Mark Ryan (Dark Matter Consulting) on AI, Scale, and the Future of the Insights Industry

In this CEO Series “futurist episode,” Lenny Murphy sits down with Marc Ryan, CEO of Dark Matter Consulting, to map the industry’s next bend in the road. From his start as a receptionist in Toronto to building internet measurement in Silicon Valley and leading research and product roles across major players, Ryan has watched disruption arrive in waves and he argues this one is different.

Together, they explore how AI is accelerating decision-making, why “scale” keeps breaking pieces off the insights iceberg (UX, CX, analytics), and what it will take for established firms to evolve into a new model: expertise up front, automation in the engine room, and relevance far beyond the insights department.

 

Transcript

Leonard Murphy: Hello everybody. It's Lenny Murphy with another edition of the CEO series and I'm glad to have you taking time out of your day to spend with myself and my guest. Uh this is one I've really been looking forward to. Um so my guest today is Mark Ryan who is the CEO of Dark Matter Consulting and also who's been around the industry for a very long time and he'll explain that. But uh I just before I even get there, uh Mark and I have been going back and forth on a variety of topics for a while and we're just going to have fun. Um this is the futurist episode. Uh we're going to talk about all types of stuff um that I think will be interesting. And after Mark introduces himself, uh I think you'll understand exactly why we're going to go in that direction. So Mark, now with that uh that set up for uh everybody, tell us your origin story and then we'll dive into the fun stuff.

Marc Ryan: Sure. Um, so my origin story goes back to Canada. I'm Canadian. Um, and I, um, did a survey in college for the National Valley of Canada and was looking for a summer job and ended up applying to a small consultancy in Toronto called Miller Brown, which was a recent acquisition in the heyday of acquisitions that they were doing and it was a receptionist job. So you know I always take pride in the fact that my first job in the industry was as a receptionist picking up the phone answering the phone and as you know was the case at the time they would uh make the everybody you know on the team had to do something and so I would be doing assetates slides things like that and you know I kind of knew my way around computers and that kind of made me stand out in that regard that you know moved into part-time work after after the uh initial summer job and then I ended up taking a full-time role there. Um, one of the things I really liked about that job and I think one of the things that set me up in sort of the way I approached my career was the guy I worked for was a behavioral psychologist. And so he loved to do everything. Like he was the biggest experimentter on the planet. We were testing, you know, this in the 90s. We were testing eyetracking software, all kinds of really wacky stuff. And he let me go get a modem, my first uh connection to the internet at work, 2,800 baud modem, this big wire running over my cubicle to my desk. Um and that modem basically ended up sort of you know vaulting me into the internet uh in a big way because uh at the time in the 1996 the internet had sort of started coming up and we had launched uh middle brown interactive a online measurement firm to measure the effect of online advertising. Uh it was in Silicon Valley actually was in the city of San Francisco south of market and we were doing a online advertising measurement and it was this thing it was in the distance. We had acquired a team out of Hotwired which had rolled out the first ad online like a 50% click rate back in those days and we were testing these things and what happened was um the team that was working at Hotwire decided uh from the Hotwired team most of the people who were actually doing the day-to-day research just up and left they wanted to go do their own startup it was you know heyday of the internet
 
Marc Ryan: and I got a call uh from the guy who was running the division at the time and he said hey you're the only one I've really met that has a modem that knows anything about the internet. Do you want to come down to San Francisco and work on this internet thing? Which is what I did. And that sort of cascaded me into, you know, the world of digital, the world of the internet. We were testing all of the, you know, the online ad formats as they came out, the first video ads. We tested, we tested wacky things like cursors that would change into Ronald McDonald fingers and things like that to help sort of drive brand uh online. And it was really a fun time that bolted me into a variety of startups. I ended up working and moving from what you would consider more of like the marketing research side into the media research side and I worked for a number of startups which uh went through a variety of different hands in terms of uh ownership but in essence Jupiter media metrics which was you know the big 800 pound gorilla of online media measurement uh in the late 90s the internet blew up in 2000s that entity got broken up and as part of that breakup I ended up at a company called Net Ratings Net Ratings was the online measurement leader of the time at the day.
 
Marc Ryan: Uh, and I worked there for a number of years and that's really what kind of get it started getting me into product development from more of like the analytics research analysis side of the business and that was a lot of fun because you know we got to do really cool things like try to invent new forms of research new new ways of approaching the market. Uh, that resulted in a acquisition by Neielson and I am in the Neielson company. At the time, Neielson was owned by private equity, so it was interesting days. We had a lot of fun building new technology, working on measurement approaches and methodologies. Um, and then a friend of mine called up and said, "Hey, I want you to come to this startup." And I went to a private equitybacked uh startup that was in the marketing research space called uh Insight Express. I worked there for a number of years as the chief research officer and migrated into a co-CEO role and where we successfully exited the business to Canar. So full circle from receptionist at the front desk working brown all the way back to you know selling to uh Canar and that was um that was you know that was a while ago and I spent nine years
 
Leonard Murphy: I'm proud to cancel.

Marc Ryan: At Canar in a variety of different roles. I worked on helping build the media you know measurement the brand measurement side of the business. I worked in the panel side. I relaunched the panel that they have there, the canth profiles panel. um I went to a chief data officer role and then um after that I left and went to an enterprise software startup and I would say that's probably one of the most interesting things I did you know in the last 10 years or so done a lot of marketing research I know insights but then doing decision science at a relatively high-tech um startup that was running like pabytes worth of data real time for customers was a real difference maker and I learned a lot there uh took that into a role at YouGov worked there in a in a product role and since then I've been uh doing my own thing. I have my own consultancy Dark Matter Solutions and that's what I do today. I work with a variety of companies and startups and a lot of the the uh types of firms Lenny I know that you also take a lot of interest in are the ones I tend to bump into day in day out.
 
Leonard Murphy: Uh that's great you brought up uh talking about the early days. Uh I remember going on CompuServe in 1997. Um yeah, in 97

Marc Ryan: Yeah. There was a time when um you know WPP who owned Canar you know uh Miller Brown wanted to make a real big play on the internet. They knew that this was going to be important. Like, you know, in the 90s when you wanted to do internet advertising, there was a guy down the hall, you know, who had his, you know, hair in a ponytail and, you know, sit there in a dark room with a hoodie on. And, uh, that was who you went to. And they wanted to try to mainstream that a little bit. And so Eric Salama who ended up being the CEO of Caner but previously to that was uh in you know a chief strategy officer role at WP came to Silicon Valley and we all got piled in a car or people there from you know the different agencies as well went down to visit excite at home uh and for those who don't know what excite home was that was one of the largest you know internet providers uh

Leonard Murphy: Yep.

Marc Ryan: You know on the planet at the time and they had their own website and portal and what have you and had you know the big you know NASA room with all the screens on the all showing internet connection

Leonard Murphy: Yep.

Marc Ryan: Speeds and what have you. Back in the heyday. Yeah.

Leonard Murphy: Yeah. Well, fun times and actually, you know, so that's interesting. There there there's our framing uh uh system for this conversation, right? Because you you obviously have been in the thick of things uh from I was just a lowly user, right? You you were uh kind of pioneering what the uh what the digital world um would look like or at least trying to understand it and measure it uh from the get-go.

Marc Ryan: Mhm.

Leonard Murphy: And yeah, over the since you you left you got and hung out your own shingle. Um I have I subscribed to your Substack. Um the uh which by the way folks if you don't uh we we'll make sure to kind of put a link in there, but you should subscribe to Mark's Substack or follow him on LinkedIn uh because he posts there as well. But I I've loved the level of thinking that you've been putting out there about tell me whether you agree with this.

Marc Ryan: Sure.

Leonard Murphy: Here's my my take with when thinking about this is you've come out and said the world's changing period.

Marc Ryan: Go for it.

Leonard Murphy: Not even changing, it has changed. It has fundamentally shifted already. So um uh and we have a pretty clear view of trajectory over this the short period of time.

Marc Ryan: Yeah.

Leonard Murphy: Although the longer we go out, the less clear that is. But the short period trajectory is driven by a a massive transformation at all levels of this industry. Driven by the value of data, but how we use it, uh how we acquire it, uh how we process it, the things that we do with the outputs of that have massive implications for the business models uh of the inside space overall. and are utterly disruptive and we can b**** about it and moan about it and say wait but no and it makes no difference whatsoever.
 
Marc Ryan: Yeah.

Leonard Murphy: Um the uh the world has simply it is an accelerating pace of change and I think you tackle those things those topics on a regular basis uh not in any hyperbolic way but simply uh very thoughtful let's break

Marc Ryan: Heat.

Leonard Murphy: It down and and look and think about it through your experience uh of you know kind of riding this digital train for a long time in very large organizations so that is my take um is that did I get it Right. Or are

Marc Ryan: Uh you know absolutely I mean the world I think I think you said it right I mean the world has changed right I just don't think it's come home to roost yet for a a lot of folks and one of the things you know everybody learns over time particularly when you try to affect change is that change does take time right and even when changes happen you really can't recognize the change until you look backwards, right? There's something about, you know, meaningful change that's only visible in a rearview mirror. And I think one of the things is we're living through a very interesting time today where, you know, the new technologies that we're seeing come out there, you know, particularly speaking to, you know, topics like AI, they're they're targeted at data, right? You know, the fundamentally when you look at what they they're part of the knowledge economy, right? What is insights? But you know it's bit but it's the sort of foundational piece of the knowledge economy. You go back 60 70 years when you wanted to understand what was going on with consumers. You went to the the research department that they knew everything. They had access to all the data and over time that department's been fragmented into different you know subsections and different parts and different different uh functionalities within the business. Um and here comes you know something new. It's just going to tackle that and try to, you know, take little pieces off the bone, so to speak, of what the the value is that you get from those parts of the industry.

Leonard Murphy: Yep. Yeah. Yeah, I I I know from over the years of our conversations, I think that you would agree with this that the I started talking about the idea of insights and the data that we collect as just being a uh a spoke on a wheel, you know, feeding into a central hub and uh around the the time or big data, right?
 
Marc Ryan: Yeah.

Leonard Murphy: That was the the term. um not recognizing what was what was missing was a system to synthesize that information, unlock more value. Um that's what AI does, right?

Marc Ryan: Yeah.

Leonard Murphy: So, you know, we are now a feed and we we were moving in that direction for a long time. Um but the the exponential increase I do think it's exponential. I don't think that it it it is moving faster than anything that I have ever observed um uh of adoption of those technologies that don't just create efficiencies but unlock new possibilities and uh potential for value creation across the entire stakeholder chain right from individual users or individual practitioners uh to large organizations um is just mind-boggling

Marc Ryan: Yeah. I I think in general people just have a hard time conceptualizing the rate of change, right? I think the the analogy maybe just works best is everybody most people are familiar with the concept of Moore's law which existed for you know decades which is like every year um well it's actually every 21 months or
 
Leonard Murphy: Yep.

Marc Ryan: so the summer of 2016 I don't know the exact number but just under two years the speed the processing power of computers doubles right so that's why you know we went from you know the Motorola handspring in the early 2000s to iPhones today which are like a million times more powerful than what you would have had just you know 1015 years ago and that Moore's law has been you know true for a long time if you

Leonard Murphy: Yep.

Marc Ryan: Think of like the equivalent in terms of like AI and you know the Moors law for AI the research that they've done points to the number between being somewhere between three and seven months of a doubling in processing

Leonard Murphy: Yep. I was thinking about six months was what was what I was thinking.

Marc Ryan: power yeah exactly and I think that the interesting thing about that is we we know that and once again it's just looking backwards like if you played with any of these AI systems chat GPT when it first came

Leonard Murphy: Yeah.

Marc Ryan: Out like in the late 2002 at least when it started to go viral in late 2022 or in early 2023 the way it work yeah about like it is now you know infinitely better we can see that you
 
Leonard Murphy: Yep. about this time. Uh yeah.

Marc Ryan: Can recognize that in the responses you get and that's two years and so just you know think about it in that context I think the the new thing that people are talking about is is they've stopped talking about like the the processing power or the the amount of effort and amount of of um you know power of these models. They're not talking about like the how long they can work before losing context or failing to be able to keep a task going and that is also doubling every six or seven months.

Leonard Murphy: Yep.

Marc Ryan: So, you know, not only are they getting smarter, but they're also being able to do more longer. And that's that's been the challenge with like Agentic and stuff so far. It's just been that, you know, AI slop or context raw and things like this where models just kind of run out of of memory or just can't, you know, keep a process going long enough without messing things up.

Leonard Murphy: Yep.

Marc Ryan: But that's something that obviously showed up and the the industry's been tackling. I mean, these are industries that have got billions upon billions upon billions of dollars in investment, right?

Leonard Murphy: Yep.

Marc Ryan: And so to expect them to not address these things would be naive.

Leonard Murphy: Yep. Yeah. I Yes. In my own personal experience, I've run into a token limit once. Um and and rightfully so.

Marc Ryan: Yeah.

Leonard Murphy: I it's one really really really long intense you know uh um uh chain on on my current favorite uh perplexity um but that was it right everything and and everything since then there is never been any challenge uh that I've run into from that standpoint of it's actually I see the memory not just memory but I see it adapting uh on an ongoing basis to my preferences. Um uh it anticipates the type of things that I use it for.

Marc Ryan: Yeah.

Leonard Murphy: Um and not in a limiting way but actually in an expansive way. Uh the uh which is just really just really interesting with the point being that you know I it like you I am I am a consultant.
 
Marc Ryan: Yeah. Mhm.

Leonard Murphy: So you know that is how I I feed my family. uh my limiting factor is my time. So that's it. Um so I'm able to produce more that's probably higher quality honestly. Um uh faster that increases my bandwidth and allows me to take on more work and do more things s than I could before. So that is a superpower. It is an amplifier for me. um to uh and that that's unexpected going two a year ago.

Marc Ryan: Yeah, that's funny.

Leonard Murphy: My thought was that it would uh limit or devalue me. So, right uh Yep.

Marc Ryan: That's the one topic I want to I'm going to pick up next at least on my um substack. But it's just the concept of like you know what is do what does all this mean for different cohorts of individuals um you yeah where's the value right you know the this research came out of Stanford that said

Leonard Murphy: Yep. Because where's the value? Isn't that ultra right?
 
Marc Ryan: That like if a job can be completely replaced with AI it's already showing to be that that's happening but new graduates are still being hired right if you're if you have a job that is augmented by AI as a new graduate you can come in those jobs are still there and they're still available they're not declining um and you know the research that they did suggested you know one thing which is that like if you're a a veteran of a particular industry you are well positioned to be you know to deep dive into the AI world because AI is going to augment you kind of like you were just saying like AI as an augmenttor

Leonard Murphy: Yep.

Marc Ryan: For veterans is just a huge win and it's getting easy to getting easier to augment people even if they don't know a lot about AI right it's very easy to go in and ask those questions and you see folks like open AI now going out and hiring people who don't have AI skills but they're just like hey you know a lot about investment banking you know a lot about you know healthcare like you know they want
 
Leonard Murphy: Yep.

Marc Ryan: The veteran talent to come in I think you know from a talent perspective that's great for the veterans because they still can they can bring all that sunk knowledge that they've got in you know an industry or a particular area to the table and AI helps make that better. I think the the interesting the second most valuable cohort is going to be the new graduates and I think this is the other thing that most people assume that that's the job those are the jobs are all going to be eliminated but I think if you know the the newer people they bring something to the table that AI doesn't have yet which is like open minds right you know when you think about um people like you know little kids will solve problems that adults can never solve because they just look at it through the lens they completely you know, unreserved of bias and they can just look at a a problem and think of a different way to solve it. And that's what new employees bring to the tables.
 
Leonard Murphy: Yeah.

Marc Ryan: They think differently.

Leonard Murphy: Yeah.

Marc Ryan: They're not, you know, biased by everything you've been doing for the last 20 years, right? So, when you think about, you know, how a business gets oified into its processes and stuff like that, like bringing in new hires from, you know, the colleges and what have you, they're going to bring a new way of thinking to the table. And I think that's still going to be valuable in the future. I think the the folks ironically that need to figure out like how to sort of solidify their career path the most if you think about jobs and what have you are the people who are in the middle because the people in the middle have had that time to oify it they're like no this is how things are done this is what my customers want you know you know they can certainly augment what they want with AI

Leonard Murphy: Right. Right.

Marc Ryan: But if you're going to say well do I want a Lenny who's been doing this for you know 25 30 years who you know augmented by AI or do I want someone who's been doing it for 10 years augmented by AI it's like well I'll take the more experienced person if I can get or I want a different way of thinking.

Leonard Murphy: Yeah.

Marc Ryan: So I think you know the interestingly I think there's a sort of a difference shaping up in what you would have thought was like you know career stability um that comes out of this. Um, you know, obviously we'll have to see what happens over time, but it is an interesting thing we're seeing where like, you know, open perspectives, new ways of looking at things are valued because the old way of looking at things is already well understood by AI systems and the only thing you can augment it with are what the veterans really know. It's that institutional that skill knowledge that they've developed over decades.

Leonard Murphy: Yeah. Well, at least you and I both uh uh hope that's the case since fall into that category.

Marc Ryan: Yeah. Exactly. Yeah.

Leonard Murphy: But and I I think that you're right that what I've like you I work with a lot of of early stage companies but you know also you know some more established companies and and with the the early stage companies obviously this is intuitive for them. So uh it's natural to uh they're going to utilize technology to uh automate any process that can be automated and create those efficiencies of scale. Um and that is their their advantage speed and agility and uh etc etc from that standpoint. Now the established companies we get into business model challenges right they have a difficult time uh adapting to that change because so much of that middle layer of the business is process and it is you know human-driven process um either process conducting the process facilitating the process or managing the process this is the way we do things um which certainly high quality you know etc etc. No, no question about that. But it doesn't deliver value. It's simply a necessary component to get to the value creation. And that is that is a hard chasm intellectually, emotionally, and business-wise to navigate through that. Uh I vividly remember vividly a few months ago having a conversation with the CEO of one of the largest companies in the world that research companies. How do you communicate to your shareholders, right? Uh that look, we're at X dollar's revenue now. Here's what's going to happen over the course of the next 18 months. That revenue is going to decline because we simply cannot charge as much for these things because clients know we don't have people doing them, right? So the uh so so it's going to come down although our profitability is going to increase dramatically uh because we will we're not paying the people to do those things. Um and there's this huge middle layer of process that is just going to go away. But that is it's difficult.

Marc Ryan: Yeah.

Leonard Murphy: How do you have that conversation with investors with you know internal stakeholders? Hell even your employees. Um, I don't I don't envy anyone who is running a established research company today in trying to navigate those things as a CEO. Uh, Right.

Marc Ryan: Yeah. And I I think that you know it's it's a challenging one. I do think it will take longer than we expect for some of those companies to sort there'll be the transformation there. The other challenge is that you know some of the big research companies they're really good at doing you know face tof face interviews in you know one of the 10 11 different dialects in India in you know in you know in rural areas of the country or in Indonesia like there's some markets where you're still operating in a very manual hard to find group of populations and so that will stick around but to your point 

Leonard Murphy: Sure.

Marc Ryan: Think there's everybody's going to be looking for ways to like a gentic ally rebuild their backends. I think one of the things that's a little bit unfortunate that hasn't happened yet, at least I haven't seen it happen yet, is like, you know, when you get into sort of a a senior or like a seale role within a business, you learn very quickly that investors value businesses, you know, along a spectrum. You're either a services business, you're a data business, or you're a technology business.

Leonard Murphy: Yep.

Marc Ryan: Technology businesses are the most valuable. Like think Google, right? Data businesses, yes, they're valuable, but you know, that takes a little bit more work. And the services businesses, they they, you know, turn and cough. They just don't want to have anything to do with it, you know, because they're really hard to deliver value against.

Leonard Murphy: All right.

Marc Ryan: Um and I I think what you're you're speaking to there points to like well there's actually a AI sort of opens door to to a you know a fourth model which is yeah a serviceled business but an AI automated back end which is kind of like a hybrid between services which there's a lot of industries that are servicesled but if you can take out the cost of operation the reason services are so devalued is because if

Leonard Murphy: Yeah.

Marc Ryan: You want to grow the business you got to hire more people right if I want to deliver more market researchers I got to hire a larger operations I want to do more, you know, of anything that's service related,

Leonard Murphy: Right.

Marc Ryan: I got to hire more people to make that work. And that's one of the reasons why they're not valued very highly versus technology where I just get more people to buy my technology. I build it once.

Leonard Murphy: Right.

Marc Ryan: And um I do see that like there will be this thing emerging where they actually can create more valuable serviceoriented businesses that can lean on an automated agentic type backend to deliver that value.

Leonard Murphy: Yep.

Marc Ryan: But to your point, it's kind of like how do you how do you describe that to investors? Because that's not necessarily a business model that's, you know, well understood or even proven yet to be successful by anyone. But the idea is that like, hey, we might not sell as much revenue-wise, but we're certainly going to deliver massive profitability and the value on the business should be higher.

Leonard Murphy: Right.

Marc Ryan: Like the multiple you get on that profit should be higher than what you would as a just a plain services business.

Leonard Murphy: Yeah. There's a uh there's I can't remember the name of them, but they're they're a consultancy and recently over the past couple months they've they've been leaning into the idea and they they've coined the term service software. Um uh which fundamentally is what we're talking about, right? a thin layer of of service and expertise with a s*** ton of AI to create uh scale and efficiency.

Marc Ryan: Yeah. Yeah.

Leonard Murphy: The uh and and I think that's that's right to your point though I would argue that even some of the valuable companies that we have seen recently uh scale you know this acquired by meta uh fundamentally they are any of the AI training set companies are taking humans taking their information and delivering that which fundamentally they're they're gig networks right?

Marc Ryan: Right.

Leonard Murphy: They are they are scaling service. This service is just kind of a micros service type of model. Um but yeah, we see those companies getting billion dollar valuations.

Marc Ryan: Yeah.

Leonard Murphy: The um I look at that and think that there is uh there is a path for insizes especially in kind of the panel side of things. real panel, not necessarily programmatic to get investors to see, look, we are actually in in the business of facilitating human uh information and experience and delivering that at scale to drive greater AI efficiency. And that can be a gamecher for many companies in the industry. Um if we're part of that tech stack if if we stop if we start thinking of ourselves as part of an ecosystem and a stack uh that is unlocking more value from this core asset which fundamentally is just

Marc Ryan: Yeah.

Leonard Murphy: Information overall. Um but again making that transformation for established businesses that have very specific models and and you know pricing strategies etc etc is going to be a challenge. Um they can conceptualize it to to make the leap.

Marc Ryan: Yeah. Yeah. I made an analogy a number of years ago. I called like um market research the Patagonian toothfish of the indust of industries. And u it's an analogy goes to the fact that Patagonian toothfish most people would think of that and they think that sounds like a disgusting fish. I don't want to eat that. But um it's actually a fish that was discovered by a chef down in Chile and he's like Patagonian toothfish. He cooked it up. He's like this is really tasty. Can you start importing this to New York? And he started selling in his restaurant as Chilean sea bass at 30 $40 a pound which is now considered a very high you know highly valued premium fish. uh it's you know market research is something that's highly valued it's premium but people don't like the name right is something about that and um I think that's been a challenge right I it's funny I did a when

Leonard Murphy: Right.

Marc Ryan: I was at Insight Express I did a presentation for the team and this I guess shows my age but when I was first getting into research you know we would do cross tabs and I remember talking to the data processing individual and she said there you can only have 80 columns of data points running across the top of the cross tab and then you can have as many like rows as you want but only 80 columns and I was like okay that's funny all right then I'd get rid of my 80 columns and you had to think like you know men is like three columns me or unless you did you can right it had

Leonard Murphy: Right. Well, it had to fit on a print out, right? I mean,

Marc Ryan: to fit on a print out and um then you know years later it wasn't until years later that I realized that the reason they use 80 columns is because market research is an industry that evolved from early days of computers um punch cards were how you fed data. Like I would do a survey, punch some cards out and feed that into a computer. That's how you counted results in in a survey. Uh you know, marketer researchers were some of the early innovators in the computer field. And um the software had been written with 80 columns of punches that you could use. You couldn't go beyond that. And there all these things that that had been built into this software that had not evolved. And at the time the the I guess the point of the conversation I had with these folks is like if I if I think about market research we have a tendency to try not to evolve to be afraid of evolution because while we were not evolving our software and sticking to 80 columns Oracle was being born like the relational database like completely different way of looking at processing and using data and look what that turned into.

Leonard Murphy: Yeah.

Marc Ryan: It turned into Yeah.

Leonard Murphy: Right. We're using Windross and you know everybody Yeah.

Marc Ryan: Exactly. And so I think there's something there's like a lesson to be learned there like marketing research had this value it was delivering in counting and and measuring things for uh companies and a new technology came along that did a better job of like organizing cataloging data for measuring and counting things. Now there's an entire industry around business analytics and stuff that we don't really call part of the industry because we let it Yeah.

Leonard Murphy: Well, Esmart does. They they try to at least. So, right.

Marc Ryan: Exactly. Well, we let it we let it shut off, right? And it's now something different. You know, we kind of try to encircle it and include it in in the numbers. But I think that there's a risk of that happening again. Like at the moment that we're in today, it's very easy to say, well, that you know, all this synthetic stuff or all this stuff happening over here, that's kind of a different machine, different animal. And you know, uh, we can do that, but you know, it's just it's all we're doing is sort of reducing the scope of of the problems we can solve on on behalf of our clients.

Leonard Murphy: All right. It's a great point.

Marc Ryan: And that's Yeah.

Leonard Murphy: I mean I remember what CX UX all those that was just part of market research, right? My first job in in research was running a customer satisfaction uh uh phone room. So you know K room um and uh there was no you know this was something different. It was just, you know, that was just one of the things that you did within market research. So, to your point, we've seen this kind of cving off uh of different things. Social media analytics, we had the opportunity to to own that. We did not um uh you know, we let that go on go off and do its own thing. Um and I think you're exactly right on the synthetic example. I know the companies that I'm working with that are building that well.
 
Leonard Murphy: So, so it's an interesting dynamic. Uh, the companies I'm working with building that aren't selling to research organizations.

Marc Ryan: Mhm.

Leonard Murphy: So, you know, the where they have the most traction is generally outside of the research organization. Now, we could argue it's because researchers think that it's less uh, you know, lower quality or whatever the case may be. And it may be, although the folks that I'm working with, I don't think that's the case at all. Um so we've seen this two things happening this uh this democratization of insights driven by technology across the board right and that's caused these this kind of breaking off or fragmentation of uh of the industry. uh but technologies enabled the product manager, the brand manager, the marketer etc. to to do these things in the way that makes sense for them across the board. So this decentralization and fragmentation uh but that's perpetuating this overall kind of breakaway where the the things we think of now is just market research uh has become this smaller and smaller uh component that is I I I used to disagree with folks like Ray Pointer and and uh even my partner Greg that the the insights industry would go away and it would simply be a function. Um uh I can't debate them as much now. I I think we are down going down that path where what we think of as the industry is simply going to be subsumed into probably into just analytics overall, right?

Marc Ryan: Yeah.

Leonard Murphy: Data and there will be functional roles that know how to address specific business issues.

Marc Ryan: Yeah.

Leonard Murphy: Uh advertising uh advertising measurement, advertising testing, you know, new product concept testing. And maybe some organizations will keep that centralized um but most probably will not uh Yep.

Marc Ryan: When you I I think the way to look at is when you look at every single successful I liked your your analogy there. I think of it like is the industry as an iceberg and every single cav cving of a portion of the industry whether it's UX or whether it's you know CX medalli is a great example of that or even you know BI analytics things like every time one of those things breaks off um of the industry you know it's doing so because it's solving a problem that I think we always fail to really truly recognize which is it's usually because they're solving a problem of scale, right? What did Medadia do for CX? It let you know, you put a a code at the bottom of a receipt, you know, from a grocery store for feedback survey at scale across every single user who's ever been to that store, right?

Leonard Murphy: Yeah. Yeah.

Marc Ryan: That solves social media.

Leonard Murphy: And they use social media data as well, right?

Marc Ryan: Same same problem like you want to understand what people are thinking. What do we do? We figure out how to do that at scale, right? And I think that is the thing that we always failed to recognize. um is that sometimes scale is more important to businesses or scale of data scale of answers is more important to businesses than 100% of the quality question right so I think you know when because scale means you're applying

Leonard Murphy: Right.

Marc Ryan: Data to more problems if I can't if my choice is don't apply any data to this problem and make a guess get you know a fantastic data set but it will cost me you I got to build my own panel because nobody has enough these people. It's going to cost me a million dollars. Or I can take a middle path which is like all right, I can use some validation exercise some you know synthetic and everybody hates that word but some sort of AI augmented approach using trained on real human data. Yeah, why not?

Leonard Murphy: Yeah.

Marc Ryan: Right? If I can be confident in my decisions because that's all these people do, right? That's all business is trying to do is be confident in the decision that you know all research is about decisions, right?

Leonard Murphy: Mhm.

Marc Ryan: you know, I got to make the right decision. I got should I go into this market or not? Should I launch this product or that product? You know, should am I is my business headed in the right direction?

Leonard Murphy: Yeah.

Marc Ryan: All those are decision, you know, oriented things. And if I can use something to help me make better decisions at scale, then I'm going to do that.

Leonard Murphy: Yeah.

Marc Ryan: And I think we generally tend to get caught up in quality and and all these other things. I'm a huge proponent of quality, don't get me wrong, but I think you kind of need to look at the different other things people are considering. It's not always just 100% about quality.

Leonard Murphy: Yep. Well, you were I mean you lived through that with Canar, right, when uh when Zappy merged, right? And uh and I'm sure you remember the the conversation uh from Stan at Unilver at that point of something to the effect um if I can uh get 80% of the quality uh at 20% of the time um

Marc Ryan: Yeah.

Leonard Murphy: And 20% of the cost then why the hell would I not do that right and think about in the automation play and of course you guys were partnering with Zappy and I know that you were internally going oh

Marc Ryan: Yeah.

Leonard Murphy: Crap especially Miller Brown you know the cannibalization of revenue but to your point it you saw clearly that's what was going to happen there was no way once that genome was out of the bottle there was no way

Marc Ryan: Yeah.

Leonard Murphy: To put it back So that driver for uh to make the business decision with good enough information based upon what the business decision was. The more tactical my rule of thumb see if you agree the more t tactical it is the more good enough is good enough. The more strategic it is then the less good enough is good enough right that that we need more accuracy.

Marc Ryan: Yeah.

Leonard Murphy: Um,

Marc Ryan: Yeah. I think and I think the thing about you know solving the scale problem is you you do exactly what you said like you bring down the cost per decision, right? So testing an ad doesn't cost me $25,000. It cost me $5,000 and now it's down even way lower than that. Like so there's been there's a declining cost of decisioning but what you're trading off is people are now applying it to more decisions. So you actually a lot of companies that go down that road end up eventually becoming bigger anyway because you know each individual decision doesn't cost a lot but there's more decisions that are being actioned there. I think the other lens on this that we need to keep in mind, you know, if I think about the marketing side of the equation now, advertising, what have you, you know, that industry is being massively affected right now by uh by two different things by like, you know, agentic workflows and generative AI like you know, hello where is, you know, AI having an impact where's the first place it's having an impact in you know, creative

Leonard Murphy: Yep.

Marc Ryan: Development. And so we now have, you know, companies and that are not having to contend with, hey, you know, I've got, you know, 100 ads to test for my brand this year. They got thousands of ads to test, you know, thousands of variants to test.

Leonard Murphy: Yep. Yep. And you can test live AB I mean effectively at scale.

Marc Ryan: How do you how do you do that? Yeah.

Leonard Murphy: Yeah.

Marc Ryan: Yeah. So I think, you know, Yeah. the the world is definitely changing that and I do think you know understanding like the scale needs but to your point I think you said this earlier in the conversation you know the question is if you if if I'm at a company established market research firm today and I look around I say you know I do agree like this scale problem if we can figure out how to take what we do and and make it scale with AI that'd be great it's like I would ask the question like is that something you guys are are prepared or able to do like or is your entire team just used to doing it the non-scaled way that
 
Leonard Murphy: Right.

Marc Ryan: Understands why I have customers that are asking for that like that's your whole life and your livelihood. If that's the case like this is when you need to start considering like how do I do this in a creative way from a business structure perspective. how do I like spin something off or how do I make a you know a particular investment in a partner to help me get there because you know one of the things we have to contend with uh you know this is what I wrote about most recently is you know what Clayton Christensen says in the university dilemma is like that the people who are stewards of your business today are the ones that are keeping you from innovating like you're end up being a victim of your own success

Leonard Murphy: And that's a tough one. The So we uh I want to be conscious of well I'm not conscious about your time. We'll go on for for hours but I do want to be conscious to the listeners. Right. So, uh, one day I aspire to do like a a Rogan type, uh, conversation.
 
Marc Ryan: Yeah, right.

Leonard Murphy: We just talk till we're done talking, you know, but I don't think we're quite there yet. The, uh, so, so I want to kind of take a couple specific areas. Um, and hell, now I just lost a train of thought. Um, the the dilemma. So we have seen massive changeover recently at the uh for many of the large uh companies in in the space right a change in leadership across the board the and there's I always have two thoughts with that the

Marc Ryan: Yeah.

Leonard Murphy: The first is well you're you're getting ready for sale um because they're always getting ready for sale at some point right um but I I don't know if that's necessarily the case in this and the second here's my

Marc Ryan: Yeah.

Leonard Murphy: My second thought usually is somebody's got to write the ship um and guide them through transformation and I think that's so now I think it's a sequence they're trying to find the perfect person to write the ship and guide them through transformation to get them ready for a transaction down the road uh and because there's a profile to all the folks we see being hired right now uh that just screw dreams business transformation, right?

Marc Ryan: Yeah.

Leonard Murphy: They're all from one of the big consultancies, you know, uh or they're pulling from outside the inside space, you know, from uh technology. Uh and and I don't know that people pay enough attention to those things that those are strong signals on where the industry is going. You were you were part of Canar, right? I've only been an outsider. Am I reading that right? Is that what you think when you see these things happening? Um,

Marc Ryan: Yeah, I mean look, um, when I think about Canar specifically, you know, but, you know, big research in general, when I think about Canar specifically, you know, they're obviously owned by private equity. Private equity doesn't like to hold things longer than, you know, five, six years on the outside and it's been that time frame. So, they, you know, they need some level of exit. you know, I don't think anyone's banking on a massive, you know, multi-, you know, billion dollar um IPO at this point. And so they've Yeah, I think they're looking at a different type of transformation now.
 
Leonard Murphy: Maybe numerator, All right.

Marc Ryan: And that sort of, you know, you read the tea leaves and new hires and that seems to be the direction of of travel. But um and it's it's good to see the desire for transformation. I mean, just bringing in fresh perspective is a huge deal. I think the challenge that the established players the bigger companies have is being uh handcuffed to you know what you're already doing today and how do you manage transformation and look I think both companies are are hiring new people to try to do this and they both need people who are good stewards of businesses to come in and say hey you know let's make sure the piano works let's make sure that we're converting enough you know of our revenue into profit that we're paying down debt, we're doing all the things we need to do from P&L perspective. And if we do that really well, we're going to create cash and that cash goes back into the business to help us invest in transformation. And I do think that they're making the right moves in that regard. The fundamental question I think, you know, they'll have to figure out is, you know, can they do that fast enough? Like, can you create that free cash flow that's going to just reinvest in the business to double down fast enough? And can you do that fast enough with the pressure coming from up top from a board level saying hey we need to see you know in the case of can we need to see an exit you know or we need to see some the next step actually you know the irony of it is that you know it may end up being that if they can position themselves well enough you know whether you look at you know K or Nos or or Tuluna or someone like that they actually do present a relatively attractive target for you know someone in the who's like more of an AI native company, right? To look at and say, "Hey, this is a great way for us to augment our our data collection, data curation process, you know, but they need to get to a certain level of transformation to do that to make

Leonard Murphy: Yep.

Marc Ryan: That, you know, attractive enough uh for those companies.

Leonard Murphy: Yep. I so as we record this in early November um because it's probably not going to publish till January you know this week uh GWI announced the uh their integration with Enthropic um feeding the data few weeks ago morning

Marc Ryan: Yeah.

Leonard Murphy: Console announced their uh their APIs the a little before that we had uh scent now integrating into Snowflake uh to to feed that way.

Marc Ryan: Mhm.

Leonard Murphy: I continue to hear more than rumors that the AI companies are circling around a few different type of companies in our space. Um uh particularly panels the So I I think you're right that there's there's evidence that's that's there that we can see.

Marc Ryan: Yeah.

Leonard Murphy: there's noise out there or or you know whispers that we can't see yet but I I believe is happening because it just makes sense um for that to occur.

Marc Ryan: Yeah, of course.

Leonard Murphy: So yep yeah yep yes yep yep

Marc Ryan: And I think even Pure Spectrum got into the game, you know, Phil when he was there launched like the, you know, training data uh features like, you know, and I so I do think that look, there's there's there's too many companies to all get the like a crazy great exit to one of those, you know, AI firms. But I do think that is like if I'm, you know, a senior executive at one of those firms, you know, that's maybe one of the things and I'm being told I need to find an exit, you know, that's one of the the locations that I would look because I do think that there's uh value in what research companies create there that's not going away anytime soon. And you know, one of the things, you know, this is one of my axis to grind. One of the things I think we've kind of been enabled to do more recently as an industry is invest in, you know, foundational research. Um, where's all the foundational research coming from the in the last, you know, 10 years? It's been from the likes of Google or from, you know, academics, but you know, more increasingly from people like Open AI and Anthropic and things like that, companies like that. And I do think you they're all working on models to improve and to increasingly throw out more research into the ecosystem that just that pairs very well with companies that try to understand humans, right?
 
Leonard Murphy: That's it.

Marc Ryan: And and so if you can take these companies that are trying to build like a scientific revolution with their models and pair that together with people trying to be the scientists of understanding how humans work, there's there's an interesting sort of value proposition that gets created there by having that connection to the consumer, you know, particularly from a panel side and then having, you know, connection to trying to build, you know, models that understand how humans work. And so I'm I'm really optimistic about the the opportunities there.

Leonard Murphy: Yeah. Well, it's you know, I mean, Meta always had an incredibly large internal research organization, right? Um, and we're we're doing very progressive things. Um, and of course now they've funneled that into AI, although you know, you don't hear uh llama mentioned at the same level as some of the other models, but I'm not counting them out um for that very reason, right? They have access to immense amounts. Same thing with Grock. you know the people get you know whatever irritated Elon but he he he bought the largest data collection platform in the one of the largest data collection platforms in the world for the pure purpose of getting training data so you know those those that wasn't a bad decision right I mean uh so I I am to your point I'm optimistic uh as well that we'll see good things come out of that and obviously you know I
 
Marc Ryan: Yeah, not at all.

Leonard Murphy: Still have a hard time seeing exactly. Enthropic is such a nebulous.

Marc Ryan: Mhm.

Leonard Murphy: They're not nebulous isn't the right term. Uh they seem to be engaged with everybody. It's clear what their business goal is. Their business goal is to be the the enterprise infrastructure for AI that and I would say they're probably they probably have already won that battle is my guess. So I think that that is where they will wind up. uh where so they'll be Google right in OpenAI will be Apple. So the uh if open AAI can survive their their instability and internal political challenges and and external legal challenges that they're going through.

Marc Ryan: Yeah. Yeah.

Leonard Murphy: Um but it's obvious where those those go overall. Yeah. I don't even know where I was going with all that other than I I agree that the the the rightful uh the logical path forward is for the the AI companies to get much closer to the insights providers because they need access to the data uh and it is their job to learn how to unlock value creation from data.
 
Marc Ryan: Mhm. Yep.

Leonard Murphy: So I I fully expect to see some interesting things like that happen.

Marc Ryan: Yeah. I I think well I think one of the the the next battleground um you mentioned this earlier. I think the next battleground for insights to figure out um is you know like insights becoming a function not a department right and these companies that are now coming into the space they're not going to the insights department that's where all the established players have their contacts right you go I've always said like the insights team when you have a question in organization you go to the insights team the insights team says here's the syndicate data we have this is what it says here's what we've done in the past or your third answer is let me go talk to one of my vendors. Um that's where you know the vendors that exist in the market today have done really well. I think one of the things that's novel that we're seeing is new vendors coming in that are coming in above that team. They're going direct to the brand manager to the product manager to the UX designer. Like they're they're circumventing an established process. And I think the the ground upon which insights will have to wage its battle in the future will not be you know fighting your way in the door of the insights team.

Leonard Murphy: Yep.

Marc Ryan: It'll be trying to stay relevant amongst a larger uh cohort of potential buyers um that are upstream of the insights team that don't think like insights people. They don't think through operational lens.

Leonard Murphy: absolutely.

Marc Ryan: They don't think with like you know uh what's my confidence interval going to be on this study. they're just thinking about solving business problems and I think that's going to be the next territory and there's going to be some companies that do and a lot of these enterprise plays you know whether it's someone like Anthropic

Leonard Murphy: Yep.

Marc Ryan: Or Gemini or OpenAI or even Microsoft with their platform wouldn't count them out are are really trying to go front and center with everybody else in the organization and they may end up being the gateway to you know to getting access to insights. So I I do think I I think what GWI did was really smart, you know, finding you and and they're leveraging, you know, what Anthropic created in the first place, the MCP platform to make it happen, which also

Leonard Murphy: Yep.

Marc Ryan: because MCP was adopted by OpenAI and it's been adopted by Google, it get makes it interoperable with those platforms as well. It's really great kind of move and I hope to see more of that in the industry. I think we have a great opportunity to start to become more interoperable with the ecosystems that are being created in these companies.

Leonard Murphy: Yep.

Marc Ryan: And the more we do that, the more we engage um on some of these forwardleaning technologies that are maybe a little unproven but worth the risk, I think uh I think the more we can continue to solidify the

Leonard Murphy: Absolutely.

Marc Ryan: value that that insights can bring.

Leonard Murphy: Yeah. Oh, Mark, you've been uh incredibly generous with your time and and we will have more of these conversations both privately and publicly, but um is there so we're we're recording this in in November of 2025. Um, if you delivered one urgent message to uh or one key takeaway, right, uh that that you want folks to hear uh and to to think through that will likely impact the next six months uh uh 12 months of the business. What would that be? Just try try and distill your wisdom into the into a nugget, my friend.
Marc Ryan: Yeah, that that's that's hard to do, but I mean I'll try my best here. I think look, I think um insights is in is in a a really good place as far as the value that it brings. But we have to be careful we don't repeat what we've done in the past, which is to ignore what's happening around us. Um, sometimes, you know, the best thing to do is to is to find ways to lean in on something that you don't understand. And it's okay to say I don't understand everything about AI. I'm, you know, call myself, you know, a guru on that front, but leaning in is part and parcel of what we need to do. And so, I think businesses need to find ways to get in there and experiment and not get hung up in in our own bureaucracy.

Leonard Murphy: Yep.

Marc Ryan: And so my advice, my one piece of advice was try to find a way to focus on these ideas and to share these ideas, right? I think one of the things most people have have an aversion to is like sharing ideas that they've got early on, you know, and ideas are like um they're like plants, you know, they grow in the sunlight.

Leonard Murphy: Yep.

Marc Ryan: Keep them locked in a closet with no sun and no sustenance, they die.

Leonard Murphy: Yep.

Marc Ryan: And when I say share, I don't mean just going to your existing clients. Find different people and talk to them like, hey, I'm in this. This is what I do. I do, you know, qualitative or I do quantitative and find someone else, someone who's not a customer. Share your ideas and what your business does because I think you discover new things in those conversations.

Leonard Murphy: Yep.

Marc Ryan: The thing to remember is this stuff is gonna it's going to be adopted first not by our traditional clients. it's going to be adopted by people who are cutting different types of uh you know paths through the industry and if we can take what we do today and share that more broadly with a larger group of potential acquirers kind of like what GWI did that got them into anthropic you're going to see more success and so my biggest thing is like you know might not understand it all but there's an opportunity to really share what we do and the value we bring um as an insights industry and most likely uncover some unexplored parts of the industry and some opportunities worth pursuing.
 
Leonard Murphy: Yep. Well, you just played good cop. I'll play bad cop. Um because the alternative there is to be the frog in the boiling water.

Marc Ryan: Sure. Yeah.

Leonard Murphy: Um and that will happen. There will be that there are are people that through whatever reason uh will not make it through this transformation. Um the uh and we saw we've seen it before, right?

Marc Ryan: Yeah.

Leonard Murphy: So yeah.

Marc Ryan: Yeah, definitely seen it before. Yeah, hopefully not.

Leonard Murphy: Uh but I encourage everybody I think everybody who listens to to uh to this show, they're they're the folks that are thinking ahead. So, uh, I don't think we'll have any frogs around here. I think that Yeah, I think that they'll take 100%.

Marc Ryan: Yeah, everybody can take this as maybe a as a prompt to uh go out and find a non-C customer and describe your business to them uh and see how AI applied to that can help and maybe it'll open a door.

Leonard Murphy: 100%. Uh, Mark, thank you. Uh really I knew this would be a fun conversation and and uh uh at least in our idea of fun I think right hopefully uh hopefully others will enjoy it as well.
Marc Ryan: Thank you. Yeah, exactly. Exactly. Yeah.

Leonard Murphy: Where can people find you Yeah.

Marc Ryan: Uh yeah, you can find me at darkmatter.com. Um if I'm on LinkedIn, Mark Ryan, Mark with a C. Um and um if you want to check out my Substack, it's gray matter unloaded. It's kind of where I unload my brain. And uh gray matter unloaded.com and you'll find me there.

Leonard Murphy: And on that note, it is it is worth it. I I will publicly apologize. I I have not uh subscribed from a pain standpoint because if people have not discovered Substack yet, uh my god, I mean, I subscribed to like a 100 Substacks. So, I I really I literally can't afford to pay everybody.

Marc Ryan: Yeah, I'm in the same boat, right?

Leonard Murphy: Um yeah, which is a real flaw with the Substack model, by the way, at least for people like me.
 
Marc Ryan: Yeah.

Leonard Murphy: Um, if if somebody Subtac is listening, I have some ideas on how to change that. Um, but uh if I wasn't already at my limit of paid subscriptions, my friend, you would be at top of the list of folks that I would Oh, yeah.

Marc Ryan: Well, I I'll have to say right now it's a free-for-all. I haven't enabled uh payments yet on my substack. So, no guiltree for anybody who wants to listen in. Yeah, I appreciate the plug.

Leonard Murphy: Okay. Well, good. All right. Good. I I I didn't recall whether I saw that or not, but fair enough then. Now I don't feel bad. So, never mind. I'm not going to apologize to you, but it's definitely worth subscribing, guys. It's uh it's always brilliant stuff. All right, thanks.

Marc Ryan: Yeah.

Leonard Murphy: Uh thank you so much. Thank you to our listeners uh for giving Mark an excuse to to do this in a more structured way than we probably would normally do it. Um and thank you to our sponsors and producers. And that is it for this edition of the CEO series. We'll be back with a new one soon. Bye-bye.