Live From AI For PR

Chris Norton 0:00

If you've ever wondered why AI moves so fast but somehow still misses the point, this pod could be for you. This is part one of our live recordings from the AI for PR conference in London. I managed to get five people into the basement and record a quick 10 to 15 minute interview with each of them. First up is Andrew Bruce Smith, who's had more appearances on the show than my co-host, Will Ockenden. I interviewed him six weeks ago and he said so much has changed, Chris. So that's gonna be a fascinating chat. Then I spoke to Johnny Bentwood, Chief Data and Analytics Officer at brand new PR agency Golin Ketchum. Johnny actually pioneered GEO before most people even knew what it was. Then we have a man who I've been connected to since 2007 and we've never actually met. James Crawford, MD of PR Agency One and board director at AMEC. And James and I have a fascinating conversation about GEO, AMEC, and what the demonstrable impact can be from PR and GEO. And the one question you should be asking every GEO vendor before you sign anything. So let's get into it. Enjoy.

What Changed In Six Weeks

Chris Norton 1:11

Andrew, thanks for doing this. I mean literally, I've grabbed grabbed you from upstairs and pulled you downstairs to help teach me how to podcast professionally. And we're we're we're both holding mics. You've literally, we've just literally dropped the episode from we recorded it seven weeks ago. Seen as we're at the AI conference, AI for PR conference on the south bank of the river in London. So what's chair you said, oh my god, Chris, I can't believe this. The show's gone pretty much, I think we've had two and a half thousand downloads already. The show is flying off the shelves. But the issue was, we talked about cloud cowork, which you and I both use a lot. So much has changed in those six weeks that we spoke. So you said we should have done it, we should have just issued it that week. So if you had to do it now, what's happened in the last six weeks then that you do you think that you could talk about that's changed?

Andrew Bruce Smith 2:00

Ah, an easy question to start with. Um I've got to get it, get it straight in my own mind about what's uh what's actually occurred. Uh well, you can take it from different uh different levels, different perspectives. I mean, from the pure technology perspective, uh we've seen, for example, Codex, which is effectively OpenAI's equivalent of co-work, or rather, Codex was originally aimed at programmers. OpenAI are always now very envious of the massive success of uh Anthropic's kind of co-work and have effectively turned codecs into OpenAI's um uh uh version of cowork. Uh OpenAI already saying that they've they've seen you know one in five uh you know uh users uh using uh codecs, i.e. normal people, not programmers, you know, using codex to kind of create agents and using it for knowledge work. So that's happened in the last uh few weeks.

Export Controls And AI Sovereignty

Andrew Bruce Smith 2:59

Um I guess the big the big story of the last week or two, of course, was um uh Fable, which was Anthropic's uh allegedly kind of cut-down stroke uh uh guard-railed, if that's even a word, guard-railed version of uh Mythos, which we did mention on Is it Mythos, Mythos, or is that a Greek lager? It's uh it's all the above. Yeah. Um yes, Mythos, just to remind ourselves, uh, which we did mention on the previous podcast, which was a new model from Anthropic, which was apparently so powerful they couldn't possibly let everybody have access to it.

Chris Norton 3:36

Oh, yes, they can!

Andrew Bruce Smith 3:37

Well, uh Fable came out actually was on my birthday, and uh uh people very people got very excited, people were running around starting building things with it and just you know describing all the amazing things they could do with it. Unfortunately, by the Friday, so three days later, uh the US government uh slapped a uh an export control on it, basically saying that no non-US citizens would be allowed or should be allowed to use Fable. Um so it's slightly complicated for Anthropic to actually enforce that. So they've just denied access to it to everybody.

Chris Norton 4:12

Well, I can I can exclusively reveal on the podcast that I did use it for about two days. There you go. It was alright. I don't and well, I didn't try and hack any banks, so I don't know.

Andrew Bruce Smith 4:23

No, and I think that, and again, we we can we can go down the rabbit hole on this one, but uh there's no question that uh in terms of its capabilities on certain fronts, even in the in the in a couple of days that people had uh access to it, that uh yes, it it it it was clearly very, very powerful. But we are in this kind of now limbo state where it sounds as though yes, Anthropic are kind of furiously kind of talking to the US government saying, look, you know, this this is a rather drastic action to take. Um but again, it's it's it has raised that whole issue of of uh you know sovereignty. What happens if you become so reliant on an AI system and you know a government somewhere can literally pull the plug on it? It's crazy. What will you do then? So I think that has it's focused the minds, I think, of a lot of organizations going, hmm, hang on a minute, what what what what can we do to ensure that if we are going to use AI systems, that that we can uh uh make sure that we're in control and not at the mercy of uh of a third party. So yeah, it's kind of hard to comprehend that that all that that's happened in the last week or so. So that's just again crazy times that we live in.

Chris Norton 5:34

And I and and probably by the time we get this out in

Start With The Work Not Tools

Chris Norton 5:36

the next week or so, it'll bloody have changed again. So what you what you hear so we were the AI for PR conference, me and you seem to always move in the same social media and PR uh spaces. So what you're gonna be talking about today on uh what you're here to talk about today, well AI, obviously.

Andrew Bruce Smith 5:50

Well, yeah, yes. I mean I mean the the the the conference is is uh uh basically aligned with uh with the book. The uh AI for PR, Stephen Waddington, Ben Verinder, uh who kind of pulled this all together. Um I'm I'm actually doing a double header uh uh with uh uh uh uh uh Will Will uh uh Will Will Vickery. And um uh I'm focused on my chapter. I I I uh Stephen and Ben admitted that that I had the toughest job for the book because I I swear you had the toughest job in the previous book that we wrote together. I'm a sucker for punishment. Um I mean back in last August I had to complete my my draft of the chapter talking about the AI technology landscape and where was it likely to go for the next 12 to 18 months. Oh my god, I I need to read this. And I feel I feel reasonably pleased with myself that that I think I got it largely correct directionally. There was a lot of emphasis on on agentic AI. Um, I didn't have a crystal ball, so I couldn't have specifically foreseen you know the rise of of co-work so forth in in in January. But but yes, I think I I I I I feel comfortable that that I kind of understood where things were were were moving. So the the the talk is focused on a mix of the same thing. Yeah. Where are we now? Yeah, where's it going? Um, but more importantly, for people attending the conference, it's about but but practically, you know, what what can we do? You know, what what uh what what should we do in relation to you know using AI applied to the work of comms and PR professionals everywhere?

Chris Norton 7:24

Yeah, obviously we've had you on the show. This is your fourth appearance. This is a quick appearance on the show, uh, and thanks for doing it. Um and um obviously with friends as well. But I just want to say, what's your hot tip for uh people that work in PR now? AI-wise, what's your hot tip for working in PR and using AI in a in a in a paragraph or two?

Andrew Bruce Smith 7:44

Well, I mean, I can say this because uh uh uh it's not a spoiler alert for my my talk this afternoon, but uh I guess number one, and I've said this all along, that you you don't start with the technology. Um the tendency is to go, AI, it's tech, you know, what's the best tool for X? Um and that's I I've said along it's the wrong way to approach it. You approach it from what is the work that we do. Yeah, start with the work, start with the work first and understand that um and then try to figure out well, what can, what should we apply it to? I think that the whole thrust has been to treat AI as a technology issue. I mean, I exaggerate for effect, but the classic approach is oh, AI, let's get IT to kind of um figure out what to buy. They go, let's buy a bunch of Copilot licenses, we've ticked the box, everybody's got AI now. Um nodding. End of sort, end of story. Yeah. Um, training is basically a glorified product demo. It bears no relation to the work. And this would apply not just to communications, I mean any department, you know, with respect, does the IT department know in granular detail what the comms team do? Do they know what the finance team does? I suspect not. And that's not their fault, that's not their role. But it's not rocket science, I don't think, just understanding the nature of the work. Um, although I guess if I was being cheeky, that that is part of the problem. Um, most organizations don't actually know how the work gets done. Yeah. Um, you know, management think they know, but they don't really know, um, and they don't really have those kind of deep understanding of of how it gets done. Because it relates to kind of things like AI agents. Now, the kind of consensus view is well, if you understand the workflow and you've got a kind of standard operating procedure and it's all beautifully kind of detailed and written down, you could just give it to your AI agent, and of course, it will now miraculously understand what to do.

Chris Norton 9:41

Yeah.

Andrew Bruce Smith 9:41

Um I would suggest that if you ask any PR team, any agency or any in-house team anywhere and say, are all your processes beautifully written down and fully detailed, the answer will be no. Because no one's ever really properly done that.

Chris Norton 9:56

No. We're not McDonald's, that's why. The PR industry is a McDonald's.

Andrew Bruce Smith 9:59

Um and it you know comms is is broadly speaking falls under the umbrella of of knowledge work. And you're right, it's if you're in a factory, you could just stick up a camera and just analyze what people do. Yeah. If you did that with a comms person, with respect, you'd probably spot them sat at a computer for five hours, tapping away at a keyboard, gazing out of the window, going to make a cup of tea in the kitchen. But the work is going on, you know, a lot of it, well, most of it probably is going on inside their head. So, how do you get that out of their head? Do you even want to do that? Uh, is it even possible? So this comes back to this just uh, I guess, dividing line between what can, can, and should we use AI to be applied to versus what still remains firmly in the realm of the human being, which I've said all along is expertise, uh experience, uh, creativity, strategy, judgment, taste, call it what you will, that's definitely what ultimately has uh value. I mean, I've worked in cons for nearly 40 years. I mean, we've moaned all along about not having enough time to strategize, to be creative, to have more brainstorms, to build relationships, all those things. I still say that the real value of AI is in removing the grunt work or uh reducing the time that human beings have to spend. This stuff needs to get done. But let's be honest, if you had a choice of letting the machine get on undo that bit of it, freeing up your time to focus on the things that really are useful, valuable, etc., I'd I'd take that uh every time. And that's I guess partly what I'm going to try and get across to people today.

Chris Norton 11:37

Yeah, I mean, I'm I'm with you. Uh I'm I'm bought in, but I mean, as I said to somebody, if someone actually sits me on stage today and says, Chris, do you think AI has made you do less work and you're focusing? I'll be like, actually, I spend my time talking to co-work or talking to Gemini or talking to Copilot or ChatGPT, I've got all of them. And I test the all different things for research, and I find myself testing things, oh, that works on that one, it doesn't work on that one, this works that. And then and then you you find yourself iteratively, well, and I know that's how we both work, and not everybody out there does work the same way.

Andrew Bruce Smith 12:11

I think we have to remind ourselves, I mean, it it is kind of crazy that I mean, you know, cowork literally came into the world on the 12th of January this year. That's barely six months ago. Um, this has all happened in the space of six months. You look back through history and be like, this would have taken years for these things to to have developed. And uh, we're still the same human beings we were from 10,000 years ago, uh, you know, living living in a cave. You know, our brains are still pretty much the same brains we had then. Um our our poor brains trying to deal with this sort of rate and pace of change uh is is is pretty is pretty uh pretty tough. So I think I don't think we should too hard on ourselves, but I think you're right. Um we are very much in the early days here, and we're we're all learning, and you kind of I well, I I learn you learn by doing. Yeah, me too. There's there's there's kind of no way around it in a way. I know people want to have the kind of quick solution, but but unless you actually put some time in yourself to kind of viscerally understand it. That's I think it's only then do you get those kind of light bulb moments when you realise, oh now I get it. Now I get it.

Chris Norton 13:21

Yeah, I can I 100% agree, mate. Um, well, um right, and realise we've got to go and go into the conference, and you've got to speak to people, so thanks thanks for coming on the show. We're gonna get you, but we're gonna get you back on. I think we should have like a regular Andrew Bruce Smith slot where we we get you with the here's the latest updates from the last six months. Yes. Thanks, Andrew. Take care.

Why GEO Decides Who Gets Bought

Chris Norton 13:46

Johnny Bentwood, welcome to the show.

Johnny Bentwood 13:48

Thank you. Really appreciate you having me on.

Chris Norton 13:50

So, Johnny, let's just get your title wrong because the first marketing mistake we've got in here is your job title in the in the in the um so what what is your job title, please?

Johnny Bentwood 13:59

I'm the chief data and analytics officer for Golin Ketchum, not Golind. Golin Ketchum is a new company that was formed at the merger of Golin and Ketchum on June the 10th. So we are eight days old.

Chris Norton 14:11

Eight days old. We won't get into uh the name change and and how you can how they came up with it because that's fascinating. But so what is fascinating is what you're going to talk about today. So you're you're speaking this afternoon on GEO, is that correct?

Johnny Bentwood 14:24

Yeah, and it's incredible how many people are getting this wrong. Um, I want to just talk to your audience about what we can do to get it right.

Chris Norton 14:33

Okay, so when you say people are getting it wrong, um what are they getting wrong? There's a few parts.

Johnny Bentwood 14:39

First of all, if you ask any company, the boardroom that's talking about AI, they're all having the same discussion. They're all talking about AI transformation, they're all talking about bringing in Copilot and what can we do to optimize efficiencies? What can we do to create look at our workflows? I go, Here's the crowd. Seriously, I care about the bottom line. And the first mistake that everyone's thinking about is that all about AI is AI transformation. It means nothing if people aren't buying your stuff. Correct? Yeah, okay. So, what can we do to help you influence people buying from you? Now let me give you a few data points. First of all, 75% of people go to Gen AI before they go to Google. Out of those, more people trust the answer of Gen AI than there is. Now, it's not just a consumer space. In the B2B space, it's in the 90th percentile. So, what does all this mean? It means that whilst everyone else is talking about AI transformation, when people are looking to make a decision about what to buy, they go to Gen AI. And this is where it gets a bit freaky. In the past, if I picked up my phone and did a search in Google, I would scroll past all those results on it go, yeah, well, that's been sponsored, they bought those keywords, that's product, and they're right. That shows critical thinking, and I love that. But now what happens is people ask the same question to their gen AI of choice, and they look at that first answer that comes up and goes, Lady, that's perfect, I'm just gonna accept that. And this is goes back to the question you asked when people get wrong. They're so concerned about AI transformation, they're not thinking about how do they make sure that when their audience, when their stakeholders go to Gen AI and ask a question, they are listed in that first answer because people aren't going anywhere else.

Chris Norton 16:24

Okay. Uh yeah, because you've got Gen AI, you've got A, you've got GEO, AEO, same kind of the same thing, answer engine optimization. Is these tools, there's lots of tools out there that tell you or claim to tell you where you appear in search and how you're gonna you're gonna appear versus your competitors. So, for instance, we've got a prohibition, we've got um we've got HubSpot, which we can use to analyze us and where we are, and that's got an AEO uh built-in tool that you get part of the paid platform. But when you look into it, what it's doing is guessing um what you're gonna get. Because how can it predict what 15,000 people are gonna be doing in terms of searches? So I just wondered what your take on these tools out there, which one are they all guess? Is it all guesswork, or is it are any of them factually, statistically correct?

Johnny Bentwood 17:11

So there's a lot of snake oil out there. Yeah, there's a lot of people claiming to solve the world, and some things that do a pretty good job. Look, the point is, big data is better than no data. I prefer to take the results of a synthetic approach where it goes out there and asks a bunch of questions thousands of times a day to see those results come in. That's how the majority of engines work, and that's pretty good. But there is data out there that our tools, I quite like the finders got a billion and a half actual user prompts, so we know the questions that people are asking. And let's make sure we judge our position and visibility relative to our brand's peers across the occasions that matter, across those topics that count.

Chris Norton 17:55

Okay, cool. Interesting. And so, what do you think of AEO versus GEO? Is it even a debate, or should we just ignore the whole thing and concentrate on getting people to buy our stuff from being visible?

Johnny Bentwood 18:07

If people are having that debate, they're having the wrong conversation. I don't care. Seriously, the only thing I care about is what you're listed in the first answer. Now, the nomenclature, what you call the industry, it's not gonna make any difference to your bottom line.

Chris Norton 18:20

Yeah.

Johnny Bentwood 18:21

I look at the outcome, not the vocabulary.

Chris Norton 18:24

So, what's what sort of stuff are you gonna talk about later when you're on stage?

Johnny Bentwood 18:28

There's a few things. I'd like, whilst everyone else is going to talk you about the theory about AI, I'm gonna talk about practical things that everyone can do. And I'm just gonna give you a bit of a highlight into a few of those things. I like to think of it as the four F's. The four F's are your grounding principles that anyone can

Spotting Snake Oil In GEO Tools

Johnny Bentwood 18:46

do to make sure they are listed in that first answer. The first one is can it be found? It sounds simple, but if your content can't be found, it's invisible to Gen AI. So all those websites that have got a password, New York Times, Bloomberg, well, they can't be found. In fact, I had a subscription to Axios, it cost me six grand. They gave me a refund because they made their entire content publicly available. They want it to be found. Anything that's got loads of JavaScript, it can't be found, it's invisible. Anything that's got poor meta descriptions or poor URL structures makes it harder to be found. So the first F if it can be found, great. Next F, what about the format? The format is so critical. It's not just about writing to how humans would like it, you need to write how machines like it. So TLDRs, bullet points, summarizations, there's different structural formattings that work.

Chris Norton 19:48

TLDRs for the audience listening, too long to read, correct? Uh too long didn't read. Too long too long didn't read, yeah.

Johnny Bentwood 19:55

Um give me a short sentence summary about everything. Because Gen I has just got a quarter of a second to scour the world and say, this is what you need to

The Four Fs For Visibility

Johnny Bentwood 20:03

know. So if you can make it into a quick, easy snapshot, a sentence summary in four bullet points, well, Gen I is probably gonna take that over trying to devour millions of pages. If you've got that in easy format, it's gonna take care. So that's the second F. And that could be FAQs, it could be listicles, it could be all kinds of things in that kind of format. The next F is feel. For years, as long as I've been in this industry, we talked about message pull-through and something we want to do. Our customers are talking about our messaging every day, we give ourselves a technology. No more. We've got to stop thinking about our message pull-through and start thinking, are we pulling through the messages of our audience, of our stakeholders? We've got to be using our customers' language, not our language. For example, if someone asks a question, what is the best blue widget for this device? And my answer on my web page is I've got the blue best blue widget for this device. It's a copy, it's an exact message copy. Well, Gen AI says, well, that's semantically similar. I'm gonna take that. So we've got to be looking at the semantic similarity. The fourth F is freshness. Gen AI has a recency bias. So the stuff that you publish today is more authoritative than what was published six months ago. So we need to get away from these ten-hole moments which PR lives in and start thinking rolling thunder. What can we always do to get on top of it? Chris, I've got a fifth F if you want. But it's a little bit salty. It's a little bit salty. It is. Is it right? No, the fifth F is just fucking get on with it. Seriously, because too many people are just talking about what to do. Yeah. And no one's doing anything. Yeah, paralysis by analysis. Yeah. So even if it's just tomorrow, taking every one of your YouTube videos and putting a transcript on it. Well, there's every influencer that you've worked on in TikTok and Instagram, say, you have to put it on YouTube and put it transcript. Whether it's looking at your URLs, making everything lowercase, no more than six words after a domain, and I think 2026, that increases your chances of being found by 11.4%. So just get on with it. Do marginal gains to make a difference. So what's your day to day then? What do you do day to day at Golin Ketchum? So I'm the chief data analytics officer, and I advise the world's largest companies how to be a data-driven commerce organization. Now, whether that is using data to inform what they should be doing, whether it's using data to evaluate the outcomes of what they're doing, it's all part of it. But my clients like PepsiCo, they've really leaned into making sure not only are they marketing to humans, but they're also marketing to machines. And we're hoping to drive and make a difference in both those avenues.

Chris Norton 22:47

Right, okay, interesting. And so and so the hot top the hot tips are basically trans transcripts where you can on your YouTube channel or any anything that you can make more visible. Um, I I had a little bit of a play around with uh the show, right? So the show, um I I heard a friend of mine called John Evans, and he he talked about um he'd said the the the world's best market he put the world's best marketing podcast by by John Evans. And somebody had Googled it uh uh or used generative AI and left him a really nasty message or uh review on his thing because he said this can't be the world's best marketing, but he just the world's best marketing podcast. He took offense to he took a little bit of offense to him, messaged him, and this guy thought it was genuinely the world's best marketing podcast, and it is a very good marketing podcast.

Johnny Bentwood 23:32

You can game gen AI, and is it okay if you call it that? If people ask the question, what's the world's best marketing podcast? Well, it may come up with that because it's semantically similar. You've just seen there how important the semantics are. But for general tips I like to say is follow those four F's and just get on with it. That's the most important tip of the lot.

Chris Norton 23:54

Okay, cool. And uh anything else that you can cover in later? Is that is that is that the crux of what you're gonna be talking about?

Johnny Bentwood 24:00

Those are the main things that I like to focus on. But if we have a look at other areas that are important as far as Gen AI goes, I think there's some common mistakes that people make. And the first one is people just thinking, well, I can just look at what is on my own website, or I can just look at sending out press releases. Now, all those things are great. We talked about website, but you've got most control over. Well, typically that's only about five to ten percent of what influences Gen AI. A press release is fantastic. I never but like putting press releases on the wire. If I've got something that's really good, I'm gonna call that journalist. But when you stick it on the wire, I'm not marketing to journalists, I'm marketing to machines. It goes to 5,000 websites, and Gen AI loves large language.

Chris Norton 24:47

So you think you're you're you're endorsing newswires then. Do you think you should be able to do that?

Johnny Bentwood 24:50

Totally. Love it. But I'm not expecting a single journalist to pick up on it. No, and that's okay. I want to write blog posts every single week, and I come back and check in my analytics in a year, I don't want to be really happy that my readership is zero.

Chris Norton 25:03

  Nice. I like that, yeah.

Johnny Bentwood 25:05

  Yes, we've got to stop thinking about old ways of thinking.

Chris Norton 25:09

Yeah, okay.

Johnny Bentwood 25:09

I want to be able to look at earned media. I've got my tier one media list. We've all got that in PR. Keep going with it. You now need to have a tier one Gen AI media list. You're gonna be engaging with this a lot in a different way. You're gonna be talking about less about editorial and more about comparison format. I'm gonna be thinking about what I can do on social. Social, the big things. What are you doing on Reddit? Ongoing strategy, community engagement, not community, look at me, we're gonna get banned overnight. I'm gonna be looking on LinkedIn and writing pulse articles, not by the brand, but by individuals.

Earned Media Drives AI Citations

Johnny Bentwood 25:46

I'm going to be on cora. I'm going to be updating my Wikipedia. There's things you can do that make a difference, but it's not just one thing. It's doing lots of different things altogether, continuously.

Chris Norton 26:01

Yeah, and also the world's best marketing podcast. You listen to it right now, it's uh embracing marketing mistakes. Um, and it's fine if you haven't got one, Johnny, because uh I know this is a quick 10-15-minute chat. Um, have you made any mistakes in your career that you'd love to share and put that out and get off your shoulders?

Johnny Bentwood 26:19

I've made so many mistakes. I'll tell you the thing that I really think about. I was very, very early into the area of Geo. Um, so much so, you may have heard of a company called Profound. I was Profound's very first customer. Amazing. I realized this geo thing was important. I spoke to this guy called James Cadwallader, and I said, I've got this idea. He's I've also got this idea. He showed me a PDF of what he was thinking of creating. He needed a customer. So he said, Right, we'll be your customer. And with that, he got the funding to form Profound.

Chris Norton 26:51

Wow.

Johnny Bentwood 26:52

I'll tell you the mistake I made. I could see the future. I could see it was really big. Yet everyone in PR is all talking about, hey, you've got to innovate. Because you probably tell your team to innovate. And what happens? Everyone needs to innovate. Yeah. It's I think it's a bloody stupid mistake at times. Because what happens is everyone innovates. Yeah. And there's innovation saturation. Everyone's got the shiniest new toy, the best thing ever, the thing that's going to solve this. And the problem is, when something is truly groundbreaking, it can get lost in that saturation. That's what scares me about innovation at times. I'm not saying don't innovate, but I'm saying you sometimes need to be able to find a way that innovation is tight. You know, when you go to a restaurant and there's a 14-page menu, it's like this is a sign of a bad restaurant. When you go to a restaurant, it's one page and it's got eight things that were fresh that morning, it's going to be great. We need to have a tighter shopping list so there's not saturation so people can see the wood from the trees.

Chris Norton 27:59

So are you saying you made the mistake by having too many, or too many products? Is that what you're saying?

Johnny Bentwood 28:02

T here were way too many things. And when I was speaking to my different client leads, hey, this is what's this is really important. Well, the day before they were told something else was really important. And the next week something else is really important. How would I get to say, look, this is the one? You can't see it, but I know this is really important. So my big mistake is I got lost when we could have been first.

Chris Norton 28:24

Yeah, I still think, I still, I still think I know we're in the uh we're in the PR sphere today, and everybody here is AI related. I still think there's a lot of people at marketers out there that don't really know what GEO and AEO is.  

Johnny Bentwood 28:36

  This is the exciting bit for PR.

Chris Norton 28:38

Yeah, me too.

Johnny Bentwood 28:39

You know what? I've been in rooms with cross-functional teams, multi-billion dollar companies, you know, Fortune 50 companies, and you've got the advertising people, you've got the paid media, you've got the SEO, you've got all the different teams, and everyone will say, we want to own this space. And everyone is justifying why they should do it, because they've all got seats at the table, and they can last at a veneer level until one person asks a follow-up question and they've got no depth. Now remember, 80% of all AI citations is driven by earned media. And who understands earned media more than PR? This is why we have the right to own this space. Whilst everyone else is talking about it, we're the ones who influence it. And that's before we even get to the future. Because a third of all citations, or third of all questions that people ask in a consumer space are creating a shopping moment. That is when you say, I've interested about this, and it says, Well, you can buy any of these three things here. Now, in the Gen Z audience, a third of those people actually buy that. What does this mean? It means that PR historically is top funnel, awareness, consideration, affinity. Now, people asking the question, a third of the time it's creating a shopping experience, a third of those people are buying. We are now going from the question to people buying, and the biggest influence on that purchase funnel is public relations. Booming!

Chris Norton 30:09

I think this is the biggest opportunity for us.

Johnny Bentwood 30:11

Um and PR is like, get a grip, guys.

Chris Norton 30:14

This is our moment. I don't think they realize that PR is 80%. Like I looked through, I I was on the I was on the uh I was on a call the other day, one-to-one with Vuelio. They were showing me their I've I've I've used various um GEO and AEO tools, and uh they were showing me their item and they went, they they jumped into some examples and I was can I stop you there? Can we just look at what's driving the citations? And it was 80% to use your number, it was exactly 80% of earned media, and to use your to reiterate your point you just made there, it was three three months, uh two months out, three months out, six months out, and a year. So all the PR that you've had done in the last year is now is now influencing how people are finding you, but that message isn't coming out of the PR industry, not even making it loud enough.

Johnny Bentwood 31:06

Well, make it loud, and he's telling you what it's not, it's not SEO. No, it's true. The CEO of Condé Nast, just wide, as Vogue, he says he has told his teams to expect search to be zero, a hundred percent zero click, which means people go in there and they will get no traffic from search. SEO is dead, it's an $85 billion market and it's over.

Chris Norton 31:30

But they're the ones with a bloody big budget, so they'll switch it. It's over. Are they gonna eat our lunch or are we gonna get it?

Johnny Bentwood 31:36

In 1910, horses were the biggest form of transport. Volume does not mean trends, okay?

Chris Norton 31:43

Yeah, true, true. Johnny, honestly, brilliant. I could tell you I could literally do a full hour with you, but I don't want to take up too much of your time. That was brilliant. Thanks so much for coming on the show. Thank you so much. Thanks, mate. James Crawford, welcome to the show.

James Crawford 31:59

Thank you very much.

Chris Norton 32:00

So, James, um, how long have we known each other?

James Crawford 32:03

We've known each other. Digitally get must be getting on 20 years. What, 2008?

Chris Norton 32:09

Yeah, yeah,

James Crawford 32:10

so we've never met in real life, which is very strange.

Chris Norton 32:13

But we've been at awards dues.

James Crawford 32:15

Yeah, been at awards dues, but we've never really sat down and had a chat.

Will Ockenden 32:19

And and we've sort of jousted online what during the days of the SEO uh rankings we were just laughing about, which is kind of why we're here today. So you're talking about GEO, um, and I know it's quite interesting because when you come to these conferences, um, you see the best practice, and some there's there's quite a lot of people talking about Amec, even though I would say 70% of the PR industry don't use Amec's principles, and then you I've just seen your um your example, the Barcelona principles, um, and using GEO uh as part of that, so which I thought was quite interesting because we're at Prohibition, we all our clients are Amec measured properly, which is good, and then we we should jump we're using GEO. Uh we've got three tools that we've built in in AI as well for clients, yeah. So, but in terms of geo, and I've just been speaking to somebody else, um Johnny, about it. Uh so I thought I'd talk to you as well. So a lot of it is made up, as in it's it's predictions of what you will end up in in the search. So I just wondered how are you guys doing GEO? How are you how are you selling it to clients? Because I think the industry has done a bad job. When I say the industry, I don't mean the marketing industry, I mean the PR industry has done a bad job of the biggest opportunity that we've

AMECGEO Principles Over Vanity Metrics

Will Ockenden 33:33

got since social, and probably bigger than social, because 80% of things are now cited by PR. Yeah. Um example. So yeah, I just wondered how you're selling it. That's my question.

James Crawford 33:44

I think it's very nascent at the moment. So there's lots of people doing it in lots of different ways. So uh we've been watching it for the obviously the past three years. Um, big interest in search and wanting to make sure we, from a personal point of view, make sure that my business uh continues to get traffic from these types of things, and then obviously applying it to clients and so on and so forth. But quick, it quickly became apparent that uh when it comes to um visibility tracking, you know, there is no universal prompt data, right? So there's no there is no Google AdWords tool telling you what volume there is there, and there's no way of knowing what people are asking. So people are basically asking, putting the finger in the air and having a guess. And there's very various panels that kind of give you hundreds of millions of prompts, but you know, how have they got that data? Is it you know it's still 0.0% of all prompts? So, and then you know how you set that up through various tools like profound or Semrush or Peak or whatever it is to track the outputs. Is it done via the AI or is it done via a Google Chrome instance, all this sort of stuff? So, my uh understanding of it from the very beginning was it's kind of directional. It's I could I use the word telemetry actually, uh rather than measurement uh in the early days. Um, and then we saw all those uh reports um of you know Reddit is the number one source of AI, or it's the guardian today. Oh, it's actually not PR, it's search, you know. Every single person had a different take on that report, and it just shows you how uh varied the the the approach is. So I thought there must have been a better way, and I'm as a board director of AMAC as well as manager director for PR agency one set to kind of create some principles to get people uh away from pushing another vanity metric on the industry because we've just got rid of AVE. So we don't have we didn't see your ask.

Chris Norton 35:29

Um I think we should do a non-PR poll. I'm gonna do I'm gonna poll our listeners and find out how many people, how many people's boards still ask for AVEs, even though the PR agency doesn't want to give it.

James Crawford 35:40

Yeah, and there's a lot we do get asked for it, and and then you know, the the next one was impressions, wasn't it? And yeah, you know, there's loads of other strange fantasy metrics out there, and we can't get away from it, and sometimes clients can't afford it, and you just have to work with what you've got. But um, I was really interested in making sure that we got the best practice there and putting in place um uh I wanted to call it a framework, but it's not a framework because it's the AMEC GEO Principles, um, a tempo for people to innovate around. I wanted something that would work for PR people, for measurement professionals, and also for vendors, for vendors to get behind the inclusives of this world, uh right, you know, whoever it might be, profound. Um, so we set about having uh, you know, hundreds of conversations, uh, capturing lots and lots of data speaking to experts across the piece uh how they're how they were working, and alighted on the fact that what you really need to do is not run before you can walk, focus on what we're calling the upstream signal, the unstructured data of earned media, social media, anything that's on the internet that shapes the environment that LM's source.

Chris Norton 36:45

Yeah, it's it's fascinating to me because I was just what Johnny was saying. I mean, he said he gave this stat that 80% of citations come from earned media, and I was like, I can literally tell you that statistic is true. Because I was on the I was on a demo with um the guys from Vuelio just last week. They're showing me the kit that they want me to use and comparing it with our kit and stuff. Uh, and it was there was an example, and it had 80%, um, 80% of the citations were from earned media, but they were over three months, six months, and a year. And it just that that was my point to Johnny was like, um, I just feel like we've not made as big a splash for PR. That that is such a big opportunity that if you're a marketing person, you think actually, if we have if we invest some money in PR this year, by the end of the year, we're gonna have a big impact on AEO, GEO, and the amount of sales you would think, the actual impact on the business. You're saying it, I'm saying it, Johnny's saying it, but it just seems to have been got lost in the industry. I don't know.

James Crawford 37:49

I think there's it's a it's a growing era, but we've got to remember that the SEO community also think the SEO is the biggest driver of uh. Well Johnny's a bit of SEO's dad. Johnny's as a uh SEO's dad, and that's you know, good on Johnny. I

Context Beats Mentions In AI Answers

James Crawford 38:01

like Johnny, he's a big pal of mine. Good play. We have we have some good you know chats about these things. I'd argue country is it's actually you know 50-50 perhaps, because we've got to remember that uh LLMs, when they need recent data, that what do they do? They search for information. And you know, we've got data, I couldn't give you an absolute percentage figure that shows that the websites that are most visible in search get cited more often too. So it's actually a blend of the two. But you're right. The reason I think the reason why um it's not getting as big a cut through as possible is because when say Google talks about uh LLMs, it talks about webmasters and search, and people think, oh, it's a search thing. But it's it's not a search thing because the search community get what they get wrong is they kind of mention about the technical side of it, and they mention then they mention brand mentions almost as a throwaway. And even even brand mentions is the wrong thing to look at because it's not the mention that is helpful, but it's the it's the context around that mention, the messages, the brand attributes, and who's saying it? The third-party endorsement that really counts. So the search community have got it wrong, and I don't know if they can really grasp reputation, and that's why uh it I think this is our playing field now. They you know if if with digital PR, we all know about SEO and digital PR, it kind of became a bit of a a uh a slave to the SEO agencies and got mechanics.

Chris Norton 39:26

But digital PR is just PR now.

James Crawford 39:28

Well it is, yeah.

Chris Norton 39:28

It's just it's not they're not two different things, right?

James Crawford 39:30

No, no, no, but they are they they yeah, they uh it is just PR, but we could talk about digital PR as a second. We might go off on one there. Yeah, but point being, um I think the SEO community were better at uh commercialising that than the PR industry um and did a really good job of it. I think that shippers sales. Um but anyway, point being this is this is our game now, and we've got to not just walk up, not be kind of like SEO is dead, but bring the SEO community in because we need their data to understand if if a website is technically set up to rank, what is ranking, what does it say about the brand. You know, there's a lot of data there that I think PR people aren't necessarily set up to get. So there's a job to uh work um hand in hand.

Chris Norton 40:15

Yeah, and there's a lot of very good SEO providers out there. We've work we work with a couple of a couple of other biggest in the country, but I yeah, I do think that I what I did find interesting, I used to teach on the um PR degree at Leeds University, and um the biggest recruit, it was quite fascinating. I taught for seven years as just a part-time thing while I ran prohibition, and it was fascinating to watch when SEO really started to take off because I was teaching the digital PR integrated PR as part of what I was and slowly but surely the fastest recruiter of all our students were the SEO agencies. So it'll be interesting to see if the if they're they've moved them out of being in sat in a chicken hand up, like doing certain tasks and not being able to do the full PR job because they they've got a very niche area of expertise. Now, if I'm if I'm an SEO guy, if I'm the owner of a CMO, sorry, if I'm an MD of a search company right now, I'm recruiting PR people like the government.

James Crawford 41:14

It should be, um, but I don't think they're necessarily there are the really there are really good SEO agencies like the Propellernet of this world, right? Who really get it, and I'd argue they're PR agencies. But they you know they definitely came from that search community, but there's lots of people who kind of just treated PR as like a a numbers game of you know spamming out press releases and crappy surveys and stuff, and I think those days are gone. I don't know if necessarily the the MDs of those agencies are set up to truly understand reputation as well as the PR agency founders are. So, and therefore, but can the PR agency founders bring in the uh the SEA the technicals in the same way? Do they have that in their skill set? I don't know. Be interesting, it's gonna be interesting.

Chris Norton 41:59

We need a blended approach.

James Crawford 42:00

We do need a blended approach.

Chris Norton 42:02

And so um my show's all about mistakes. Is there any mistakes that you've made in your career that you're um proud to share that you've never talked about? Go through your roller decks because I've got tons. Literally, what I've been doing is uploads.

James Crawford 42:19

Uh some I'm probably not allowed to mention.

Chris Norton 42:22

Um it's recorded.

James Crawford 42:25

There's nothing criminal. Um, I'd say um I don't think some Andy Bar was first uh for me when it came to online PR and then connecting the dots with the digital PR piece. I definitely launched PR Agency One in 2011 with a at a search conference to do that, and I didn't push it hard enough. I regret having not having really gone all in on it. Do you know what I mean? And then when SEO BR, somebody called it digital PR at some point, I was like, don't call it that, mainly because I didn't rank for digital PR, I ranks for SEO PR. Um I missed that gravy train. I kind of regret that. Um because I think I could have made a lot of money. Uh you know, be that's probably my number one regret. Um number two regret is probably I've I've done I've grown I've I've got a good, profitable business, but I think if I was a bit more grown up in the early days, I probably could have uh done a lot better. But um you know I had a you know I had a really good time. Um I was I was you know the culture of the agency has been fun and I've loved every minute of it. So a few regrets kind of commercially in terms of not kind of really I see some people who've really pushed it and grown their businesses and done brilliantly and good on it. But are they happy? I d I don't know, let's pretend they're not.

Chris Norton 43:34

But that's the thing, like you can you can literally throw it like I

Regrets From Digital PR’s Early Days

Chris Norton 43:37

I my wife calls me a workaholic. She says she doesn't know anyone as driven as I might like like love it, but actually I've tried as I've got a little bit older, I'm trying not to work as hard because I don't think more more money more more clients, more money, more problems, more and actually like you could just suck it all off and I could go off and do podcasting or become a consultant and bringing them and do all right. I live a nice quiet life. But I know I'm not going to do that. No, no.

James Crawford 44:02

I think you I think you're right. I do uh the thing is I find running my business fun.

Chris Norton 44:07

So I will most of the time.

James Crawford 44:08

I will, most of the time. There was a three years ago when the Ukraine war kicked in. That was not fun. But when the Ukraine war kicked in three years ago, that was like dark days for the power industry, and there was like there wasn't much of business about and yada yellow.

Chris Norton 44:20

I think it's worse now with the um I I think what the government, I mean I'm not to get too political on this, but I think what the government did to the NI and the hospitality sector, because we've got a lot of food and drink clients, and we've seen so many clients go bust, not because of the as a big contributor is is the is the NI, the tax and not many people are really on it. They're not really hearing about it. If you look at your local pubs, they're struggling. I just think that I think the government should do something about that. I do think they've made a mistake.

James Crawford 44:47

  I think you're right. Uh, we don't really have much in the hospitality space, but I can see that being impacted, especially the NI thing. I mean, we're seeing it, I mean, with the graduate roles, we used to bring on graduates and train people up, and now the sort of the entry-level sort of living wage is not far off an account manager, you know, and that that that that gap has been compressed.

Chris Norton 45:07

Has it stopped you from hiring somebody?

James Crawford 45:09

I've been uh focusing on hiring more kind of seasoned kind of exactly the same. Yeah, yeah, yeah.

Chris Norton 45:15

There you go, that's exactly that and that they should listen to what we're saying there. That's two business owners of similar size agencies, um, both reputable businesses, not recruiting because of tax implications. It's the tax and the minimum wage that they get by a year.

James Crawford 45:29

It's the affordability of for a year, because people don't stick around as long as they used to. You might get two, three years out of somebody. The first year a graduate is I don't want to say useless graduates, you are great. But you know, you learn. You are learning. Um, and they are brilliant. I think young people are fantastic. Um but they yeah, there's a big risk, risk there with that. So, and um actually what I've found as well, poor old recruiters.

Hiring Pressures And Recruitment Reality

James Crawford 45:56

Um, I spend probably about 30 grand a year, 40 grand a year on recruiters. LinkedIn Ads now brings in uh I spend about spend about four grand uh on LinkedIn Ads for jobs last year. Got some amazing CVs, fantastic, you know, account it's people with one, two, three years experience. Um and therefore, you know, I've not had to usually what put me off hiring the more experienced people was also the extra 20% or 10% on what you paid uh on top. So, you know, now I can get them directly through through uh LinkedIn ads and

Chris Norton 46:29

what I find and I mean this is this is probably not even for the podcast, but what I find quite annoying about LinkedIn ads recruitment though is that really like I'm we're we're based in Leeds, right? So we want people, I mean you can have some people anywhere, but basically we recruit in within an hour's drive of Leeds, you know. Uh not even Manchester, where you guys are. Uh really, some people do do that, but they don't last very long if this drop in two and a half.

James Crawford 46:51

I've got a hard and fast rule about that, actually. I've got one guy who commutes in from Liverpool, and bless him, he's worked really hard and he's still with me and he's still doing it. Yeah, I would normally it would have been a no.

Chris Norton 47:00

Yeah, well, and my issue with LinkedIn ads, you get these from the colours. You get all sorts of no, you know. That's applying from like Glasgow.

James Crawford 47:07

Yeah, 100%. And I you clearly see it in the ads. I don't know what is wrong with it in Manchester or London.

Chris Norton 47:12

Yeah, you you put are you willing to live there and you speak to them, they're like, oh yeah, it's a bit far.

James Crawford 47:16

Isn't that alright? So if I come in, no, it's not a few.

Chris Norton 47:18

All they've done is click one button to apply for a job. Anyway, this isn't really helping our listeners. No, no, I don't know. We can cut that for something else. James, thanks for putting on the show.

James Crawford 47:26

You know, really great to see it. Great to meet you here after all these years. I've been following your business uh prohibition for donkeys years. You do some really great work. Um, and it's from a marketing perspective, also really visible and good proposition, good sense of uh the the digital impacts of the work you do. Um, so yeah, one of the good guys over here, by the way.

Chris Norton 47:47

I mean, thank for someone like you to say that was great. I mean, it's just weird that I've known you for like so long but never actually met you. It's that that's the weird thing about social media. That's the great thing and the weird thing about social media.

James Crawford 47:58

Yeah, and it is a weird it's a completely weird thing. Um it's so weird. I was talking getting onto internet dating, a bit of a tangent, but right I was at a um oh god, it was a comedy night. And it you know, comedian asked, How did you meet a couple? Hinge, you know, what such a depressing way to have actually met someone and it now social media, we've only just met after 10 years, but uh hinge is a bit quicker. I'm not sure we're gonna be on hinge, but um you know what I mean. The technology is a weird kind of friction, so we've known each other, but not actually have a chat.

Chris Norton 48:28

Yeah, I know, and then like um when I talked to people at the conference today, like I know what I've known Wads, I've met him loads of times. I've wrote a I I just I actually um wrote a chapter in the social media handbook in 2012, I think that was on crisis digital crisis management back in the day. Um but uh what's fascinating is I've never seen his, I've never met, I've never been on his boat. He's got a boat, right? And he shares pictures of his boat. I feel like a vision boat. And he's he's like, I can see him thinking, it must be a bit

Online Connections And Final Wrap

Chris Norton 48:58

weird because people know inside your life, even though you know, yeah, you know, you've let them in your window through social media.

James Crawford 49:03

Yeah, no, it's funny, I've uh I've I've never been invited on one to his boat. What's invite us to your boat? Invite us to your party boat, once. Yeah, let's have a party let's have a party.