Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Earned First

Earned First artwork

Burson's Red Surtida on GEO's 'credibility paradox'

Jun 9, 202635 minAudio

In this episode, Burson's Asia-Pacific head of intelligence and transformation Red Surtida joins Earned First founding editor Arun Sudhaman and research and insights editor David Blecken to explore the findings of the 'Credibility Paradox', a study that examines how AI answer engines assess and surface brand content. The conversation covers why corporate leadership content is underperforming as a credibility signal, how employee experience is emerging as a reputation asset in the GEO era, the channel dynamics that vary significantly across industries and markets, and what it means that business decision-makers find AI-generated answers more believable than the general population does.

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Welcome, everyone, to the Earned First podcast. My name is Arun Sudhaman. I'm the founding editor at Earned First. I'm very happy to be joined on today's podcast. We actually have two guests. We have with us from Burson, Asia -Pacific Head of Intelligence and Transformation, Mr. Red Surtida, who's based in Singapore. Red, welcome to the podcast. Hey, Arun. Glad to be here. And we have

alongside my colleague based in Tokyo, Earned First's Research and Insights Editor, David Blecken. Hello, David. Hello. Hi there. So we're going to talk about, well, GEO, Generative Engine Optimization. Broadly, it's got to be the... topic of the, not just the month, but of the year, I feel, in the PR world. Every day, I feel some people are talking about it. And they often talk about

it as if they've just discovered it, which is a little disconcerting, I find. But we are with people who definitely haven't just discovered it. Burson has just released a credibility paradox study, which they completed with a company called Profound. And it throws up some quite interesting findings into how GEO is evolving. And David is more familiar with some of these findings. So he will be perhaps adding some color and insight

and questions as this discussion progresses. But one of the key findings I think read that we could start with is that showing up in AI answers. isn't really enough. It creates something of a, perhaps even a visibility trap. Yes, now we know that AI is summarizing brands, presenting brands, showcasing brands to millions of people via AI platforms. But how much does that matter

if people are not actually believing what those AI platforms are actually telling them? Yeah, I think that is where the shift is happening. The way we look at it, the first phase of GEO was about just showing up, right? A lot of brands, a lot of organizations wanted to be present. But visibility alone without believability is exposure without influence, right? So visibility

alone does not drive action. If your goal as an organization or as a brand is impact, you need to be credible. You need to be believed by your stakeholders, your audience. And if you're just focusing on just showing up, your impact is often short -lived. And this is where reputation will come in because it's just not about being visible, but also having that credibility with

your audiences. can i um can i just uh something that i found interesting uh about this research and some of the sort of assertions that were made um around the research um whether you know this uh the fact that visibility is not enough but hasn't that always been the case i mean why is that different now with uh with AI, with GEO?

I mean, hasn't it always been the case that just sort of being present and being visible is not really enough to, you know, to build? Yeah. And that's exactly the point. Because right now, a lot of brands and organizations are just focusing on being visible. But if you take a look at the other channels, the other touch points. that we play in like social media uh earned media paid media even right just being there if if

you're spending money um running paid media ads but if your content is not credible then you're just wasting money and that's the same thing for ai and what is what is it that ai is actually rewarding because obviously brands are going to make lots of claims And presumably the AI platforms are looking to verify those claims in one way or another. How does that actually

work and what does that mean for the reputation investments that companies are making? Yes, that's a very good point because at the end of the day, most brands, most organizations, what they will do is they will build an ecosystem, or at least the advanced organizations that are looking into getting into AI answers. They will build a content ecosystem tapping into earned media, owned media,

paid media, just to be able to get that a foothold in the channels that matter to AI. But what we're forgetting would be even if you have that content ecosystem. But if your target audience, your stakeholders, don't believe the content that you're trying to push, then that is where the visibility gap comes in, the credibility gap.

And yes, it's very easy or relatively easy to... to get AI systems to consistently favor your content, right? We still need to take a look at the people angle because brands are forgetting that at the end of the day, their audiences are still the key decision makers, right? AI is just a touchpoint. It's an influencer, but your customers

will still make that final decision. And do we know Because these algorithms, you know, we don't necessarily have a lot of insight into how they work. So do we know, for example, I mean, we hear that AI will prioritize authoritative and media sources. So you could say perhaps it will prioritize better journalism. perhaps make the

claim that you know better journalism is more likely to scrutinize claims that brands are making are these like are these reasonable assumptions to make or you know are we kind of getting ahead of things yeah i think those are very reasonable assumptions to make and the way we look at doing this is looking at patterns right most of The audits that we've done, we've done, I think we're up to 70 plus now audits across different industries,

across different brands, across different markets. Patterns will show up where there are certain channels that will impact your visibility, like earned media. You also get visibility through, for example, Reddit, Wikipedia, like those types of channels. But it also depends on the industry and the markets. For example, if you take a look at markets like Japan, China, Korea, the relevant

channels within those three markets are either semi -closed or closed ecosystems. An example would be WeChat. They have their own LLM. They have their own ecosystem. the AI that people have access to within WeChat will only get information from WeChat. So even if you publish something from a very reputable publisher, for example,

if it's not on WeChat, you won't show up within that market. So it really depends on the market that you're in, but also the industry. We've seen, for example, within consumer tech, Reddit is quite important within that channel. Review sites are quite important within that industry, rather. But if you then shift into an industry like enterprise tech, they're both within tech,

consumer and enterprise. Review and Reddit might not be as important. A lot of factors are playing into it, and that's why we advise clients to not just rely on what's readily available online about advice on geo, but actually do the audit because you need to know what are the relevant channels, what are the relevant formats even, like are videos important for your industry, for your brand? influencers important right we've

seen um brands and Industries where LinkedIn is quite relevant like they show up in in citations um and then of course you need to take into account believability uh if you have seen the the report it says that leadership is actually quite low when it comes to believability so even if they are being like LinkedIn is being cited uh audiences are not believing the things that they see on

there as much. David, I wondered if you wanted to come in now if we talk about this leadership problem because you had led our coverage of this report with the headline that leadership lacks credibility in the GEO era and that the study found that that uh corporate leadership is largely ineffective in building reputation in this context yeah um i mean um yeah i mean there are i have

a few questions about this but uh you know do you have a sense of why this is uh red what you know what why um was this a surprise to you when um you know when this finding came out of the research Was it surprising or was it something you expected? For me, it was a bit of both. It's surprising that it's that low. But if you take a look at how people are writing content on LinkedIn,

for example, because it is the main channel for leadership content, I can see why it's low. It's the way leaders are creating content on their channels. um and just open linkedin and you will see a lot of of click baity content right now um and i think that is where uh the the gap is coming from and i think as pr professionals are

also to blame on this because we have uh made it so that that is the case that is the environment leaders are uh creating content on so we i mean do you mean it's just that um there's you know is it is it the um question of authenticity you know there's just not enough authenticity uh from leaders they're they're playing it too safe or they're you mentioned click baiting stuff

but i mean often it's you know we see virtue signaling and uh all manner of things like that but um is it is it ai generated content that's to blame This would be kind of ironic, I think, that AI is actually to blame for AI platforms. Exactly. And I think it's also because LinkedIn content, even X, are being cited by these LLMs,

by these AI answer engines. So they are showing up. But then when people see the results and they see it's from LinkedIn, that is where the gap comes in. It's exactly what you said. Some of them can be AI slop. Some of them can be virtually signaling. So there are a few factors playing into it. And I think it's a good opportunity for executives, for leaders to now take a look at how they are creating content and how can

they be more relevant, more credible. now that AI is actually citing them. And do you have any words of advice on how they can be? I mean, is it just a case of being less polished, being more human? Yeah. What would you say to them? I think aside from getting in touch with us to help them with their content, definitely being

more authentic. Because a lot of leaders that we see online are either trying to fit a mold because they see other leaders doing it, their peers doing it that way. So it seems inauthentic. But also lacking proof when it comes to the way they create content. It can seem to... too focused

on selling, too marketing, and not enough authenticity around what they're publishing. Do you think the industry has perhaps overestimated the value of executive visibility while underestimating the value of proof? Would that be fair to say? I think that's a very complicated situation.

I would say it's a bit of both because we do need proof, even if AI wasn't around, right? But because executive visibility works, that's why a lot of executives are doing it. But I think in terms of how we do it, how... content is being produced, how executives are being positioned, there is a lot of opportunity to work on. And

I think AI is shining a light on it, making it one of the bigger opportunities for us within the industry to work on. Can I just ask very quickly, how did you assess believability? Yes. So we do have a tool called Burson Decipher. So we worked with a cognitive AI company called Limbic. So we have partnered with them for three

years now. And we built Decipher. And you can think of Decipher as a consumer panel in your pocket, right? So this specific tool, what we do is we... put in content so any type of content you can imagine be it social media post a press release for example and statements right about the specific areas that we wanted to cover for this study and then what this tool will do is

it will assess two things the first one would be how believable that content piece of content is piece of statement is to a specific target audience and then Second one would be how engaging will it be? So for this study, we focused on believability, which is how relevant that content is. And it gave us different scores, different forecasts for each of the piece of content that

we covered. And the good thing about this tool is it's not retrospective. So it doesn't take a look at what happened, but it's trying to predict how people will react to. that specific piece of statement right so it's it's it's taking into account the historical data but it's giving us um a prediction as to how people will engage them so what about the importance of the workplace when it comes to believability um one of the

findings that david highlighted from the study is that employee experience is one of the strongest credibility signals um when we're talking about ai generated content well or ai generated answers let's say um but how does employee experience show up in these types of platforms is it something that's being given a lot of priority to it's

it's one of the i i think it's more on it depends on the type of prompts being used by target audiences right and the good thing about uh employee experiences is that it is a credibility multiplier it does show up in ai answers but because it provides human experience on how a company actually operates it makes the answer a bit more credible, or the

content more credible. And I think you can compare it to how leaders, how executives talk about their brand, and then compare it to how employers talk about their brand, their company. And because of who is talking about the content, it makes sense that employees are more credible in the

eyes of the different stakeholders. So now, brands need to view workplace in terms of the reputation, evidence system. It's not just about talent and culture, but it's about telling your brand story as well. Because when it comes to AI, it will get content from... employee reviews uh talent rankings even labor coverage right um awards hiring signals those are things that will get

uh cited by ai all are getting signals in ai rewards so it's quite important as well so we need to rethink how we position employer branding so it's not just an hr story it's it's it's actually a reputation asset you know we've we've said um that it is very important for reputation for some time and it sort of comes and goes in uh you know you see sometimes there's more importance given to those things employer branding internal

communications and whatnot and then you know interest fades and they they go back to you know they take a back seat again so um you know is is uh geo now the the thing that will change that do you think will they actually become core reputation functions this is something that we believe in um because in terms of how we measure reputation it's not just what companies or brands

say about themselves it's also what the different stakeholders say about them right it's also how what they do right and also what people actually believe in. And there are a lot of factors that will influence those four things, those four areas. And employee branding is, or employer branding rather, is one of them. One of the other things that I found interesting is that business

decision makers find AI generated answers 10 % more believable than the general population does, which I think probably mirrors our own experiences. I certainly find that civilians, let's say, are less enamored with what they read on AI. But what does that mean for corporates? Because it's easy to assume that everyone is thinking like you. Yeah, I think it's very interesting

because in terms of us, like the... the people working within this space, we are a bit more familiar with AI and how it works. And it makes sense that the business decision makers actually would find AI generated answers a bit more credible because they know what the pitfalls are when it comes to these answers. So they are aware of the limitations. They know what to trust and what not to trust. And then if you take a look

at the general public, and the type of news that they see about AI, it's not surprising that a lot of them or most of them are a bit hesitant to believe what they are seeing. So it's very important now to take a look at the believability factor of the content, especially for brands that are playing within the B2C space. Because

the audiences, the stakeholders that you're trying to target are more hesitant to believe AI content. But for the B2B space, it's still important to know because there are still a lot of decision makers that are not as exposed to how AI works. I firmly believe, at least in the next year or

so, this is where we need to focus as communicators our attention on. Making sure that the content that we produce that gets cited by AI would actually translate into action from our target audiences because they believe that content, they find it relevant. Sure. And then another finding that I think is very relevant for us in Asia, no doubt, is that whilst there's, you know, overall, yes,

there are believability problems, there are also specific translation issues in, you know, in regions where there are multiple languages, right? companies, brands, communicators overcome the English language bias of LLMs. You know, they may miss linguistic and cultural nuances. And

then, of course, you have, as you've already mentioned, you've got local platforms like WeChat. You've got specialist closed ecosystems as well. How do you see that changing? corporate engagement strategies operate in a region as diverse as Asia? Yes, I think this is where our approach to AI visibility will come in because we created a mandate where in order to build motivation,

not motivation, in order to build your reputation when it comes to showing up within these AI systems, we take a look at the 3C model, right? So number one would be consistent narrative so it's not just looking at one specific platform you need to show up in multiple platforms and make it consistent right make sure that you are in the right channels so that's compounding authority so you know which channels are being cited which

channels are influencing ai and then Lastly, make sure that when you execute your campaign, you are actually doing it with purpose. So the chorus of content. So make sure, number one, consistent narrative. Number two, compounding authority, having the right voice. And then lastly, chorus of content. So that mitigates the gaps that we see when it comes to non -English content. So even if your content is not... everything

is not in English, at least you know which channels are important to AI. David, anything from you on that point? Well, I think, I mean, that's obviously something that it seems as if that will not be easy to address. Presumably that will take time to get right. I think it also depends on the industry because you see, categories or industries, verticals that are producing a

lot of content on a daily basis, an hourly basis even. And for those types of verticals, categories, I think it's much easier to influence or it's much difficult, much harder to influence AI because you are competing with a lot of content online. But then you have categories where it's a bit slower, where you publish maybe once a week, once a month. So your share of voice is a bit

higher. So there is a higher chance for you to influence what AI is saying about you. But that doesn't mean that for categories that are a bit more fast moving, that doesn't mean it's difficult. It will also depend on the brand itself, because if you are one of the bigger brands, you might be able to control the narrative within your space. So there are a lot of factors playing in terms of or dictating how AI will describe

or share answers. It really depends on... on those factors so it's very important for brands to know what those factors are and maybe just to close red you've been obviously doing a lot of work in this area and and reviewing how ai answer engines work is there anything in particular that has surprised you that has maybe um you know is is really made you rethink how messages

are shaped and how reputations are built compared to perhaps the conventional wisdom that we have been living with for so many years? Not really surprised, but more of, I think, annoyed because we have seen AI cite information or data from 2011, for example, or 2018. That was ages ago. And a lot of those information are outdated.

So it makes it a bit more difficult for our clients, the brands that we work with, to actually correct those outdated information. The good thing is they're not misinformation or disinformation, but they are outdated. So it is a bit annoying. there is opportunity to correct them. Yeah. And I guess it gives you a clearly visible picture

of where those gaps might be as well, which is not unhelpful. Yeah. And this is actually funny because we are now seeing brands audit their websites because there are pages that are being cited by AI that are clearly outdated something that they have published ages ago 10 years ago and is still being cited so they now need to take a look at their website uh clean things up and make sure that it's more um more applicable

to to what we have now does it mean that we'll i mean you know in in this um kind of new new era um Does it mean that we'll see more in the way of actual substance? Will we see more substance rather than just positioning, I suppose? Do you think this will lead companies to pay more attention

to, as you were saying, the website? They perhaps haven't prioritised it and there's outdated information there. you know will we now see more attention given to that and and more attention to the quality of um of content that that they put out or being produced yeah i think that is the ideal situation but of course that takes a lot of resource um and time which most companies most brands won't

be able to provide right or to to access so i think there is a move or the the ideal situation would be to move towards substance but of course not everyone will be able to do it so some brands might look for a shortcut to using airline the obvious shortcut yeah but would people actually believe what they see which is the question yeah

they need to answer yeah how do how do ai answer engines What do AI answer engines think of AI content itself? Do they have any particular preference for it or not? Yeah, I think it depends on the content because we have seen instances where AI is citing AI slow, which is the problem for most brands, right? so there is no hard and fast

rule that ai will cite specific types of content so it really depends on how it's written where it's published and some brands are taking advantage of of this uh hopefully it gets fixed down the line but with how we are or how these ai engines are structured right now There are still a lot of gaps that we need to watch out for. And hopefully

it gets addressed soon, but you'll never know how fat it will be. So brands are taking advantage of it. There's no, what do you call this? There's no rule right now. It's the Wild West. Brands are experimenting. Organizations are experimenting on the content that they're producing. Yeah. And maybe just finally, Red, as we mentioned

before, these algorithms are black boxes in many ways. Do you feel like there's a possibility they could change a lot? Because we are relying on how they perform today. um what's to say that that you know all of a sudden they won't start um prioritizing different types of media in you know six or 12 months time yeah and you'll never

know uh because if you take a look at what has happened in the past i don't know two years even one year a lot has changed in terms of how things are being cited. So you'll never know what will happen in the next two months or even by tomorrow, right? And this is why brands and organizations need to be on top of what's happening because you'll never know if your current plan, your current strategy might be outdated in a couple

of days or weeks. Okay. Well, keep us posted. And thank you very much for your time. Thank you as well, David. You can find more about the Credibility Paradox study on the Earn First website. But a big thank you for listening to this podcast. And of course, as always, thanks to our production partners at Earshot Media.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Kaizzen’s Vineet Handa on monetisation, Middle East expansion and AI enemies

    Oct 30, 202637 minAudio

    Vineet Handa, founder and CEO of Kaizzen, joins the Earned First podcast to discuss one of the more unusual independent agency stories in India. Founded in 2008, the firm sold a majority stake to Maven Corporate...

    Credit the people on this episode

  2. Jean-Léopold Schuybroek: Four decades of building independent PR

    Oct 13, 202625 minAudio

    Jean-Leopold Schuybroek, chair of PROI’s International Development Group, joins the Earned First podcast to reflect on four decades in the consultancy business. The Interel founder and former ICCO chair has spent 20...

    Credit the people on this episode

  3. Geopolitics, Mastercard & OpenAI's CCO search

    Sep 29, 202639 minAudio

    Earned First's David Blecken and Arun Sudhaman look at the backlash over Singapore Airlines' stake in Air India, Chinese brands going global, and Mastercard's global PR pitch. They also discuss OpenAI's nine-month CCO...

    Credit the people on this episode

  4. Why corporate affairs pros want change, with GlobeScan's Chris Coulter

    Aug 24, 202636 minAudio

    GlobeScan CEO Chris Coulter joins the Earned First podcast to break down findings from the latest Oxford-GlobeScan Global Corporate Affairs Survey. Now in its seventh edition, the study draws on nearly 300 senior...

    Credit the people on this episode

  5. Culture, Cannes & cockroaches with Charlotte Mceleny

    Jul 30, 202647 minAudio

    Charlotte Mceleny returns to discuss Cannes Lions 2026: the creator economy's growing dominance, the industry's fixation on "culture," and renewed momentum behind CCOs and organisational trust. The conversation turns to...

    Credit the people on this episode