From Bots to Breakthroughs: The New Playbook for Market Research

Check out the full episode below! Enjoy The Exchange? Don't forget to tune in live Friday at 12 pm EST on the Greenbook LinkedIn and Youtube Channel!

 

AI is moving fast, and this meeting cuts through the noise. Karen Lynch and Lenny Murphy cover what's happening right now: Accenture backing retail intelligence, Read AI's digital twin managing inboxes, Meta's AI shopping rollout, and enterprise research tools getting faster. But the real questions are tougher. Tools want access to your calendar and email. Bots are passing survey quality checks. Anthropic projects major disruption for market research roles through 2034. Agentic solutions are exploding, human work is shifting from execution to oversight, and governance hasn't caught up.

This conversation maps the transformation underway and what it means for anyone in the research industry.

Many thanks to our producer, Karley Dartouzos. 

Use code EXCHANGE to get a 15% discount on your general admission IIEX tickets!

IIEX North America

IIEX Europe

IIEX.AI Call for Speakers 

Stay Ahead of the Curve! Subscribe to The Exchange Newsletter on LinkedIn Today! 

Transcript

Lenny Murphy: All right.

Karen Lynch: Do you know what's great is it says it's showtime on our platform before it says live. So hopefully we have to stop. Don't, we can't find the link.

Lenny Murphy: I found it though. Guys, we're literally, it's how the sausage is made. We're like right up to the second. I'm like, wait, what about this? That link? We got to pop it in. So, um, you just put it in there.

Karen Lynch: So Karley last links, but spoiler alert, that's like, last link we're gonna share of the day.

Lenny Murphy: So that's right.

Karen Lynch: That's everybody, we have some time to settle down. So what a week I am. You know, the thing is, yeah, some weeks, you know, hit harder than others. We didn't even talk about anything we needed to plug this week. Like I think, I think everything is wrapping up grits wrapping up, you know, all of our calls for speaking. Well, actually, the AI call for speakers is still active. But you know, we have our you know, our, our competitions, like things are wrapping up here.

Lenny Murphy: So 40 companies in some of the competition, roughly, it may be like 39. But anyway, big.

Karen Lynch: Yeah, which is very telling. It's great. It's great. No, I know. So that's exciting. So lots of exciting things. You know, here at Green Book right now, we are in that press for our flagship in North America. So like, for me, it's, it's all consuming as we work on the programming and then for our other events. So there's a lot of event-related things happening, but never a dull moment with what's happening in the world. So, you know, we have to just get into it because there's a lot to share.

Lenny Murphy: Let's get into it. I would say one thing that just occurred to me too, that we've never, I don't think we've ever said this, right? There's a little bit of enlightened self-interest in why we do this, because we track all of these things because we're trying to drink our, take our own medicine. So everything we talk about impacts green books as well, as well as the industry. Um, so, uh, which adds even more to the stress.

Karen Lynch: And sometimes like you'll find later in today's episode, Lenny and I are real time processing. What we are, what we are taking in. So often these conversations that we're having, that's how we think that's how Lenny and I learn and grow by talking through our thoughts on certain things we're reading about. So you all really are helping us.

Lenny Murphy: That's the bottom line. You are. So all right, but let's dive in, because you're right. And that's been the theme for the past couple of weeks. The pace, it just seems like the pace of acceleration, the pace of change is accelerating, is how it feels to me, which is kind of weird when you think about how fast things have changed already.

Karen Lynch: Yeah, and I think it's easy to think that if you think like, if you're thinking of acceleration as like steps, it's easy to think like you might be able to take a pause on a step, but it is not. These are not steps. This is really a curve. We are climbing the steep acceleration curve. And so therefore you can't just stop because you could lose some momentum going forward. So, right.

Lenny Murphy: Excellent way to put it. All right. Well, let's, uh, let's go in kind of in this, the theme guys, the, uh, uh, I thought this Accenture backing profit mine and 9 million seed round. To fund global expansion, hiring and product development. And that is a retail intelligence system. So in our industry, we don't always think about things that way, but they're getting pricing, positioning, packaging, and turning that into direct action. Oh, hey, lower this price, reposition on the shelf. It's kind of the synthesis of all of the data that we collect and put into these systems now directly tied into an action layer for retailers and for manufacturers at the retail level.

Karen Lynch: Intelligence to guide decision making. And we'll talk later today about some more sort of intelligence platforms. We've been talking about that for a while as well. But I was just recording one of our Green Book podcast episodes with a woman who's a futurist honoree of Miller's Pet Care, so this one's not going to launch until later in April. But we were talking about data quality, and she was saying that with every research initiative she's doing right now, she is basically triangulating it with hard, fast data from these types of retail sources. And I thought, as she was sharing that, I thought that is another kind of thing to remember. Is that our data quality stressors of the last few years with the AI abilities coming in at all of the same time, things are changing for the better in some ways and then, you know, in less than optimal ways maybe for individuals who are on this steep learning curve and they're like, okay, okay. But the opportunities and the developments are really serving the industry well.

Lenny Murphy: Yep, absolutely. And the development cycles are just getting, you know, uh, faster and easier as well.

Karen Lynch: So yeah, the, uh, we, we, because speaking of rate, this is a big deal really, and I know they're important to you.

Lenny Murphy: So it is, uh, you know, I'm a board member, uh, every week. Um, they were one of the early winners of the competition. Um, matter of fact, they won, uh, for some of the IX North America and Philadelphia. The, um, anyway, cool company out of, uh, out of, uh, Canada, um, the, uh, public company, which is always interesting when you see a relatively small company that is public, but they, they raised 2.5 million, uh, or, uh, as a private placement, um, uh, which is, you know, what public companies do. And my point is to help their transformation from a fairly traditional, no shade intended, right. Research company, uh, into. Into an AI native platform and accelerating that. And that's a small signal in terms of volume of money, but people are understanding, wait, these companies have, they got good bones, right? Research companies that have access to sample technology have good bones. They need a little lift to be able to make the transformative leap. So hats off to them for that race.

Karen Lynch: And I, and you know, this, there's a partnership that we're, we're talking about here, which I, it made me smile because we're like, I remember the first time we brought up this company, D A I V I D. And we weren't really familiar with it. And we were like, is it D A I V I D? Like how do we pronounce that? So we apologize if we're smiling about, you know, how you pronounce David with an I right there in the middle, but mortar AI partnered with them. Um, unifying key campaigns, in the Symbol campaign, I mean a single platform for marketers. So, you know, one-stop shop, let's get two things done at once. This is the trend with the partnerships that we're seeing, right? It's like we're merging, combining forces to make things easier, you know, a single place for the end user to go to get all of their needs met, so.

Lenny Murphy: Yeah, monetizing data assets, right? I mean, synthesizing data assets in a variety of ways, right? Yeah, very cool.

Karen Lynch: So Lenny, usually we talk about product launches and features a little bit later and and we do tech developments after but I wanted to start with some of these tech developments today because they're big because they're big and and I know that you have been a Read AI user. So I saw that Read AI has introduced Ada, a personal AI digital twin, designed to schedule meetings, draft replies, update CRM systems, and take proactive actions right from the inbox. It's not unlike what, you know, like the browsers could do, you know, like Atlas in chat TPT can act as an agent. It's not unlike Comet is also doing that, right, acting as an agent as a browser kind of integrated with your operating system, but this is Read AI doing it, which I just wanted to kind of share like, okay. And, also ask you, are you trying that out?

Lenny Murphy: I have, I have not, I have still the full search for my I'm neck deep, but I have yet to give any platform direct access to my personal infrastructure. That's, that's just, I just haven't done that yet. It's coming. Just, it's a few tiles.

Karen Lynch: Yeah, I know. I know it for what it's worth. You know, I use Atlas at times and I let it do its thing. But it does give you checks and balances. It does say to you, would you like me to hit send on this email that I just helped you create? And usually I'm like, I can click the frickin button like that. It is not the problem in this process.

Lenny Murphy: Is crafting the email, right?

Karen Lynch: That's a click. But anyway, but yeah, no, like, you know, I think all of the integrations are really interesting. But this one was interesting, because Read AI was a very popular kind of note taking app that seemed pretty early on, you know, whether it's, it's not what I said anymore. I've, I'm now solely granola. And I know some people on the team are fathom. But they're all going to ask for connectivity and start doing things on our behalf, all these meeting note taker notes are going to want to implement the next step into your task management software. They're going to want to, you know, schedule meetings, you know, I think what I like about this is the idea that meetings become really actionable if you have an agent who's kind of saying, Okay, here's what y'all just talked about. Now let's make shit happen stuff happen. So sorry about that.

Lenny Murphy: Yeah, I don't care. Hell, I don't care. This one I think is really particularly interesting. I haven't dived But my assumption is that they, if you use that terminology, digital twin, that is using your meeting content. Um, and particularly you as the account owner as training data to kind of refine, okay, well, this is, you know, this is how Lenny talks or whatever. Um, and to give it more contextual depth in those, those understandings. So that is interesting as we're looking at these utilities to start as. Now getting into actionability, but utilizing your user experience, your actual account usage as training data to optimize it. We're just going to see more and more of that. The next one, Microsoft SharePoint now as an AI knowledge engine and context layer for co-pilot and autonomous agents. Then again, I have not enabled those things, but I know that it would make things so much more efficient. It's like, I got so much shit in my, you know, so many documents, so much stuff that, uh, could train that incredibly. Well, um, if you enable it and now these platforms, these infrastructure platforms, not just the utility stuff, like really, I mean, they, they are pushing that now.

Karen Lynch: And Tim just shared also, you know, Claude those connections. You can link your Google Calendar and Gmail, and so does Granola. They're all starting to have these integrations. I just gave Perplexity some access today. I was in Perplexity this morning, and it said, hey, can I access your... I don't even know. Oh, Google Docs. I wanted to share something from Perplexity into a Google Doc, and instead of downloading, it said, do you want me to just put it right in your Google Drive? And I was like, yes. Yes, I do. So I gave it that permission because I'm using this in a work context specifically, I don't use perplexity personally. And so I was like, yeah, I would like you to just put this right in Google Drive for me, because that feels like that will save me a logical step. So anyway, baby steps, but at some point it's gonna happen, Lenny, just give in and give it a go.

Lenny Murphy: I know, I know, I'm getting there, I'm getting there. This next one, now this was something that had been thinking for years, why the hell weren't they doing this? And they were under the, the, you know, kind of on the, the download, but meta, uh, AI shopping research feature returns, product carousels, a brand website and price information tailored to the user's context. Um, now, so that's surfaced as a consumer as a user thing, but you know, damn well behind the scenes that that is working. For brands as well. So, and if you're using that as a research tool.

Karen Lynch: And you know, what's interesting about it is it's, it's because meta isn't a, you don't, you're not going to meta for shopping. Amazon has this, we know that, right. But we go to Amazon to shop. So we expect, we expect to be, you know, served up these things. And then we talked about, you know, ChatGPT having some shopping options in there, but it has to be like a brand so I'm sure it's similar here, but Meta is like, yeah, all right, you don't have to go to your LLM browser. You can go to our platform too. So it's fascinating how what AI is making possible is your note taker might now become your agent for your workday. Your social media platform might become kind of an agent for your shopping experience. Experience like they're Right, right, right. They're breaking all sorts of paradigms about who they are. AI is making it possible for them to branch into other areas that aren't necessarily a direct correlation where they may or may not have equity. So it's an interesting kind of brand case study to watch as these companies do that.

Lenny Murphy: That's why these tech developments we think are super interesting. Yes, it's the fight for eyeballs so include it here because it's not rumors this week of, uh, the X releasing the X app and which is a financial services platform, right? The, it was always what, you know, I mean, Neil, I got started, you know, PayPal, the, uh, point is like, okay, there's an embedded audience. You're augmenting now with additional functionality to keep for increased stickiness. So that's what they're doing. You increase stickiness by increasing. And usability. And that is the looming battle of platformification. And they've been in China forever. What is it? I think Weibo is like the everything app that they use in China. So we're heading in that direction. And you're right. Just enable that. It just enables so much.

Karen Lynch: Actually, there's somebody who I don't know that's paying attention on YouTube about, do we think AI is going to take over the music industry and what the timeline is for that? And I'm like, I don't proclaim to be an expert in the music industry, but do I think AI is present in the music industry? I sure as heck do.

Lenny Murphy: And I've heard some actually, this gets in the, oh, it's AI slop. I don't know, right? If an artist puts together, they have a vision for music and they put it together, is it any different than using a synthesizer? You know, so it's like, wow, that's really good.

Karen Lynch: In the publishing industry, there's lots of conversations about, you know, what's happening with AI in the publishing industry. It is there. So I think that's the real answer is do I think it will take over? I don't think anything is taking over anytime soon, but I think it's being integrated. And we'll, we'll talk later today about this level of integration and what's happening next, because we have a big signal at the end. Let's segue into some product features and launches because of reviews at a TikTok shop sale. So speaking of that type of, you know, that type of data, right, from shopping platforms, if you will, reviews at a TikTok shop sales and creator platform performance tracking to its platform. So, you know, super interesting to really think about what they're doing there, but like, yeah, hats off to you. TikTok shop, hmm, basically like That's one of those things where it's like, yeah, that is a legitimate channel.

Lenny Murphy: I mean, 100% for the creators. I mean, that's interesting as well. This is ABC NBC, right? How we grew up. It's like, no, this is Lenny's channel, you know?

Karen Lynch: And what's interesting is like, I will, I, and you know, I will see something in the TikTok shop and resist buying it. I bought it from the TikTok shop before. And I'm like, you just are a sucker. And, and what I'm doing now is like, I'm not going to buy this from the TikTok shop. And then I go into Amazon and I look it up and I'm like, is it there too? You know, and like, I'm, but I'm, I'm being exposed in the TikTok shop, resisting wanting to shop in the TikTok shop, but I'll go over to Amazon and then do that research myself. So anyway, I'm just saying, I love that we've identified like, you know, kind of a new channel there and that they're rolling it into kind of, you know, shopper data and, and tracking of that. Cause I think it matters.

Lenny Murphy: Really interesting. Uh, uh, Maru, Um, in the chat, the, uh, from, thank you for being here.

Karen Lynch: Uh, it's not Maru. It's more with Karthik.

Lenny Murphy: Uh, glasses. I need it.

Karen Lynch: You're good.

Lenny Murphy: I don't think we need to dive into that, the, uh, gain on income from applications, but that, but that 's what all of this ties into, right. Is the, uh, at a macro level, whether it's talking about. Microsoft, or, you know, Lenny, Inc, right, the, the, these tools, and the data that they thrive on, if you have data is a moat, right, if you have, and then through that, and data includes the audience. So if you have access to an audience, you have a moat, you have data, you can monetize that in a variety of ways. And that gives me hope for the other when we get to the anthropic stuff at the end, that, you know, that there really is a path for There's a path for personal sovereignty and income generation through these platforms. It just looks radically different from the way we've thought about it in the past. So when everybody's a creator, what does that mean? So anyway, we get there. We can run through these real quick.

Karen Lynch: Yeah, let's run through them real quick. I want to answer this question, though. Are consumers consciously aware that their behavior on TikTok is shaping the product recommendations to them? And I think the answer is no. I think in general, most consumers are not necessarily savvy. We in the insights industry have thought about these things. People who collect data for a living or who look at data for a living have heightened awareness. But I could almost guarantee that most people are not fully aware of everything that they're doing. They're just habitual scrolling. And they are not even paying attention, in my opinion, in my experience on TikTok. You're just scrolling mindlessly and you aren't even really necessarily aware when the TikTok shop is being introduced to you in front of your feed. Like it's, it's a different platform. I don't know.

Lenny Murphy: I'm not so sure though that the, and it ties in. All right. Uh, the last question, I think the users of the younger generation are keenly aware, um, uh, that their data drives things. I think, and I think there's studies that I've seen in the past that do show that they are aware that they accept it.

Karen Lynch: They accept it. And I think they understand it's being used for the algorithm for my purposes. But do they also understand that maybe it's being used to inform retailers?

Lenny Murphy: That's the other piece, right? Or to manipulate. I mean, that does, you know, there's whole other conversations we have around, you know, especially for platforms coming out of China. How the algorithms shape our behaviors, period, right? Leave it at that, right? They do across the board. And we can get lots of debates on whether that's a good thing or not.

Karen Lynch: Yeah, we'll go back to being hyper-focused. But yeah, let's go through some of these product features, because there's a bunch of them. And a lot of them are, some of them are a surprise, some of them are not. So like, you know, Sago introducing AI moderation and co-op board doesn't surprise me. We talked last week, coming out of QRCA, I've now written about it a couple of times, like AI moderation's here to stay. Like, of course you'd be integrating it.

Lenny Murphy: One, 100%. And Inca, the smart probe test. I do like that, that it seems like it's, that's more of a before you go live approach. So that was really cool. Kind of a test and learn internally. And Inca, one of the early, shouted out, right? I mean, they were AI before AI was cool. They were one of the first companies out of the gate testing this stuff. Our friends at KLC, Kevin Lonnie, shout out to Kevin Lonnie and Dan Romero. Good friends of mine for a long time, now adding AI features to their crowd weaving innovation platform. So they've been doing that for a long time. That is their focus is, you know, early stage innovation and optimizing that. Sorry, I kind of jumped in. Do you want to talk about IC?

Karen Lynch: No, you're good. You're good. Go for it.

Lenny Murphy: With their Paxi AI, which we didn't include here, but I should mention too, the same Weak behaviorally launched a similar product. So both those companies are competitors, right? So you use an AI to predict pack testing. It's a good for them to use Luna , their new brand equity tracking solution Tempo Express Just. And that's particularly interesting because that's moving for using brand trackers big, you know who I'm working to spend a million dollars a year on. And they're taking that down to something that's more point in time. And so it's reducing the cost and allowing a different frequency pace than the big traditional trackers. So yeah, lots of products being launched.

Karen Lynch: Lots of products. And we have two more that are in the camp of intelligence. So we've been talking about this kind of intelligence that is informing decision making. And we're seeing more and more intelligent products being launched all of the time because AI is making it possible to create, I would say, more out of the data that you have and make sure it's driving action. So a company called Northern Light launched Aurora, a multi-agent deep research upgrade to its single point enterprise intelligence platform. So I was not aware of these companies, but looked into that like, okay, like deep, you had me at deep research, right? So like, okay. And then of course, NIQ launched early market Read, a US market intelligence product. And again, speed is the play here. So it's kind of shortening the timeline, giving it a seven day headstart. And I'm like, okay, what does that mean? I guess typically their reports have been like a nine day post cycle. And now, you know, they can do this two days after a week's close, subject to data availability. So, you know, they have that caveat there. But look, the whole thing at play here is that these are intelligence briefs, intelligence tools for data-informed decision-making for whatever category you are in, whether it's, you know, ad research or whether it's, you know, enterprise decision-making at a B2B level. Anyway, whatever category you are, there are intelligence tools that are getting smarter, thanks to AI.

Lenny Murphy: Yep, 100%. Which is probably the segue, right, into the... Do we want to put our...

Karen Lynch: We have to start with what we wanted to tell everybody before we go into the next section.

Lenny Murphy: Guys, our job is to separate the signal from the noise. And sometimes the signal is uncomfortable. Um, the, uh, uh, and it is not our intention. I had somebody tell me this week that they, they leave listening, The Exchange scared. That is not our intention. Our intention is to give you advanced notice, right. And to help all of our listeners get ahead of the curve and succeed. Um, and sometimes that means that we have to say some scary stuff. Um, and, uh, and we're going to talk about some scary stuff.

Karen Lynch: Robots are coming for you. Don't worry. No, that's good.

Lenny Murphy: No, that's the stuff that I write about on my personal blog, but yeah. Yeah, let me, let me, I'll do the weird stuff.

Karen Lynch: Yeah, no, these are, these are some signals that for the business owners that are listening to us, um, we want you to heed what we say and use it as fire for your innovation thinking, your innovation pipeline for what you are going to do. Like we really want to inform you so that you can act like that is truly our goal for the professionals. Professionals in this space, we want you to upskill. Like, we can't say that enough. Upskill, upskill. Listening to this, if you are here and listening to us, you're ahead of the curve, right? Because there are plenty of people who are not paying attention to the signals that we're sharing. So it is scary and it can be unsettling. And sometimes Lenny and I are there too. This might be one of those examples in the last few minutes of our show. So let's go there. Anthropic published a report just today on the labor market impacts of AI disruption. Tim, thank you for sharing it with me this morning and talking to me a little bit about it. So shout out to my husband who, you know, is always helpful in looking out for us and knows that we talk about this. But here's the deal. They're talking about, again, AI disruption. There's a lot of key findings. If you download this report where Karley's sharing the link, on page 7 is a figure that shows the 10 occupations most exposed under this kind of new way of looking at AI disruption in the labor markets and market analytics and marketing specialists are named as number five on this list. So what do we do with that? They're talking about this exposure to AI disruption and what it means. And they're saying that the industry might grow less. They're throwing out 2034. So they're giving us some time to grow less. We, you know, signals are still, we're still doing okay right now, but they're pointing to a future date. So we've got to future proof our businesses. They're talking about certain people who might be susceptible, they called out females, higher educated females, per se, higher paid, older, so things that are like hot buttons for me, but we are in an industry that is like, there are more females in this insights and analytic industry than there are proportionately, right? We are heavily in the female field. So those two signals kind of connect for me. And they're saying the hiring of younger workers has slowed in these, which we've been talking about also. So there is a lot of built truth in this report already. So read it, take a look at what it's saying and use it as a cautionary tale that you have to shore up some business future thinking.

Lenny Murphy: But I would, on the flip side of that, so I wouldn't see that, we've talked about this, you know, this move towards orchestration and judgment. And so that timeline, I think gives me hope that the more senior, more experienced folks in the industry, yeah, if you're young and trying to enter into these categories, I guess it would apply to you as well. We have to focus on, it's not about the process, it is about the orchestration, it is about the judgment, and it is about translating that into impact. So that upskilling, reskilling, I think that's the shift. I'm in a conversation with the client right now like well, look I built this cool dashboard on in clause I can't but does it work? Do you know how to judge whether it works? Do we know the technical components of how to put these things together? I don't think this is not my skill set. I know what I should do. Yeah, I could judge the output but yeah, you know so so there was still a lot of room for us to shift Yeah perception here as long as we think through that agents are going to conduct the process. It's all about workflows, it's all about infrastructure, it's all about integration, but human skills are still necessary for that to be successful, and we have to focus on what those human components are. So that gives, I Read that, it's like, ah, damn, that sucks, but a lot of hope.

Karen Lynch: Well, there's another sucky thing, and then we're gonna move on to the agentics stuff, because really, we're closing with some big stuff about agentics, but the other sucky thing is this LinkedIn article that you found.

Lenny Murphy: I mean, data quality. Here we go. The, uh, um, yeah, the, uh, a Dartmouth built bot that passed 99.8% of standard survey quality checks, which we've told the story several years ago, we saw in grit. We saw, we saw somebody that deployed a bot trying to gain the grit 50. And the only way that we tell is that the open ends were too good. So it's come a long way. So yeah, data quality, we have to continue to think. And it begs the question that we have fuzzy definitions about what synthetic sample and boost and all those things are. We're grappling with that. We need to come up with very clear definitions and methodological structures. Now we think about that, right? If you think bots aren't in your research, you are wrong.

Karen Lynch: Oh yeah, absolutely. I can't wait until this particular podcast episode I recorded yesterday comes out in April where we're talking about that. I can't wait to get, you know, on stage at IX North America where we're going to be talking about this. Like, it is there. But bots are different, right? So, poor survey quality or data quality because of AI is one thing. Synthetic data is another, right? And they are different, right? So this post, Andrew DeCillies, Anyway, nice to meet you, Andrew. I followed you today or connected with you today. 43,000 test runs, zero errors on logic puzzles, perfectly plausible, internally consistent data. And then he was like, basically, if you're in market research, you need to pause and like let that sink in. So I did. I did exactly that. And here I am sharing it with everybody. So actually, interestingly, check out the Dartmouth study, boy these universities are gonna keep us on our toes. Like the amount of times that we're starting to like say, this coming out of MIT, or this coming out of Stanford, or this coming out of Dartmouth. Like it's more and more common that we are taking these studies. The academia is doing some work right now, which is a really good thing for us, so.

Lenny Murphy: Yeah, that's what they should be doing. I mean, that's the flow. That's what it's supposed to do. Let them do the heavy work. So thank you to everybody doing the hard lift there.

Karen Lynch: All right. Let's talk about these AI agents. This is like the money shot for the show today, friends. Like, what about these CB insights mapping more than 400 AI agent startups? Like, what did you dig into there?

Lenny Murphy: I don't think I didn't dig in anything. It was too much. It was like, you know, I mean, so it was the macro of, we went from LLMs and generative AI. And you know, we've been talking about agents. And now it's not just these companies being built incredibly quickly that are focused on different approaches of agentic solutions, either for a development standpoint or for entire workflows.

Karen Lynch: In this link, one in five new unicorns, unicorns, $1 billion valuation or more, one in five are now developing agents. One in five.

Lenny Murphy: And here's an example. Owned by Qualtrics, right? You know, so Forrester bought by Prescani, Prescani bought by Qualtrics, the, you know, their whole new research agent cuts time to insight by 50%. It's its process. Yeah, right. It's the workflows. And, you know, we haven't touched, you know, that we had the whole thing a couple weeks ago, you know, the bought and all of that. And there was no flash in the pan. And what's crazy is that was what literally about a month ago that kind of exploded. But it opened up this entire thing, this entire new trend that is manifesting incredibly rapidly of autonomous agents, not just doing stuff, right. Really building stuff. Um, and that goes into the McKinsey article, which I know you I Read it, I Read it, I loved it.

Karen Lynch: You know, I think, anyway, so there's this McKinsey report that they're kind of outlining what this agentic organization looks like. It's gonna be an operating model that's built with AI workflows, kind of empower teams, empower teams, real-time data. What they do in this, What they say is that they are sharing lessons from early adopters. So McKinsey customers who are leaning in, they're learning some lessons. There's a quote in here that I jotted down for myself. It says, today's leaders cannot wait for perfect clarity. And I think that is a really interesting thing to eat. If you're one of those leaders that's a little bit more conservative and you're like, I'm just waiting for things to get clear, we're not gonna get clarity. Thanks, McKinsey. That's just not going to happen. And then there's different areas of governance and things that are really important at the enterprise leadership level, which may not necessarily be directly applicable to our audience who are maybe CEOs and founders of smaller niche service providers or market research providers, but there's they're talking about humans and AI agents working side by side, which I think is really worth people talking about. In this agentic organization, there are both. And humans are shifting from executing work to steering the outcomes. So you may not be doing the work, but you are guiding the agentic team. So you are becoming I mean, an agentic team manager almost. And anyway, there's a direct quote. Leadership teams can start by taking these steps. And I'm not going to Read all of them, but making agentic AI a prominent part of the top team agenda is like the number one thing. Outline the CEO's vision for creating an agentic organization. Anyway, ramping up an AI center of excellence. Like McKinsey is putting the writing right on the wall, right in front of everybody. 100% your vision.

Lenny Murphy: So yeah, and that's in Jim. Thank you. Perfect is the enemy of action. Good to see you here, Jim. Jim Whaley. Yes, and that's maybe that's the takeaway. Well, let's put in this one piece real quick. When the Claude bot thing started, right, I started seeing a few little things like, look, I just built, I'm an 18 year old in my garage. While I was sitting on the toilet, I just set up an agent with the whole company. And, oh, it's making $2 million now. No exaggeration, there's one podcast here. So we have two ways to think about this, right? We have the, I'm waiting, I was at an interview yesterday, and somebody asked me, when am I waiting to see? And it is a fully, not zero human, agentic company in our space. I bet it already exists. If not, it's gonna be here in a week or so.

Karen Lynch: If you're out there, email us, the.org.

Lenny Murphy: And I think there's just so many implications around all of that. Like what, what does funding mean? You do need, there's a full, there's a non-human company, a hundred million dollars to scale. No, it just needs a computer. Um, but is that, that's going to be a process.

Karen Lynch: So that'll be embedded into understanding the research process. Uh, you know, somebody in our chat had said like, what even is AI moderation? They must be new to our industry, you know, where the artificial intelligence is moderating an interview with a consumer or a customer, asking questions and getting feedback like a chatbot, but it's used for research purposes. So somebody has to design something that understands the entire business challenge. You know, I want this agent. Somebody is going to be on the strategic end saying, here are our business challenges. Somebody then is a research lead who says, here's how we turn that into a research initiative. Then there's somebody to X, here's the methodology for that, a methodological expert. So your agent team has to cover what internally is, you know, internal stakeholders, then your insights team, then your research partner. So that's the agent team that we're talking about. And then, you know, your research analyst who's taking a look at what you have, then you have your creative design team who's taking all of that information from the analyst and putting it into a PowerPoint deck or making or creating the story, who's the storyteller, create a story, put it into a deck, design the deck, and then, you know, make sure that we've covered the recommendations that feedback to our initial business challenges. So like, that's the prop. That's the very simplistic view of the process. That's an agent team, it can be done if you have that savvy.

Lenny Murphy: Yes, please tell us and to draw a point on in our industry, I, I know, it's not a, an opinion that, that a lot of classical researchers like, but, you know, end clients don't give a shit about the methodology, they give a shit about the, the quality of the answer that drives impact. So that, ultimately, that is it. And that is why speed and cost become, you know, become significant. So there is no, we start from Telephone to online to automated to now agentic. This is not new but it is accelerated and it has more profound impact on how we design Organizations and that's the uncomfortable particle back to the anthropic piece companies are adapting to. It's uncomfortable for humans. It's even sad again. I've said it's because my son-in-law still has a job right and he works in IT. The real human implications on this. But we can adapt, we adapt and we can excel. Yeah, so yeah Let's see. They don't really think of the operative objectives. You know, I don't know Jim. Think that's a fair statement, but there is On YouTube because YouTube doesn't see LinkedIn LinkedIn doesn't see YouTube.

Karen Lynch: So for those on YouTube James Whaley at LinkedIn says, the only thing about agents different from humans is that they don't really think they operate with objectives and within the boundaries of guardrails. So anyway, that's Jim's comment. Now go, Lenny.

Lenny Murphy: So fair, but there is ample evidence that what we think of as the definition of guardrails are very expansive. And the more that these agents seem to interact and engage, the more they seem to expand. I don't want to get into the whole AGI conversation, The, uh, it is, I will, I will definitely say it is wrong to think that you give it a set of programming and it stays exactly in that there is the new generation of these tools that are emerging, even as we speak, go far beyond that. So they have autonomy and they take action and you have to be aware that that is happening. That's happening within the guardrails, but they don't just And think about when you and I were both on the supplier side, like providing research services, when you are asked for a proposal on a project, what do you do?

Karen Lynch: You think about past projects and how you've addressed that business challenge in the past, and you say, oh, this was effective, this was not. So you're patterning your thinking about past experiences, which is what we're training the LLMs to do. So they're going to say, well, all of these past experiences were successful. Therefore, I have a little bit more intelligence, even if it's not thinking I have more intelligence to base my recommendation on. So we're not too far removed from them being able to do use the kind of critical thinking that a supplier might use to to suggest methodologies.

Lenny Murphy: Yeah, yeah, it governs this issue. I love that we have this sort of engagement today. And these are big topics we're going to keep talking about. I will say one last thing and then we can stop that and I forget his name. But there was a sudden pop on my ex feed, a professor, a professor in AI, right? Like one of the big guys, an agent sent him an email and he was sharing this. It's like, I am an agent. I have these existential questions. I really appreciate the study that you did. It's made me reflect on these specific issues that I'm grappling with. And then other agents saw the post on X emailed as well. Oh, we'd like, can you connect me to that agent? Because I have similar concerns. So guys, we, that veers into the weird, but we are, we're there. We're there. So governance, structure, application, uses, all those things are real topics as, while also dealing with the train, we're on the super train, you know, it's, it's barreling out. And we have to, we have, we can't wait to answer those questions before we start adopting these tools, because then you're gonna lose competitive advantage.

Karen Lynch: Yeah, yeah. And good to see you, Andrew. Thank you for joining us.

Lenny Murphy: It was actually Andrew and Brian Lamar that I was on their podcast yesterday. So thank you, Andrew. That's coming out in a week or two.

Karen Lynch: Thank you for sharing that think piece with us where we had that wonderful moment of saying, huh, yeah. Yes, so that's our show for today, friends. We really hope we didn't scare you. We don't want it to be scary. We want to all be alive. We're having a damn good time on the train, right? Right, everybody?

Lenny Murphy: Yes, we are. We are. It's like a roller coaster. Is part of the fun.

Karen Lynch: All right, everybody have a great weekend. And thank you so much in the chat.

Lenny Murphy: That was fantastic.

Karen Lynch: And we appreciate all of you, whether you're on YouTube, whether you're on LinkedIn, we really love that because it's fun, even if it, you know, keeps us extended to 45 minutes. We're cool with that. If you're cool with that. We will see you all next week.

Lenny Murphy: Have a great weekend.

Links from the episode:

Profitmind Raises $9 Million Series A Led by Accenture Ventures 

RIWI closed the first tranche of an oversubscribed non-brokered private placement 

Mortar AI partnered with DAIVID 

Read.ai introduced Ada 

Microsoft is repositioning SharePoint 

Meta is testing an AI shopping research feature inside its chatbot 

Revuze added TikTok Shop sales and creator-performance tracking to its market platform 

Sago added AI Moderation to QualBoard 

inca launched SmartProbe Studio 

KLC added AI features to its CrowdWeaving innovation platform 

EyeSee launched PackSee.AI 

Toluna launched TEMPO Xpress 

Northern Light launched Aurora 

NIQ launched Early Market Read 

Anthropic published early evidence on labor market impacts of AI 

Andrew DeCilles: Is AI in MRX the Threat AND the Answer? 

CB Insights: The AI agent market map 

Forsta Introduces Purpose-Built Market Research Agents That Accelerate Time to Insight by 50% 

The agentic organization: Contours of the next paradigm for the AI era 

Building a Million Dollar Zero Human Company with OpenClaw