search, long-tail queries have been rising in popularity because people want personalized information. They don’t just want a therapist near them; they want a therapist in their exact location who provides this specific type of therapy. They’re looking for that extra layer of personalization. My team specifically has been really focused on this, mainly because one of our core business outcomes for the year is increasing brand awareness about the fact that we’re accepting insurance in specific states.

We started our insurance rollout, I guess, late last year, and we’ve been creating personalized content to target people in specific cities within the states that we’re now announcing insurance acceptance in. That has been top of mind. One of the ways that we’ve been driving local traffic is through schema markups. That has been really helpful in just signaling to the Google algorithm and LLMs that this content is intended for people in a certain location. You should show this when they search with that keyword, specifying their location. I think that’s how we’ve been thinking about it.

I know six months ago, something that was really popping off in SEO, AEO was programmatic. A lot of people were using programmatic SEO to target specific locations with content and filling in the specific location for that particular content. That, I think, also signals a return to targeting more localities with content.

Ashley: It goes back to what you said about helpfulness, and what does it mean to be helpful for the user? Sometimes, if you’re finding a solution, but it’s somewhere that’s a completely different locale, and it’s not at all relevant or available to you here, potentially, that information might not be as helpful at that specific point in the patient journey.

Caroline: That can maybe have a harmful effect, too. If you’re looking for specific information in your locale and Google is returning things that aren’t relevant, that’s frustrating. I would personally probably try a different search engine to try and get the information that I need. There’s also that negative effect, too, that could happen.

Ashley: I think users are much more willing and adept at just bouncing from different search platforms. If they don’t find the right answer on Google, they are going to go to ChatGPT. They’re going to go to Reddit. That might be my first spot most of the time. Everyone is much more fluid to taking the search journey across platforms. Maintaining that visibility across all of them is hard. What’s the biggest difference that you’re seeing optimizing for the LLMs compared to just classic Google search?

Caroline: Two years ago, it was very easy to hack the Google algorithm by keyword stuffing. One of our earliest techniques was going on the page and manipulating different keyword densities in different areas. Early this year, that stopped working. The algorithm got smarter. It started recognizing that we were trying to hack the system and realized this content isn’t genuinely valuable. With the Google algorithm, it’s easier to trick. It’s easier to get low-quality content ranking high if you know the exact ways to manipulate the algorithm.

AI search is too smart for that. It’s going to be able to scrawl the content on the page accurately and see what you’re doing and realize that you are not a helpful source, so it’s not going to cite you. I would say that’s the biggest difference between the Google algorithm and LLMs right now. I think we’re learning more about how LLMs cite different brands and what they value every single day. If you ask me in a month from now, that answer could be completely different.

Ashley: Yes, the landscape has really changed quickly. As Google has evolved, and a lot of it, they’ve adapted, and sometimes they haven’t adapted to users. That’s why people are frustrated because they’re searching and all the answers are the same, and they’re not feeling like it’s actually useful or helpful information to them. The behavioral search journey has evolved, and people are going to be educating themselves often way before they ever go to a practitioner. How has that changed how you think about the search journey? Where are you focused on?

Caroline: I would say our strategy when I first joined BetterHelp was very top-of-funnel informational queries. It was people searching for depression symptoms, mental health condition symptoms, just trying to educate themselves via Google, to go back to what you were saying. That’s not happening on Google anymore. That’s happening on TikTok. That’s happening on people’s social media algorithms, where they’re getting this content to pop up, and that’s where the educational stage is happening. They’re going to Google for more bottom-of-the-funnel queries.

They’re discovering, “I’m experiencing symptoms of depression because this popped up on my TikTok feed.” I think ADHD is a huge one where people are going, “Oh, I didn’t realize this was a symptom of ADHD until this person popped up on my TikTok talking about how they experience it as somebody with ADHD. That’s making me reevaluate my mental health. Now I’m looking for more information to confirm this hypothesis that I’ve created for myself.” They’re going to Google to find solutions and treatments and figure out how to cope with this self-diagnosed mental health condition.

Our strategy has evolved from targeting those informational queries to targeting bottom-of-the-funnel content. We’re focused more on branded queries now because we know that those people, when they come to our page, they’re high-intent users, and we want to capture that audience. That’s where our strategy is evolving. That also has to do with LLMs and AI Search. People might see a TikTok on mental health and then go to AI to get more information and say, personalize their queries. “I saw this TikTok. I also experienced this. What else should I be looking for to confirm this idea that I might be experiencing this condition?”

They’re getting that top-of-the-funnel educational content elsewhere and then coming to Google when they’re ready to converge or even just ready to learn more about next steps to take.

Ashley: It’s really across all of the channels. It’s that you can’t really view it in isolation anymore. Thinking about what that means for measurement and how are you evaluating a success, what are you looking at in terms of how do you track progress and visibility?

Caroline: I think AI Search, like AEO, GEO, has been top of mind for a lot of executives. It’s very much like the hot new thing. Let’s capitalize on it. There’s an opportunity to be at the forefront and really make sure that your content is up to par. That can be shortsighted because AI Search right now has very little ability to prove overall business outcomes. There’s no way to track conversions from an AI Search platform. Traditionally, in a lot of in-house SEO teams, the key metric has been conversions. That’s the KPI. How many people are signing up for a demo or signing up for the platform? That’s how we measure success.

With AI Search, the user journey becomes harder to track because maybe somebody finds out about your brand via an LLM, but that LLM didn’t cite your page. Then they go to Google to find out more about your product, your company, and offerings. You can’t connect those two within a clean data tracking source. It becomes a lot of piecing the puzzle together. I actually worked with my team to create a Looker dashboard that tracks ChatGPT UTMs because I saw these UTMs when I was using ChatGPT, and I was, “I know we can track UTMs, let’s do that.”

We have this rudimentary dashboard that gives us some insight into the user funnel, but it’s missing a key chunk, which is, well, what happens when somebody learns about the brand without a citation? What does that user funnel journey look like? It’s trying to put these pieces together and use all the data that we have access to, to demonstrate to executive stakeholders why this is worthwhile.

Ashley: Like you mentioned, there’s a lot of hype around it and how important it is, and how much do you invest into it and how much do you throw into it. I think people are trying to figure that out and justify where do I spend my marketing budget and how much it actually does really change from traditional SEO. That’s really helpful to hear. Going back to the social-driven behavior and how that’s influencing, what insights are you using from the research on that team or your team on social, and how is that applying to the content that you’re actually creating on that more middle and bottom funnel?

Caroline: Yes, our social team has taken on the role of informational brand awareness educational content. For most people now, their social is going to be the first touch that they have with the brand. We’ve been trying to use those outlets to really drive brand awareness instead of focusing on conversions for them. Obviously, I’m not on the social team, so I can’t give you a hard answer on what their main KPI is, but I think it just goes back to creating brand awareness, PR, collaborations. That is our main function of social channels is getting people to learn about the brand and discover it. Then it goes back to working in tandem with bottom of the funnel for SEO.

We’re linking our site in our social posts. We’re creating UTMs to track traffic from Instagram, Reddit, or Quora, even. That is helping us understand the user journey and where our high-intent users are.

Ashley: Looking back at the last couple of years, is there one aha moment, this is how the consumer journey– something that stood out to you that’s changed your strategy, changed how you approach things? Anything that you feel around there that was an inflection point or like, “Oh, that makes so much sense.” Anything like that?

Caroline: Yes, it’s super basic, but our CTAs on page were really boring, and they were small, they didn’t stand out. Back in September of last year, we embarked on a project to redo all of our CTAs and create bigger, more value prop-driven CTAs that are aligned with our brand as a whole. Our CTA strategy before this was very focused on each individual article. If it’s an article about depression symptoms, the CTA is going to link directly back to depression symptoms and then encourage the user to sign up from there. Now, we’re trying to focus on this company-wide goal, actually, of creating more cohesion across marketing channels.

We want what you see on our website to reflect the TV advertisement that you saw for our brand. We want a cohesive brand image and voice. To do that, we created these CTAs that are driven by our platform-wide value propositions. You can switch therapists at any time; that’s a big one for us. If you’re experiencing symptoms of depression, or you live in a rural area, or there’s a therapist shortage where you live, online therapy is available to you. Taking those company-wide value propositions and using them to create CTAs that really pop on the page. That was super successful for us.

It’s so basic, but it’s like the more people are encouraged to sign up for your product and try your product, they’re going to do that. That’s why advertising works, is because people see that, they subconsciously or consciously make a decision to continue the user signup process. That’s been really successful for us.

Ashley: Sometimes the simplest things can be a light bulb. Thinking about the next year, what do you think is the biggest misconception that teams have about AI Search, and something that you want to rectify and change the way they think about it?

Caroline: I think there’s a lot of overestimation on the value of AI Search visibility. Teams that focus entirely on AI Search and abandon their SEO fundamentals will really feel the effects of that. It’s so important to have a strong presence on Google, not just for showing up in LLMs, but also for high-intent users who are looking to find more information on your brand. One of the things that we’ve been really focusing on recently is creating really solid, transparent product pages for each of our types of therapy that we offer. We didn’t have that before, which blew my mind.

If I was a user and I was searching for us on Google, here are all of the things that I would want to know immediately. I want to know how much it costs. I want to know when I can start, how soon I can get matched with a therapist. These things were hidden in our content. I think it just goes back to targeting high-intent users through SEO. AI Search is not the end-all, be-all. We’re still working on figuring out the direct correlation between AI Search and actual business KPIs, just because there’s no tooling set up yet to track that.

That’s why it’s important to remember the fundamentals. Remember that high-intent users are still trying to look for your brand through Google, and make sure that you don’t forget about them.

Ashley: I think everyone knows the mantra, SEO is dead, SEO is dead. That’s not true. SEO is still king, Google is still king, and when someone is looking for a provider and actually for a local search, they go to Google. That’s where you’re going to go. You may find a lot of great resources on social, on ChatGPT, but when it comes to booking and finding your local doctor, you’re still going to go to Google to do your conversion, lower-bottom funnel search.

Caroline: Especially because I think ChatGPT often has outdated or incorrect information when it comes to those really specific queries. It’s helpful in the sense that it takes all of the information on Google and it consolidates it into a specific answer for whatever you’re looking for. Oftentimes, I’ll use AI Search to be like, “Hey, I was working out, and I got a weird pain here. What does that mean?” Then it gives me a generalized answer from all the information online. The times that I’ve used it to search for a really good restaurant in my area, oftentimes that information is wrong, and it’s frustrating. I end up just going back to Google anyways because their algorithm for those local searches is much better.

Ashley: It really is helpful for that. You don’t know what you’re quite searching for sometimes, especially symptom-related. When you’re describing things that feel not quite right, it can help from there, but then you have to dig further, and that’s where Google can really help, because then it continues the journey versus just being contained in one. You can’t ignore how do you control your presence on the LLMs. We do find that if you’re not actively controlling it, you will have brands that are putting out information from years ago, five years ago, because that’s all that the LLMs can find.

If you’re not providing it with more content, it will just put out whatever it has. Even if it’s old and outdated and the restaurant’s closed, for example, or the practice is closed, it’ll still put it there. The one last question. If a brand wants to be recommended by an AI system, if there’s something that you could say a focus on in alignment with your SEO strategy, what do you think they should start doing now?

Caroline: Along with AEO, GEO, a buzzword that’s been popping off in the communities is query fanouts. There’s this whole sense of query fanouts, meaning if somebody goes to AI for a very specific search, like,” I want a CBT therapist in my area,” so we’ll just say, “I want a CBT therapist in Oakland that can take me next week for an available appointment, and that accepts my insurance.” The AI breaks that down into different queries, so it’ll amalgamate the information for CBT therapist, and then therapist in Oakland, and then therapist with availability for next week. It’s fanning out all of these queries and then combining them into a response to that prompt.

I think that also falls in line with traditional SEO optimization for FAQs. FAQs are really an opportunity for a brand to get specific and provide really helpful insights related to particular questions surrounding one general main keyword. I think what I’ve seen personally is ensuring that FAQ answers are clean, concise, and to the point. I’ve seen a lot of brands that don’t really answer the FAQ in the first sentence, and I totally understand that. One of the most common questions we get is, how much does online therapy cost? That’s really difficult to answer because it actually depends on a variety of things, exactly.

Unfortunately, AI search, they don’t want, it depends. That’s not a clean response to present to a user. You have to double down and give them a really concise, specific answer. We approach this by giving a range for how much our services cost, and then that is the first part of the sentence. How much does therapy cost? Therapy costs, insert range here. Then we use the rest of the answer to delve into, “Well, but that can depend on your location, provider availability, et cetera.” I think that rings true in Google, too. It just lines up with the user search experience.

If I’m searching for how much does therapy cost, I don’t want to hear, it depends. I want to hear a specific number so that I can start making plans and figure out how to work this into my budget. I’m looking for clean, consolidated information, and I think that’s exactly what the AI is doing as well.

Ashley: The search platforms want to be useful, too. If their users aren’t getting that specific answer, they’re just going to lose their own customers. That makes a lot of sense, and it’s a really helpful way for people to start thinking about their content, their SEO strategy, and how to show up across all of the different touch points. Thank you so much for this conversation. It’s been super detailed and technical, and it’s good to give some people some ways to think about the search landscape, and not to chase the shiny new trends, but how to start planning and integrating those thoughtfully into existing strategies.

Caroline, thank you again. Where can people find you if they want to connect or learn more from you?

Caroline: I’m on LinkedIn under Caroline Adamec, BetterHelp. I’m not really a social media user, so not there. Where else? I’m currently working on a project outside of my work for a new company called Wine Aisle AI. They’re really awesome, really excited about the work that I’m doing there. Obviously, that’s a little bit different of a sector than healthcare, but figured I would give them a shout-out.

Ashley: Do a Google search for me.

Caroline: Yes, do a Google search and see what pops up.

Ashley: Thank you again so much for joining us, and that’s it.

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.