What You Need To Know About AI Visibility

>> Listen to the episode on Spotify, Apple Podcasts, and others.

TRANSCRIPT:

 

DOUG: I’m really excited for this conversation because a couple of years ago, even a year ago, we weren’t talking too much about AI visibility, and now it’s everywhere. Megan, you do audits of the organizations you work with, and what’s the most surprising finding? And I’m getting the vibe, it’s not something that’ll make people happy.

 

MEGAN: It certainly isn’t, Doug. Unfortunately, the most surprising finding from my perspective was that most of the brands that we looked at were at, you know, less than five percent visibility in AI answers, which basically means that of all the queries that we looked at, they only appeared in about five percent of responses. And so, you know, after we’d run about thirty of these, I turned to our developer and said, you know, hey, could there possibly be a glitch in the system? It doesn’t seem likely that all of these brands are less than five percent visible. He came back and said, no, actually, you know, it is really the landscape view. And what we were able to understand from that is, you know, B2B tech and healthcare brands in general have less visibility than, say, a consumer brand. Market leaders tend to have higher visibility than some of these, even secondary or tertiary brands in these sectors, but at the end of the day, it really is a game of just establishing presence at all. And that’s really, you know, been an interesting finding and something that creates opportunity for firms like ours, but also, you know, it’s something that we have to quickly scale in order to, you know, make an impact.

 

DOUG: Yeah. And you work quite a bit in the B2B space, and we’re seeing and hearing from all the data that we’ve been viewing in the conferences that B2B it’s almost more important because for B2B purchases, there’s an increased likelihood that they are using a large language model to try and get information. So, what are you seeing in terms of the trends that help your clients, brands, organizations become more discoverable with the LLMs?

 

MEGAN: We sort of came about this by reverse engineering a couple of questions that we thought we could answer. So, you know, obviously, the goal here is to figure out how to make brands discoverable in these LLMs. And so, you know, first and foremost, and this, I think, was a really key question to ask, was what is the role and rank of the PESO media across, you know, this mix? So, not just how does PR and earned media affect AI visibility, but how does, you know, paid, earned, owned, and shared media affect AI visibility? And so, you know, that’s really the foundational piece. What we saw was that, you know, PR plays a role, but it isn’t always the primary role in the earned media channel. Owned media and earned media working together is what ultimately helps brands become known, and so that became a really critical finding. We started to think about that visibility from the lens of, you know, all the different levers that could contribute to the outcome. So, it’s not just, you know, getting an awesome placement on Forbes. It’s also, you know, understanding community boards and community channels like Reddit are becoming critical. You know, there are other areas like G2 and, you know, review sites that are also increasingly important: YouTube channels, social media like LinkedIn. And so, understanding and being able to almost diagnose for our clients where their buyers are most likely to find them, what URLs and citations were coming back against the queries that we were putting into LLMs, really became almost diagnostic in our ability to help them refine their strategy. And that strategy is almost always multichannel. Paid isn’t really playing a large role directly in LLMs yet, but in the future, I’m sure that will become an increasingly critical channel as well, but today, we think about it as earned and owned and the right combination of those actions across those lenses.

 

DOUG: Yeah. And what we’re seeing is interesting and always works well, especially since we’re in the earned media business on the broadcast side. So, it’s helping us that earned media swings in and out of vogue. Now it’s back in vogue because of the visibility, and then again your own content can help promote the fact that you got earned and tie it to that information. Another issue that comes up, though, is being described accurately or favorably. And I guess you can be accurate but unfavorable, but you’re going for like the accurate about your good stuff on that. How do you make sure that shows up?

 

MEGAN: We have a saying at PAN that I always bring up whenever I have the opportunity because I find that no matter what changes in the industry, this always sort of rings true, and that’s it starts with the story. So this idea of, you know, the very first thing you do if you’re a brand wanting to impact your AI visibility is you ask yourself really important questions around who you are, what makes you different, what are the problems that you solve, and you understand, you know, holistically, what is your brand story? Brands are using different words to describe themselves across the buyer journey, and that makes them incoherent to both machines and humans, right? So, you’re thinking about telling a cohesive story to humans all the time, but you have to understand that when you have the capability of looking across so many different channels, the way you describe yourself becomes even more important because machines need to interpret that whole landscape, and they can. And so, inconsistent language is one of the most important things to correct out of the gate. Also, you know, removing yourself from the echo chamber, right? A lot of what we see in the B2B industries is using words that, you know, are internal to a company or to an industry, but you really need to start from what are the words being used by the customers trying to find me or my competitors, right? So be really thoughtful about what words you use to describe yourself and letting the queries, the questions, lead in terms of how you describe yourself when you’re doing that exercise. And then I would say third is, you know, this idea that content that could be absolutely excellent, you know, may still be invisible to machines. And so, there’s this natural tension in the communications field that I’ve observed and that I feel myself, where we see ourselves as artists, right? We wanna build beautiful stories, we wanna move people emotionally. But at the end of the day, we also need to recognize that there’s a balance that must be struck when you’re appealing to both a human and a machine. And so, you know, when it comes to algorithms, we really have this responsibility now as communicators to help our clients understand that the most thoughtful and amazing piece of thought leadership content may not even be viewable because it’s not, you know, checking some boxes that are really critical as far as the algorithm is concerned. It’s balancing the art and science of communication now and understanding and bringing those rules to bear for our clients that makes, you know, us successful and really helps them to achieve that accuracy and favorability in, you know, the visibility that they’re seeking.

 

DOUG: Yeah. And I like that you talked about going on the offensive when it comes to AI visibility versus trying to play defense or overreact. What do you mean by that? I’m guessing a lot of it is about the consistency of content, getting it out there, earning media, working those two most popular channels to earned and owned that are popular most within the machine learning.

 

MEGAN: I had a conversation recently with an old colleague of mine who now runs communications at a large cybersecurity firm. And it really was helpful in sort of calibrating the way that I talk about and think about, even with all of my clients and my colleagues, you know, what offensive AI visibility work looks like, what offensive PR looks like.

 

DOUG: And just to clarify, this is offensive being proactive, not offensive being, like, offending everyone. Just to clarify, let’s not offend everyone.

 

MEGAN: That’s very, very true. Taking the offensive. Yes. But ultimately, what she had said to me was, you know, there was a new leader that was coming into the CMO seat. She was a demand gen sort of revenue-focused leader. And so, she was feeling like her role as the PR and communications lead was, you know, really being sort of devalued in that mix. And so, I said to her, you know, we have to really rethink the way that we think about the job of the communications person in any firm in any industry, and it’s really like you are the keeper of credibility channels. So, you start to think about credibility as a lens. It’s not necessarily where those where that credibility opportunity, you know, shows itself. You know, oftentimes, we would traditionally think of ourselves as, yes, it’s the earned channels. We built we bring the customer and the proof and the stories to those earned channels. But now it’s really about identifying all of the areas where credibility plays a role. So that’s the review sites. That’s, you know, the accurate and realistic responses that are coming back in a Reddit thread that is, you know, it’s what you’re writing, you know, from the point of view of your executives on LinkedIn. I mean, at the end of the day, it’s really being able to put that rigor around what is believable, what is true, what needs to be incorporated in this communication that I’m putting out to the world, whatever the channel, that makes it believable. And so, you know, if you think about yourself as almost like a shepherd of credibility across the AI visibility surfaces, you become so much more critical in the mix, right? And so, that’s the discussion that we had, and I sort of ripped off of it over time. And I know that she loves it, and she’s delivering that same talk track internally. But I think that that’s the opportunity. That’s what, you know, going on the offensive as a PR and communications professional looks like. Not the defensive, which is to try to validate that earned media is the number one most important primary channel in AI visibility when in fact, we’ve seen it’s not, right? It is a support channel. Owned and earned together is what makes it work, not one or the other.

 

DOUG: Right. They have the multiplier effects on each other. And, you know, obviously, I’ve been thinking back and forward given that the company’s approaching its 40th anniversary in that time even in New York. Had to walk through snow a mile even during the middle of summer, so I don’t go down there, but I was trying to look at, you know, what stayed the same. And one of the things that I came across, and you touched on this in an earlier answer, is brands know their answers to stuff, but it’s always been about figuring out what the questions people are asking are. Almost like the game of Jeopardy. If you think like, do you know the questions? And if you can figure out what the questions are, that’s going to put you in a better position to actually be answering them and having more AI visibility. And you know, now from your perspective, there are so many more tools that help you identify what the questions most probably are on that. How is that playing into this sort of whole strategy that you’re putting together for clients?

 

MEGAN: So, you know, we really back into building very custom AI audits for clients that are, you know, that are leading. I mean, we look at both the client and we also look at core competitors. But, ultimately, you’re hitting on the very, very important piece of it, which is sort of refining with them what are the queries that we wanna go test in this audit, what we wanna see, who shows up, what is the sentiment, what are the actual messages that are, you know, that are coming back in response to these questions, but what I think is really interesting and where we challenge clients is that oftentimes clients have a legacy and have a future state that they’re pursuing. And they bring on communications firms and marketing firms like PAN because they wanna change the perception of their organization in the world. And when it comes to AI, a lot of what is being, you know, funneled into the system is actually legacy messaging and positioning, right? Legacy lists and things that they’ve been included on over time. And so, like, I’ll give you the example. We did a very deep, deep dive for one of our clients who is a government contractor. They, you know, they are the leading provider of AI capability to the federal government. But when we built the, you know, the question set that we were gonna go test for them, the query set, we forced them to not just include, you know, questions like who are the top AI contractors for the US government, but also who are the top, you know, government contractors in general. And what we saw was that, you know, they’re getting a lot of lead source from their legacy positioning, from the legacy questions. That’s what the audience is asking for, right? The audience isn’t asking for the biggest provider of quantum necessarily. They’re asking who are the best, you know, government contractors that I should be working with. And so that’s where we start to see a little bit of resistance from the communications team with very specific directives to position the company for the future state, but there’s a balance required where we have to acknowledge that, you know, the core and the legacy of a brand is still driving a lot of lead source, you know, to the website and to, you know, back to the company. So how do you protect some of that, maybe reinvest in the best of that, and then start building new and, you know, putting new inputs into AI to start to change that narrative around the company? It’s required that you recognize and you sort of treat both. It’s like when you introduce a new product, right? You can’t ignore your core product. You still need to fund that, but you also have the new product that you need to support. So, you know, it’s a sort of bifurcated strategy, but that’s a really critical part and very surprising to a lot of comms teams.

 

DOUG: For the people and the machines that are watching and consuming this content, they’re both gonna be impressed. Thanks so much for joining us.

 

MEGAN: Thank you.