Defy Organizational Gravity with Synthetic Personas - with Dr. Matt Wilkinson

00:00
Daniel Nestle
Welcome or welcome back to the trending Communicator. I'm your host, Dan Nestle. You know, whoever said that many hands make light work never worked in a corporation. Companies throw many hands at problems all the time. But sometimes, and most of the time in comms and marketing anyway, many hands tend to make for flat, lifeless, and shapeless work. Let's take messaging, for example. I mean, we do everything right at the front end. We sit in on the customer interviews, we read the sales call transcripts, we talk to the shareholders and the stakeholders, and we hear everything we need to know. We hear the words that people actually use and the feelings behind them. We hear the flavor. And then it goes into the many hands. Meat grinder, brand team, legal, regional teams, regional leads, executive sponsors, everybody who sat in on none of the interviews.

01:03
Daniel Nestle
It goes back to legal. Maybe HR has a hand in something. You know, there's so many hands touching what we've created, and what comes out on the other side says roughly what the customer said. The meaning is there, but the language has been ground down into corporate paste and the emotion is gone. Then we ship it. We put it out there in the market, whether it's an ad or whether it's something that our CEO says. And we wonder why nobody in the market reacts to the words that come from an idea that we originally lifted from the market in the first place. Well, I've spent a lot of years inside that grinder, and I've attributed these problems to editing by committee, too many cooks in the kitchen, or just her standard risk aversion. But my guest today, he nailed it.

01:55
Daniel Nestle
He calls it organizational gravity. Now, he spent 15 years running commercial teams in life sciences, where every claim goes through medical, legal, and regulatory review. I mean, that's one high gravity meat grinder. So he built something to survive it. A way to keep the buyer's voice in every step of the process. And his new book, the Buyer in the Loop, lays it all out. And so will we in the next hour. Dedicated listeners may remember him from his first appearance on the show way back in June with our friend Brian Piper. He's a chemist turned marketer, communicator, speaker, author. He's an innovator doing incredible work with synthetic buyer Personas. Please welcome back to the show, although this time without any help from Mr. Piper, the founder and CEO of Striven, author of the Buyer in the Loop, my friend, Dr. Matt Wilkinson.

02:54
Daniel Nestle
Matt, good to see you.

02:56
Matt Wilkinson
Thank you for having me. Dan, great to be back.

02:58
Daniel Nestle
How are you? How are you? What's going on.

03:00
Matt Wilkinson
I'm good. Thank you. I'm good. Missing the fact that summer seems to be ending rather quickly over here in the UK but other than that, I.

03:07
Daniel Nestle
Know the time of the recording, it is. It is just before Labor Day here in the US And. And in the uk, I suppose you don't call it Labor Day, but. But here it is early September. The weather is supposed to be getting cooler, but I don't know, we're in a heat. We're still in the heat right around where I am, but I hear you, I feel you. And I was going to say it's been a while, but it really hasn't because we catch up a lot and were just on the show back two months ago, three months ago. And I'm assuming that all of my listeners, of course, listen to every single one of my shows and remember exactly who you are.

03:44
Daniel Nestle
But just in case we got some new people listening in, maybe we can just quickly say, like, you know, just fill people in on who you are, Matt, and. And why you're here. And. And, you know, I. I have to say that I want to really get into your book today because the book itself is amazing. But it's everything that kind of surrounds that conversation about the buyer in the loop that is so relevant to our work today, to what everybody in comms and what marketing are doing. So please, go ahead. Floor is yours. Who are you? Why are you here?

04:24
Matt Wilkinson
Thank you. So I'm a recovering scientist. I started my career as a chemist, as you said, and I'm also a bit of an accidental marketer. My first experience, my first marketing campaign was actually when I was working as a postdoctoral fellow at the. At the University of Amsterdam, where my boss at the time said, hey, you know, we need to improve our undergraduate recruitment. And, you know, I said, were having coffee and I said, why don't we do a fireworks lecture? You know, I was fresh out of undergraduate. And he goes, what's a fireworks lecture? So I told him, what is a fireworks lecture? Where you get a group of people in the lecture theater and then you set fire to stuff. You have explosions, bright lights. Basically, you host, you show off all the fun stuff about chemistry in a lecture theater.

05:17
Matt Wilkinson
So we had, you know, magnesium burning in oxygen to, you know, to bright light. We had, you know, we set fire to stuff. We made liquid oxygen and then set fire to cookies and a whole range of stuff, balloons full of ethylene and oxygen to blow that up. We left scorch marks on the ceiling. We nearly deafened the dean. It was amazing. We doubled undergraduate recruitment. In fact, nearly everybody that turned up to that open day signed up. So we had a lot of fun. But what it made me realize as I look back and in hindsight, was that were speaking the language of our customers. Then the students that are attracted to chemistry do. So it's not squishy biology, it's not theoretical physics. It's fun, it's exciting. And people do chemistry because it's explosive, because it's dangerous.

06:05
Matt Wilkinson
You know, it is the most fun of all of the sciences. May not be the most useful these days, but it is definitely the most fun. Yeah. So that was kind of how I got into marketing. But, you know, I then wound my way through science and business journalism into comms, you know, via an mba, and then. And then fell into marketing really having, you know. But the key thing that I've been doing for the last few years was that, is that I've really seen that with the way that AI is just sort of transforming marketing and communications, you know, that the ability to create stuff, you know, is everybody has it now, but the ability to understand your customer is one of the few things that really sets us apart.

06:54
Matt Wilkinson
And I've really been focusing over the last year, 18 months on how do we bring those customers to life and solve some of these problems where customer insight is just stuck within an organization. It's not shared, it's not useful. How do we bring it to life? How do we help people focus better on the customer?

07:18
Daniel Nestle
It's. It's a problem. Like it's a, it's an epidemic of sorts with corporate marketing and PR and you know, I think that's the real question that people are missing is, you know, how did we get from that excitable, worried, you know, very emotional buyer, you know, who had reasons, who had like, motivations, seemed to be turned on by what were saying. How do we get. Get from there in the research, right? And then, you know, when we finally put out the campaign or the, or. Or even, or the talking points or we get the CEO interview and he's. He or she's talking about something using what we think we learned from the research, it just falls flat like, doesn't resonate. Nobody doesn't do anything like what we're. What we thought it would do.

08:16
Daniel Nestle
And then you know, you as the creator of that campaign or of the, of those words as it were, you know, suddenly you're on the hot seat because they're like, well, you did the research and you created the, you know, the messaging. Why is the response so terrible? Right? It's. And, and you know, I think I want to get into that whole idea of organizational gravity because I think everybody deals with it, you know, but we've been dismissing it almost or just sort of living with it as oh, you know, just, you know, legal, you know, or oh, you know, what can I say? It's compliance and you know, in the parts of it piece by piece, but not as a whole.

09:13
Daniel Nestle
So, you know, that's, that is why this whole, the whole idea of like organizational gravity got me when I read the book. And by the way, everybody, here's the book. Fire in the Loop and post its for those of you who aren't watching. It is replete with post its on many pages, which is clear evidence that I actually read it. But what stuck with me really like from the go, is this organizational gravity concept. You know, thinking about this, the whole thing holistically, it's a force of nature. It is a natural force that just pulls you down and pulls the messaging down and pulls the creative down. And you know, you can't address it department by department. Right. In fact, it's hard to address it as an organization. So what you've done is really remarkable, I think.

10:06
Matt Wilkinson
Right.

10:07
Daniel Nestle
So can we talk about that for a minute? How do, like, what is first of all, what is the solve to organizational gravity? And let's back into it from there.

10:20
Matt Wilkinson
So I think there's two parts to solving it. One is that yet we have to recognize that there's a problem, but being able to name a problem is the first part of that issue. So, you know, and it's something I've seen time and time again. And you know, a great example, I was doing some work with a client, great strategy. We've done loads of research. We just can't get it through the system. There isn't the bandwidth in the teams to deliver what we really need to do. Yes, we know this is where we need to go, but we can't mobilize around what the customer really needs us to do because the organization's getting in the way.

10:58
Matt Wilkinson
The solve that I think I have particularly for the messaging part, is let's try and build something that captures all of that voice of customer, everything, as much of what we know about our customer in one place, and then turn that into representations of our customers and then be able to use that customer or those synthetic customers along the journey. So these are built not just of asking, you know, chatgpt or Claude to, you know, act as something. This is now, this is a knowledge base, but we're turning that into a way of that behaves and acts a little bit like our customers and then being able to use that to capture the words they use, what actually is going to be important to them.

11:43
Matt Wilkinson
And then, you know, I will use that with, you know, team that up with some of those fantastic prompts that everybody shows things like the Ogilvy website page review or, you know, look, understanding things about, how do I get, how do I go viral on LinkedIn, for example? Build those knowledge bases together as well and go, well, if I'm looking to go viral with this kind of Persona, how do I target that? How do I do that? Based on what we know about the customer, based on what we know that's important to them, how do I do something that's going to hopefully resonate with them? And so we're able to use the language that they use and then bring that through the process.

12:20
Matt Wilkinson
So whether that's from inception of a campaign or maybe even a product all the way through to launch, and at every stage, if the customer goes alongside the brief and the campaign, we would say, well, if I'm going to, you know, if legal's going to make this change, what impact does that have? So there's always going to be a need to make a change because legal is going to say, you know, we're going to get in trouble if we say it this way. But let's have a look at what's the impact. Can we work with it? And, you know, often, you know, people that are of higher seniority than the person that's in charge of the communication, the campaign, they may be a little bit distant from the research. They're a little bit distant from the, you know, the customer, perhaps.

13:05
Matt Wilkinson
And, you know, maybe they, you know, maybe they still have an opinion that. No, I want to see it this way because that's what I like. It's. This is important to me, but maybe that isn't important to the customer. Or if you're going to talk about, you know, I think the Heritage line is one of my favorites.

13:20
Daniel Nestle
Yeah.

13:20
Matt Wilkinson
You know, if you're looking at a company saying, we've been in business for 20 years, well, does the customer really care that you've been in business 20 years? Yes, they want a reliable partner to work with, but actually they probably want to know, what have you actually done for the industry? What have you Changed. What problems have you solved? So there's a way that we need to articulate these things that actually will allow us to resonate with customers and make a point about longevity without saying, hey, we're just really old. Yeah.

13:52
Daniel Nestle
I mean, so, okay, you mentioned it and you talked about it and kind of in the context of what you were just saying, but I think we. You sort of buried the lead a little bit there because, you know, the solve, of course, is the book. I mean, I think everybody needs to get the book, the buyer in the loop, in case anybody hasn't seen it yet. You know, I loved it. I really loved it. But look it, what you've put together here is an actionable, practical, accessible, and I might add, digestible because it's. You sit down, two hours, you got the book done. Guide for what you just mentioned, like in passing, Synthetic Personas. Right. It's. The book isn't like how to Build a Synthetic Persona. The book is about why we are missing this or why we let these.

14:42
Daniel Nestle
What we, why we let what we know the customer needs and wants go to the wayside during this process. And as it's churned as the, as our, as I keep saying the word messaging, and when I say messaging, I'm not just referring to it in the comm sense of, you know, the talking points and the FAQs. I'm talking about your brand marketing. I'm talking about your advertising. I'm talking about everything that you put in front of customers, customer service, customer, you know, customer communications, everything. Right? We let that messaging get picked on and tweaked and prodded and moved around by every single kind of hand in the company. And it's no surprise that what comes out at the other end looks like it's been beat up and backed down by every hand in the company because it's a flat mold of jelly.

15:34
Daniel Nestle
So there's nothing there. There's something there, but there's really nothing there. Right. And the, the key there is that, is that how do we get something in that process to give weight and evidence to our argument that, no, we have to say it this way or we should say it more like this, or we should focus on this particular aspect because that's what the customers want. If you don't have something in the process at every step, then you have to be extremely, I don't know, persuasive, powerful. You're going to deal with every single ego in the mix. And there's a lot of them sometimes. But on the other hand, if you can say, look, we have this evidence based thing here that we're taking along the ride and it's gonna help us. And that thing is the synthetic buyer Persona.

16:41
Daniel Nestle
So, you know, I realize that I'm just backing up everything you said and just saying it in sort of a, you know, different way and almost summarizing the book for people, but that synthetic buyer Persona, right, I wanna focus in on that. And, and I wanna tell our comms folks here that, you know, when we say buyer, I, I think that's completely interchangeable with media Persona, your synthetic, you know, audience Persona. It's, it's stakeholder Persona. It's everything like all that. Right? So I want to get to this a bit and pardon me again, I'm going to talk. I'm sorry folks, because like, there's a lot to get out here. There's a synthetic buyer Persona.

17:27
Daniel Nestle
It is of course, you know, heavily based within AI and you mentioned while you were talking about ChatGPT and people, you can, you know, you just go to ChatGPT and you can ask what would a, you know, what should I write? How should I write this or tweak this so that a, you know, demographically kind of this way person living in Detroit or Chicago or Tuscaloosa, you know, how should I tweak this so that it would resonate with that person? And that's, you know, that's what I think many people think a Persona is that you just tell the chat GPT to assume the personality of this demo, this demographic in this location, you know, who is a customer of such and such. But you lay out the reason why you shouldn't do that.

18:17
Daniel Nestle
So I want to just, I want to, let's roll that all up and I want you to kind of talk a little bit about synthetic buyer Personas. And you know, what is so special about the way that you've kind of put them together and why you can't just. Or why you shouldn't. I shouldn't say you can't, but why you shouldn't just go into your favorite LLM and say, act like a buyer of double mint gum and tell me what Wrigley's company should, you know, or Mars now I think should should say in their next campaign, like advise us a little bit here.

18:56
Matt Wilkinson
Doctor. Sure. So I like to think of these as basically existing on a spectrum. On one hand, we've Got what happened back in, you know, pre AI, where people, you know, very often marketers would go away and create a one page overview of who our customer is. Oh, we're talking to, you know, Claire or, you know, Susan. And there's a few things about the social media platforms that she, you know, they exist on and that, you know, and that does a pretty good job of at least defining something about who we're targeting. But it doesn't really help us put on, you know, put ourselves into their shoes. Then when we get to sort of thinking about, well, AI, well can we use any of AI with that sort of approach? And the answer is yes.

19:45
Matt Wilkinson
But again, that just helps us to define a little bit better than what you said about act as. You're then just saying act as somebody that does this. And here's a few more of their attributes. What AI is really good at doing is pattern matching and looking at lots and lots of data. And so I like to think of these things. On one hand, you've got the sort of the flat files that are difficult and people tend to put them in a drawer, not use them. You know, so the old, you know, PDFs, you know, they might be pretty, they might be interactive. But yeah, once we've read them, we think we know them and we never look at them again.

20:26
Matt Wilkinson
And the number of times that I've worked with companies that they go, oh, we never finished the project, here they are, they're old, you know, whatever it is, they're out of date.

20:35
Daniel Nestle
Sorry, can I interject there a little bit? Because you, not, not to, not because what you're saying is off. Just, you just reminded me of. Do you remember the kind of like when it was like de rigor to be like, we're gonna do a Persona exercise when Personas got really big and you'd get in a room, like get in a meeting room, you have like, you know, all the flipboards or the, you know, the sticky paper on the walls. And you know, there was this whole thing about drawing the Persona and like, you know, you get, dig, dig deep into it.

21:12
Daniel Nestle
And at the end of the day, however many hours you're paying the facilitator to be there, you have this really thorough representation of your buyer Personas according to you, not according to the buyers, by the way, but you have these and you've got pictures and you've got these like this beautiful art and they're all on, you know, what's the size of those things, like a 0. 1. A 0, a 1. And for us Westerners out here, it's like those giant pieces of oak tag or like the big. The big flipboard pieces. And like, you know, like somebody in the meeting is. It's their job to take them down off the wall and stack them up and say, I'm going to transfer these and don't worry, we're going to get them. And then you never, ever see them again.

21:54
Daniel Nestle
Like, not one time would you see them again, you know?

21:58
Matt Wilkinson
Yeah, you're the facilitator, you've worked with, takes that information away. Charges you a lot of money to turn that into something very pretty. Yeah, you look at it once and then it rusts away on a hard drive.

22:10
Daniel Nestle
Yeah, exactly. Yeah. Anyway, that memory just came to me. I had to push that because I think that's what. Because I think that's what people think about Personas. Like, oh, it's that thing that we did. And I remember it looks like kind of like a profile and, oh, where is that? You know, so.

22:24
Matt Wilkinson
So you were saying, and. And there are. There are organizations out there that do really detailed Persona and really good customer research to get into this. The problem with those are that it's really hard for a human being to actually hold all of that information in and then say, okay, so I'm going to try and make myself look like them and think like them for a while. You know, you might manage it fleetingly, but it's really hard to do that. What AI allows us to do is to take all of that customary research, all of those interviews, all of that insight from the CRM from, you know, and from what we're told, as well as what our customers are actually saying online as well. Let's not forget that AI is brilliant at social listening and collating all that information as well.

23:09
Matt Wilkinson
So we can bring all of that information together and in an unbiased way. Once we know the rough, you know, who we target, we're actually then able to build archetypes, if you will, of our customers based on data. So it takes us out of the loop a little bit and actually goes well, actually. How would you thematically group these? What are the key things that the AI finds that creates the similarities and differences between them? And then once you've done that, you can then create these representations of synthetic customers that has information not just about who they are, but maybe who they've been. So maybe look at what's their journey over the last 10, 20 years. If we've got information going back that far. What are they saying out there in the market? What do they think of the brands in our space?

24:04
Matt Wilkinson
What are the key things that are cropping up in conversation online? What are the words that they use? What are they saying? And so I like to think of this as going from being the old style, kind of static pages on the. Yeah, it solves a task, it creates a profile all the way through to something that is what I call a grounded synthetic customer, where it's built on everything that you know about a customer and more. And I think that's really when we start talking and sort of engaging with them.

24:37
Matt Wilkinson
You know, we can layer in things like personality, insights and a whole range of stuff that actually goes, well, you know, what if we ask this person, you know, if we ask this representation of our customer a question, they're going to actually give us answer that is actually going to help us look at the world through a number of different lenses. And I think that's what's really important because, you know, within any Persona there's, you know, there are representation to help us understand the world.

25:04
Matt Wilkinson
But if we can look at the world through the lenses of these synthetic customers and get them to then tell us what they think rather than us trying to, you know, put on their glasses and then see the world through whatever len that they put on, it just makes life so much easier and they're so much easier to share. And it removes that terrible sort of challenge that we have as human beings, which is whenever things get hard and we're busy, we don't do it yet. Those are the times when we need that help the better most. And I think that's really where these come into their own. Because all of a sudden you can say, hey, I need to do X for yeah. How would these things.

25:45
Daniel Nestle
Yeah, yeah.

25:46
Matt Wilkinson
And all of a sudden it gives you that, that lens, all those different lenses of perspective to go, well, you know, if I'm going to create some communication, some key talking points, let's make sure we do it this way. And this is why. Yeah, and I think it's also the why that's really important, because then it can say, well, this is backed up from these things. Let's just say you've been out and you've done 30 qualitative interviews because of a certain piece of messaging that you're creating to go through. Every time somebody comes back to you with an objection about your messaging, you're not going to be able to do it.

26:23
Matt Wilkinson
But if you go to your synthetic customer that's built off of those 30 qualitative interviews or a series of synthetic customers that are built from those 30 qualitative interviews, all of a sudden you've got the evidence that says, hey, actually, if you're going to talk about, if you really want to talk about that, you need to spin it this way or you need to talk about this, you know, from this angle. Because this is what's important to them about heritage and legacy.

26:46
Daniel Nestle
Yeah.

26:47
Matt Wilkinson
They don't care about how long you've been in business. They care on the changes that you've made over that time. Yeah, yeah.

26:54
Daniel Nestle
And, and. Okay. Okay. There's. We're getting it now. I think we're. I think it's. It's. It's becoming clearer. And I, I do love the idea of having this. I don't know, I would like to like having a database wrapped up in a personality, you know, that you can, you can question and talk to and along the way. Help me understand a few things. Okay? Help us all understand a few things. There was one word you used, and you used it once. I think that, I think is critical here to this whole process, and it has to do with, why should I trust what these things say? Right?

27:40
Daniel Nestle
So as you go through with your, I don't know, approvals, process, review process, whatever number of processes you have to go through in your organization, or even if it's just when you're talking to one person who has strong opinions or just wants to really take a red pen to everything you do, and you start to lean into your synthetic buyer Persona or your synthetic stakeholder Persona or whatever you want to label it, and you bring that evidence or you bring that perspective, additional perspective to whatever conversations you're having, use it to help you push back. And if pushback is indeed the right word. But you have this. Okay, the. Maybe one of the natural questions, I think that that other person or whoever you're talking to might say is, why are you talking to a robot? And why should I.

28:34
Daniel Nestle
Why should I believe you? Right? And you said one word. You said one word that I think is pretty important. You said the word grounded. And I think. I think grounded is the key here to why we should believe this stuff. So can you talk about what it means to be grounded? And, and, you know, and again, it's. It's not just about the buyer Personas here, but it is about that. It is about why they're. How to make that. Make them, you know, evidentiary and believable.

29:08
Matt Wilkinson
Yeah. So it's grounded in the evidence that you have about your customers. It is. And there's two elements to that. It's not just grounding them, so grounding them in the evidence that you've collected, but it's also them being able to show that any perspective that they come up with, any, you know, any pushback that they give, you know, somebody in marketing, comms, whoever it is in the business, you can actually trace that back to the original evidence that you collected. And I think that's one of the really important things to do here. You know, Everybody talks about AI hallucination at some point, and we know that AIs hallucinate, and we know that hallucination rates have gone down. And there are ways to work around that. And, you know, we can, you know, we can work around the hallucinations.

29:53
Matt Wilkinson
And Frank, these days, the AI tends to be, you know, tends to be at least as reliable as, you know, a human, if not more so most of the time, because, you know, they've got pretty big memories these days. But I think that the thing about the grounding is it's twofold is one, that it is grounded in evidence, but secondly, it's grounded in your evidence. And that's where the real moat comes in. Most companies say that they understand their customers, they talk to their customers, they know their customers, they know who they're working with. And any company that's got a legacy of working with those customers likes to think that they know their customers better than anybody else.

30:33
Matt Wilkinson
And yet they don't leverage that understanding in the way that everybody can get behind in a way that is really easy for departments to agree on who they are, who they're speaking to, who is important there. So I think that's one of the things that I find really important about that grounding is that it's. You're building an evidence based. You're then turning that evidence based into something that you can actually engage with. And not only engage with, but can then use it to trace any, you know, any. Any suggestions, any, you know, anything that's created back to that, that evidence base. Rather than saying, oh, it's a hallucination, it is the next most likely thing that an AI would say. Well, that just creates a sea of sameness, a sea of beige, if you will.

31:22
Matt Wilkinson
Whereas what we want to do is we want to. We want to, you know, we want to paint on canvases that attract our customers and Customers love what they love. And so we have to paint in their color schemes. And so if we're going to be doing that, we have to use their paintbrushes and their, you know, and their tins of paint. And I think that's really where I. I get so excited about this because it gives and it gives, you know, and I'm gonna use the word marketing here, but it gives marketers the ability to start pointing at evidence rather than their gut feelings, because gut feelings are really hard to. To defend. Yeah. You know, I think so often marketing goes, oh, you know, creative. This is why. And you go, this is just. This is better creative.

32:03
Matt Wilkinson
And you go, well, yeah, but why? And it's really hard to justify things. It just is. It just feels right. And on the other hand, a synthetic customer can turn around and go, here's the reasons why this works better for our customers. And I did it for the cover of the book. Yeah. And I did not think. I did not realize that it would work as well as it did.

32:26
Daniel Nestle
Yeah. What did you do? Just walk us through that a little bit.

32:30
Matt Wilkinson
So I have my own synthetic customers that I've created based off of conversation. I had research that I've done, you know, CRM data, you know, things that I've collected. And what I looked at for the book was, okay, I had three. Three covers that are, you know, our friend Jim McLeod had created for me. Genius, phenomenal. And he created these front covers for me. And, you know, were a bit unsure as to which way to go. And I took the covers to the synthetic panel, but I also took them to LinkedIn. And that was such a fun experiment to do. So I had, you know, I had four, you know, synthetic customers. So they were, you know, three covers. So there was always going to be a winner. And there was, but they kind of split.

33:20
Matt Wilkinson
And, you know, actually what was interesting was that the marketers split between two. It was interesting to observe. But then when we took it to LinkedIn, there was also no decisive winner. And even when. Even though you could kind of map people to the synthetic customers from the people, the way that people were voting, they generally split along the same lines as those. What we then did was that Jim went away, and Jim, being a genius, started to remove things as much as anything else and came up with a cover. And I went, I love it. And with that, we then took that back to the. Took that back to the organ. To the synthetic panel.

34:04
Daniel Nestle
Yes.

34:05
Matt Wilkinson
And they agreed. Finally, they agreed on something. And it was so brilliant to see the reasons why and all the reasons that they had objections in the past we'd managed to remove or most of them. And you know, then when I didn't take the. I didn't do the same poll. Cause it pretty much. So this is the one. This is the one. It felt right to me. The synthetic customers agreed and it looked at any of the. Most of the objections that people had, you know, had articulated online, you know, fed through as well. So it answered all the, you know, questions. And then, you know, when I then took, you know, spoke to people individually, they agreed. So it was this fantastic experiment that you could do.

34:55
Matt Wilkinson
But I didn't even realize that I could put images in front of these synthetic customers and they would understand that they would be able to understand imagery in the same way. Did they get the same emotive thing that we did? Probably not. But there was certainly stuff in there that they. That got them to vote in a specific way. And that was just such a powerful lesson for me. Yeah, yeah.

35:16
Daniel Nestle
Every, I mean, to that point, every time I open up, Claude, I wonder what the new thing that it can do that I don't know about yet is actually happening. And I have, I probably have like 14 or so YouTube videos right now pinned or flagged to watch because, you know, all these advice from Anthropic and did, you know, you can do this. And by the way, you know, here's some incredible stuff you can now do with very complicated things. It happens all the time. And I'm so far behind on all that. But that being the case, besides that, I love that example of bringing whatever the decision is that you need to make, bringing the options in front of real people and in front of your buyer, your synthetic Personas. And you made the decision.

36:19
Daniel Nestle
Ultimately, it wasn't the Personas that made the decision, I think. And again, I kind of want to continue to pick on this whole word of grounding because it, you know, it's really. It has to do with why you made the decision as well. And it has to do with, you know, what the humans are all about here. When I read the book, I. I saw that word when you started talking about grounded, that when you start talking about the fact that you're. The things that you're doing, the Personas that you're talking about are grounded. And that's why one of the reasons why they're so much different than just somebody who can go over to opening up your chatgpt and saying act as right.

37:01
Daniel Nestle
And that Grounding, it's a really good kind of catch all phrase or word for not only like giving something weight, right? You're literally grounding it. You're giving it the weight it needs to resist that organizational gravity. But it's also the catch all for your expertise that cannot be replicated. So, yeah, you're pouring in evidence, like real evidence that you've pulled from all of your data, from your customer conversations and records. You're pulling in from things that, by the way, other AI doesn't have access to. It's just your stuff. And you're pulling in all this good stuff and then you're injecting your decades of expertise into that. You know, where you're saying, analyze it this way, not that way, where you're saying, you know, this is, I want you to speak in this manner.

37:57
Daniel Nestle
You know, this, these are things that matter to you, and here's the little facts that you may or may not know. And this is the way I want the answers to be given to me and everything. And you know, and then of course, you're adding in any of the frameworks that you have to deal with, especially if you're in a regulated, regulatory, blah, blah, regulated world where you have rules and laws and everything. And I'm not suggesting you create synthetic Personas that will give you compliance guidance. But what I am saying is if that's all in the soup, right, if that's all in the mix, then you have something that's truly grounded versus somebody just going to an LLM and saying act as right.

38:47
Matt Wilkinson
Absolutely right.

38:49
Daniel Nestle
So that pulls through the entire process, right. In your case, With the cover your Personas are grounded in your own understanding of the world and your customer research. And therefore you could trust the result, but you had to kind of add that little bit more. So the grounding is there. And then you have outside expertise that you bring in to kind of enhance that. With Jim, talk about, you talk about, again, a genius Jim doing the designs, but he's responding to you in the first place. It's your brief. So you've grounded Jim in a way in that process. So the importance of this, of grounding, I think can't be underestimated and. Or can't be overestimated. That's the right way to say it.

39:47
Matt Wilkinson
Right?

39:48
Daniel Nestle
And for those of us in comms and everything in communications and marketing, but specifically people who are out there telling stories all the time, I think half the problem we have Is that you're not as grounded or the story isn't as grounded or the work isn't as grounded as it needs to be. So when you encounter the resistance, it can get blown away. Yeah, right. You know, the, the Steven Pressfield, what was it? The, the art. The. The war of art. Right, the Steven Pressfield book where he's talking about the resistance, you know, where you have to sit down and write through it. I think that's. Or I think organizational gravity is the resistance. I think the people out there who say, oh, it's just AI and, you know, there's, you know, I could just do it myself.

40:37
Daniel Nestle
That's resistance, you know, but that is the enemy. The enemy, so to speak. And the grounding is the way to get through it.

40:49
Matt Wilkinson
Yeah, I'd agree. I mean, imagine that you just, you know, you've just hired a new employee. Do you expect them straight away, fresh out the box, to be able to tell your story perfectly?

41:01
Daniel Nestle
No.

41:04
Matt Wilkinson
And as people have gained experience within the business, they go out and they speak to customers. They learn from the customers, they learn from the organization. Often we learn more from within the organization, from outside the organization, which I think is a big problem generally, that we actually tend to be, you know, organizations tend to be more inwardly focused because we spend so much time with each other rather than with our customers. And sales tend to, you know, do spend a lot of time with customers, but they don't often have a good incentive for bringing all of that rich data back in. Yeah, hopefully synthetic customers and the, and grounding them actually gives them a really good reason to do that. Yeah, but imagine then that you're doing the same with a large language model when you just say, act as.

41:50
Matt Wilkinson
Yeah, you're not giving them any of that onboarding. So I like, you know, you can think of this grounding as almost as onboarding for the models that we work with. But then the great thing is that those knowledge bases, those interactive knowledge bases can be shared so easily. So rather than us all having to look through, oh, which one of those 30 interviews that, you know, it's on page five of the 15 of, you know, of the report, whatever it is, we don't have to go searching for the information. We can just ask a question.

42:21
Matt Wilkinson
Yeah, and I think that's where the real power lies here, is that we're doing the upfront work so that we've got a repeatable resource that can be shared across teams and so that we can all align so Whether we're making decisions about products or value propositions, messaging, what is it that the sales team wants. And if we actually start to create mechanisms for capturing insight or capturing information and constantly evolving these things, actually we become far more customer focused than we ever were before. And I think that's the thing that I really believe is going to be the difference between companies that survive or fail as we go forward is those organizations that can keep laser focused on what the customer wants and needs. Particularly because AI is changing that so rapidly in so many ways.

43:13
Matt Wilkinson
Certainly in a digital space, the changes from people saying I'm going to Google it to now they go to an AI to ask the question, they may never go through to the website. The rise of the zero click search, we have to be spending more and more time understanding our customers than ever before. And almost the half life of the research that we do is shrinking as well. So they don't, that insight doesn't last as long. So we have to be going back to the well of asking questions of understanding customers. And if we're going to spend that time and effort to understand our customers, we can't just be creating a report once every year and letting it rust away on a hard drive somewhere. We have to be activating it and turn that into something that teams can widely use.

44:05
Matt Wilkinson
That's a, you know, why would you do, you know, spend five, six, seven figures on a, you know, on customer insight when all you're going to do is turn it into a report, share it, have a couple of meetings about it and then forget it. Yeah. You need to be, we need to be doing more with the information that we have. Yeah.

44:24
Daniel Nestle
Especially now that, you know, time keeps getting compressed and opinions get stale. You said at the shelf life, right. Something you were talking about just actually made me think about, you know, a perceived over reliance maybe on Personas. Once you have, once you have them going and you know, you have them with you, it's like we're kind of anthropomorphizing them as we should be. But like you have them with you along every step of the way in an organ in your process and you're asking them questions, you're saying, you know, okay, look, here's the latest iteration of this and what do you know, what do you think?

45:05
Daniel Nestle
Or here's the latest iteration of this and you know, tell me where it falls flat or you know, and you want, you know, the book in the book it, you're very clear about the kinds of questions and the strengths of the Personas, you know, but as you're going, as you're working through these, I, we have to make it clear that we're not suggesting that you delegate and believe everything that's coming out of the Persona's virtual mouth. That there is, you know, there's still a very large role for us in this process and that the humans have. Cannot delegate the decision and the discernment. Right. So as you work, as you use these, if you build them and you use them, you build them while you use them, what's the limit? Like, what are the limits?

46:03
Daniel Nestle
Where do they break and where should you be clearly focused on? No, no, this is where we're going to actually talk to the real person.

46:15
Matt Wilkinson
I think there's two places there. One is time, temporal drift, if you will. So have things changed? That's easy to understand because the world's changing and the world's changing fast. So things change and they can change overnight. We saw that with cover we've seen that with AI. Things that we would have expected to have been true for many years and now may be true for a month, if not shorter. The other side of things is when we talk about grounding, you know, if we go out and do voice of customer research about a specific topic, we've actually only asked a certain amount of questions. So there are gaps in our knowledge still. So we. That the.

47:02
Matt Wilkinson
The validity of the answers that you get from a synthetic customer really relates to how closely the questions that you're asking it relate to the information that you've already provided it. Otherwise you're getting back into that generic prediction of the most likely next word. So you have to be very careful to look at, well, actually, what was the insights that we've already gained? What is it? It's one of the reasons why so many of the studies about sort of looking at synthetic customers kind of struggle when it comes to pricing.

47:38
Daniel Nestle
Yeah.

47:39
Matt Wilkinson
And they're not very good at pricing. And there's a couple of reasons around pricing. One is because AI is being very logical. Tend to look at features, benefits and ascribe monetary value to that. And so you would look at, well, if something it does A, B and C and another option does A, B and C. Well, it's going to be the same value. But if I go out and there's a. Let's just say I've got a laptop from a Dell or an Acer or whoever it is versus a laptop from Apple, if I like the Apple brand. That's going to carry weight. And so the AI doesn't necessarily understand the value of the brand to a human, the emotion involved in a decision. People spend hundreds of dollars on buying a Montblanc pen, for example.

48:33
Matt Wilkinson
It performs the same service as a Bic Biro that maybe costs a dollar. Does it, does it do anything differently? It allows you to create a line on a piece of paper. The difference is how it makes you feel and how other people might perceive you using it. So same with cars, right? You can buy a car that will get you from A to B. And yet so many of us want to buy a car from specific brands because there's something about the brand that resonates with us. There's something about the way that it makes us feel. It's faster, it's noisier, whatever it is, or it's more comfortable, whatever it is that attracts you to a specific car, that is something that is uniquely human. And we have to be very conscious that this grounding is fantastic.

49:27
Matt Wilkinson
It's really good at being able to capture what customers have told us, what we've learned. But there is also. And they're really good at being able to sort of fill in the gaps between when you would normally talk to customers. But I would argue that actually, hopefully what it does is it gives us more incentive than ever before to go back and speak to customers again and again and spend more time understanding customers, because that is the only way that we're really going to be able to differentiate ourselves when the value of production, particularly if anything that's digital, approaches zero because of AI. So whether it's code, whether it's words on a page, the ability to create something is negligible. Now in many respects, other than manufacturing things, there is craft, there is artisanship, there is the emotion. That's what stays.

50:22
Matt Wilkinson
And so we have to be careful that we're taking the lessons that we know about our customers. We're making the most of that information, but we're still going back and we're still laser focused on making sure that we're making our customers feel right. But we can find out what our customers are upset about by using online, by finding they're doing social listening, by looking at what customers are saying through support requests, through all of that information that we're collecting, we're then able to turn that into something that allows us to then do things differently and hopefully do things better.

50:59
Daniel Nestle
Yeah, yeah, that. You know, oftentimes people who are using Personas, they just like just your standard PDF in the drawer, Persona or whatever you want to call it. Like, you can't. You can't really pull the emotions off the page, or if there's emotions kind of coded into that PDF, it's more like there'll be a line that says fears, desires, you know, things like this.

51:22
Matt Wilkinson
Right.

51:24
Daniel Nestle
But. But it's very hard to kind of. It's almost impossible, as a matter of fact, to. To emulate that from the customers, from. From the customer standpoint or from the stakeholder standpoint. And, you know, the.

51:39
Matt Wilkinson
The.

51:40
Daniel Nestle
The Personas that you're designing or that you've kind of laid out how to build a little bit in the book can do that. Now, that part, that emotional part is I think, one of those things where maybe our listeners are gonna be like, you know, we're talking about creating little emotional robots that we're asking questions to. What do you mean? I think that might be concern, but it's that understanding how something might respond in an emotional. That is uniquely human, you know. But can that be baked into the Personas and. And can you trust the Personas when they respond emotionally?

52:28
Matt Wilkinson
So, to an extent, yes. There are. There's two. There's two levels to this one. We can bake in personality, insights into them, so we can find stuff out. We can also look at. If we understand how they've reacted in the past, we can again use inference to infer how they might respond in the future. So there are definite levels of that emotion that we can start trying to build in. But there are absolutely challenges. Can we. The answer is, can we get some insight? Absolutely. But if we're going to go off and make a big decision about something, what we need to do is to then go and test it first. Get that. To create the shortlist and then go to customers. So we absolutely can make things that we try to mimic the customers with.

53:18
Matt Wilkinson
But at the end of the day, the humans are human beings, and we're precocious. We have to test. What do humans say? The challenge, of course, is that if you go to a. There are a lot of criticisms about the way that focus groups gain information. All of these things. Oh, yeah, we have to be very careful about, you know, what data I mean. Coca Cola. Great example. What was it? 190,000 Taste points when they were. When they created New Coke, you know, they tested everybody and they said, which of these do they prefer? Is it Old Coke, New Coke or Pepsi? They created New Coke. They released it to the world. And then they had to backtrack within months because what people expected when they got that red can was something different than what customers were expecting.

54:06
Matt Wilkinson
So there was a difference between what customers actually preferred versus what they expected when they opened a can or something. Yeah. So we have to be really careful that just because we might ask questions, are we asking the right questions. So even, you know, human research fails if we ask the wrong questions. Yeah.

54:25
Daniel Nestle
I love how you brought in, like, baking in social listening and sort of these, you know. Yeah. You're looking at how people communicate, how people really communicate. And of course, you can have transcripts and you can have like, audio and. But AI isn't going to really pick up on the vibe necessarily of a person's conversation strictly from a transcript. But when you put it all together, transcripts, social, Reddit, pulling in Reddit threads. Right. Talking about, like, figure out how people talk. All brilliant. Now, look, Matt, there's so much more I want to talk about in this book you've talked about. I think we've covered the bases, but life must go on. People have things to do. And I know that you have things to do as well.

55:12
Daniel Nestle
So what all I want to wrap up here basically by saying to our listeners that, number one, you need to get buyer in the loop. It is, it is a fast read, but not a light read. It is a rich read. And Matt walks through in here how you, what you need to do to build that Persona. It's not like go to ChatGPT and do this, but what are the elements of the buyer Persona that will make it effective or not? And I highly suggest that we do this. It doesn't matter if you're an organization with a hundred thousand people or three. Every organization has gravity. And you'd be what you'd be. I think you'd be doing yourself and your company a favor by having that voice to kind of keep, you know, to keep to.

56:07
Daniel Nestle
To keep to having that customer voice next to you so that you can keep asking it along the way to make sure that you're staying on track. And the Byron Loop is a great way to get you going on that. So, Matt, you know, before we go, one last question for you. In the book you have, you say, hey, you give at the end, like, if you're in sales, if you're in marketing, if you're in product development, if you're in something, you give like a kind of quick rundown of what you should do. The one Role that's not in that list is if you are in PR or in comms, you know. So the whole book speaks to communicators, I think. But if there's a specific like 30 second thing that you. That communicators. How communicators can start with this. What is that?

57:01
Daniel Nestle
What would that be?

57:02
Matt Wilkinson
Yeah, and I probably have conflated marketing and communications one because as we do because I've, you know, I've worked in marcoms for a long time. The other is that most of most many of the organizations I've worked in haven't made that big a good a distinction between the two. So yes, you've got brand, but you know it, you know comms, you know, you don't have to often have too many PR specialists within it. It tends to be an agency in the organizations I've worked with for sure. And in. They tend to have more agencies that tend to take on those sort of roles of being specialist communicators. Because they have the black books. Yeah. Which is still incredibly important. Sure. But I would say look at who are your target audiences? Think of your buyers as your target audiences.

57:48
Matt Wilkinson
Who do your stories have to resonate with? What's important to them. So if it's investor relations, one hand we know that the numbers are important. They basically investor relations are looking at trying to. Investors are trying to understand what's the upside, what's the risk. Whereas me, in terms of pr, some of those other stories we may actually be looking at what's the human impact we've got, what's the environmental impact, what's the social impact of decision making. And we have to be able to understand when we start thinking about how we're communicating with the world, we have to start looking at actually from those perspectives. What do all those groups say?

58:27
Matt Wilkinson
So in some ways having a synthetic customer group that can look at not just who our buyers are, but who are our investors, who are the lobby groups that are going to be fighting against us by being able to build those in there as well. Maybe, just maybe we can start thinking about how do we head off some of those objections to our messaging and give ourselves a little bit less trouble in the market by actually being able to talk and head off some of those arguments at the past.

59:00
Daniel Nestle
Very well said. You've rescued yourself in my mind. But Matt's fundamentally right where he says things about marketers and this and content creators and et cetera in the book. You know, just swap out what he's saying for comms. It's the same thing. This is extremely powerful and very effective. And I think we should all consider how we might want to implement some variation of synthetic voices, synthetic customers, stakeholders in our comms, NPR and marketing programs. Matt, this has been amazing. You can everybody out there. You can find matt@striven.com that's S T R I V E double n.com or Dr. Matt Wilkinson, but not it's Dr. Matt Wilkinson because there's a lot of Matt Wilkinson's out there and also, you know, on LinkedIn, as it were. Please go look for Matt. Please go talk to him.

01:00:06
Daniel Nestle
And you know, if you need to know anything else about Personas, read this great book, the Buyer in the Loop. How Synthetic Customers stop your buying. Sorry, How Synthetic customers stop your buyer Disappearing between Research and market. That is absolutely worth your time. Matt, any last words?

01:00:29
Matt Wilkinson
I just want to say thank you to everybody and I really hope you enjoy my little opus and I throw that in there just for anybody. That's a bit of a Claude fan. A Claude fan.

01:00:40
Daniel Nestle
Thank you so much, Matt. Thanks so much. And we'll have you back again sometime soon.

01:00:44
Matt Wilkinson
Thanks so much. Dan Speak soon. Foreign.

01:00:52
Daniel Nestle
Thanks for taking the time to listen in on today's conversation. If you enjoyed it, please be sure to subscribe through the podcast player of your choice. Share with your friends and colleagues and leave me a review. Five stars would be preferred. It's your call. Have ideas for future guests Want to be on the show? Let me know at Dan Trending Communicator thanks again for listening to the trending Communicator.