Join Market Logic at Quirk’s Virtual 2026 – AI & Innovation Webinar

Hi, everybody, and welcome to our session, Meet Persona Agents, Always On Intelligence for Insight Led Innovation. I’m Quirks editor Joe Rittholm and before we get started, I just wanted to quickly go over the ways you can participate in today’s discussion. You can use a chat tab to interact with other attendees during the session and you can use a Q and A tab to submit questions for speaker during the session and we will answer as many as we have time for during the Q and A portion at the end. Our session today is presented by Market Logic Software. Enjoy the presentation. Hi. Thank you so much for joining today for Market Logic’s persona agents presentation entitled meet persona agents, always on intelligence for insights led innovation. My name is Joe Rini. I’m a director of product management at Market Logic Software, and I’m really excited to speak to you today about our persona’s offering. So before we get started, just quickly, who is Market Logic? What do you do and and so on? Many of you will know us know us, in in this space. We are a SaaS software company, a market leading SaaS provider of an insights management solution that is used by some of the largest and best known brands in the world, as you can see on the screen. And we help them manage and make sense of their insights. All of their insights connected globally to ultimately empower business decisions, new product ideas, strategic approaches, and so on when it’s time to really make a decision in the field. And we after offer a host of capabilities powered by general GenAI and large language models, including our DeepSights knowledge assistant, which provides instant instant answers to business questions based on our customer’s repository of knowledge, AI agents, and, of course, our DeepSights persona agents, which I specifically wanna speak to you about today. So before I get into the personas offering, I wanna talk a little bit about the overall challenge that we see, and that’s all around connecting the dots between all of the types of content that our customers hold and going from a passive consumption of all those reports and market research documents to more of a forward looking, always on consumption of that content, leveraging the power of technology to help our consumers, our customers make the best use of all of the content that they hold. So what are DeepSights personas agents? These are really large language model powered, synthetic or synthesized customers or b to b experts built on our customers’ unique proprietary data and understanding of their customer segments, brought to life in the platform, as I said, with the power of GenAI. And with these synthesized personas, users are able to explore the lifestyle and views of the customer segmentations, investigate and get feedback on early product ideas, concepts, marketing approaches, you name it. So it’s really like being able to speak to a version of your customers or your, let’s say, b b to b stakeholders at any time, anywhere, and get immediate feedback rather than waiting for costly and timely sessions with real customers. You’re able to go into the system and speak to these synthesized personas, all built and based on your data. And now I wanna come into the platform and immediately show you what this AI persona offering is all about. So I’m in a demo platform now, and I’m currently looking at some snacks and beverages personas that have been set up in the system. Let’s just hover over one of them. Maya Lopez, she might be, one of our customers’ segmentations named the wellness oriented snacker. So all of these personas are probably gonna exist in the PowerPoint presentation somewhere, maybe in the Market Logic knowledge repository on behalf of our customers or elsewhere within the organization. And prior to solution like the AI persona agents, customers would be expecting stakeholders internally to understand these, personas and try to use them in marketing, research, and other, such activities. Let’s hover over and take a look at Maya. So she’s a thirty two year old UX designer based in London. This is gonna be fictional information, but strongly based in the understanding of this particular customer segmentation. And the big difference with the large with the large language model powered nature of these AI personas is rather than reading about Maya in a static way in a PowerPoint presentations similar to what I was just showing you, you can actually go and speak and interact with her. So I’m gonna jump into a group chat in a second. As you can see on the left here, all of our chats are are stored, and customers can also set up projects if they’re working on particular tasks and topics. Let’s come into the snacking concept, area where I’ve had an ongoing chat with a couple of the personas. And the first thing I noticed, I have been speaking to both Maya and Sophie, simply asking them to, first of all, tell me a little bit about themselves just to further explore and understand them before I go ahead and expose them to some new product concepts that I’ve been working on. So Maya tells me a bit about herself as does Sophie. I could, of course, go further here and spend some time really digging into each of the personas and try to understand the whole segmentation approach that’s behind each of them. I’m gonna start steering the conversation towards their snacking routine before showing them some content concepts. So I just wanna understand how they snack. And, again, this is gonna be based on the data and the real customers understanding of how these two persona segmentations will be snacking. So as you can see, Maya is typically having rather healthy snacks at home with breakfast and so on, And then she talks about throughout the day what she’s eating, how much money she would typically be spending, and so on. And a similar type of answer from Sophie, a little bit more of a shifting dynamic snacking routine and so on, and that’s shaped by both particular site segmentation, maybe also some aspects of the the cultural geography and so on. Right? So the next thing I’m gonna do is actually ask the personas if they could really reflect on a couple of concepts. So I’ve prepared two concepts here, a smart pack type of, yeah, nut based snack here and a twenty gram protein bar in particular pack packaging with some of the the the copy and so on on the left hand side. And I’m simply going to paste that in to expose it to both of them. And I’ve just prepared the question here for ease and ask them if they could each assess the image both of the images and let me know what they think. Would they prefer one to the other and so on? So as I do that, the model is now going to think about and read both of the image it’s been exposed to. And now based on its understanding of this the given customer segmentations for mine, Sophie, it’s gonna analyze and try to understand and let me know what it is about the various concepts that might work for them, what doesn’t what wouldn’t work for them, and, you know, provide me that feedback for the next round of iteration on the concept test. It takes a minute here as the model is just thinking through and really analyzing each of the images. And in this particular case, we’ve configured these personas to answer or give feedback on the on things like concepts in a very structured like way. We don’t have to have it configured this way, but many, many cusp customers actually prefer that when it comes to concept testing. So let me scroll back up and show you, how Maya answered this. The first thing is she actually rated according to this preset criteria that I’ve given on a scale of one to seven and one to five, the overall appeal, the purchase intent, and so on. This is totally configurable based on our customers’ wants and needs. We can leave it pretty subjective, but we can also insert this type of more regimented criteria. And I think this is a super useful way to see the type of feedback that can be given here. The strengths, the watch outs of the packs are put together, and there’s an overall kind of recommendation given, which is to go ahead and iterate on the first one from Maya. And the same thing for the second package shown in the image, and now here’s Sophie’s take. Right? So you can see extremely nuanced feedback at this point based on this customer segmentation all brought to life by the LLM reflecting on these images exposed to them. And I think you can immediately see the, you know, kind of incredible value of being able to take concepts and put these in front of these synthesized customers at any time. Imagine the next step before actually I’m gonna go ahead and have Maya iterate a little bit more and give you some feedback. You could simply ask the system to give you a takeaway image so you’d be able to take one or both of the personas feedback on the concepts and pass that downstream to internal stakeholder. This is coming in the system very soon. So imagine I would now prompt simply type in, okay. Give me Maya’s feedback put together in an infographic. And, you know, in a matter of seconds, we get this generated image that’s, showing us, first of all, who is Maya. Right? So really encapsulating who this particular persona is, but then talking about each of the two concepts shown and providing a summary, a cartoonish kind of takeaway on what the feedback that Maya provided was about that. And you could see just the, you know, the the level of quality of the image that’s been generated essentially on the fly here for us. Both of the packagings are captured in the image. And maybe I might ask the model also, hey. You know what? Could you also produce a pack that actually fits more with particular persona, of who Maya is? And, again, the model could, in a matter of seconds, provide one or multiple such examples here as inspiration for the real downstream further packaging rework and so on that would go on. So image generation, a really exciting area that’s coming to the platform in a matter of weeks. And at this point, I’m gonna simply ask just Maya because she’s really the persona that I’m interested in for this particular campaign. Could you iterate on a few slogans that would work for both concepts? I’m now gonna leverage again the model to have Maya really start to do some early stage creative work for me at this point. So the models, first of all, created an image for me, something take to take away for stakeholders, also rework some packaging imaging. And now Maya’s provided some example slogans that work to her work for her for both of the different concepts with a little explanation why. I can now dig deeper and, as you can imagine, go where I need to go, in order to move my project forward. I could ask further questions to Maya. I could take away that copy and rework it in another environment, come back to the system, run it by further personas, and so on to really refine and bring the concept to somewhere that it needs to be before actually taking it to more regimented and costly testing methods and so on and ultimately to market. So with that, I’m gonna come back into the platform. I hope that gives you a very nice taste of the power and ease with which you can work with our personas. And I just wanna cover a couple other use cases that we’re seeing customers put in play with their personas. The first is around campaign testing. Right? So in the in the demo there, I had Maya produce some slogans, but we see customers interacting with one or more personas and trying to get their feedback on sloganing and campaign, work, and the personas are brilliant at, first of all, giving you really tailored feedback about giving campaign, slogans, but as you saw, also generating iterations and other versions, as food for thought to take away. That simple lifestyle exploration piece is still very important. So, again, rather than looking at these static segmentations, really being able to speak in an interactive way to the personas is a very powerful aspect of them in and of itself. In the b to b, in the pharma and health care space, we see a particular use case around adoption journey mapping. So imagine we have pharma and health care customers who have well thought out personas like personas and segmentations around doctor or medical professional groups, hospital procurement, and so on. And our customers are able to then speak to these synthetic stakeholders, expose them to messaging, to, email campaigns, to materials that would be sometimes regulatorily required in order to move them along, for instance, a drug job a drug adoption ladder or to gauge how they might switch particular treatment and so on. And we’re finding customers having very, very good feedback with, being able to simulate this rather difficult group to actually get in person, to speak to often, and I think that’s a very powerful place that personas are starting to play a role. You just saw the concept review, use case. I think that’s a very strong one as well. Packaging feedback as well, and, of course, the segmentation across or this comparison across multiple segmentation. So you can have these solo chats. You can also have the group chats and get that kind of feedback between both, the typical segmentation comparisons you might see. So within a given category and country, let’s say. But if you want to, you can have a US based and a German, based purse personal care persona, give feedback on a particular concept, let’s say, and really use that to juxtapose what might work in one market and bring that over to new markets. So really unlocking a kind of global or, let’s say, cross market exchange of ideas vis a vis the persona offering. So what sets DeepSights personas apart? As you may know, there’s a lot of persona offerings popping up in the market, but there’s a couple aspects of the DeepSights offering that we really wanna speak about in Drive Home today. First of all, they are extremely easy to set up. We have an expert services team for taking those static segmentations and bringing them to life, figuring them in the back end, testing them, and ensuring that they’re up to our customer standards, and ultimately turning them over to the customer base in a extremely quick and, yeah, let’s say, timely turnaround time on our personas. They’re tailorable. So as I showed you, for instance, in that concept comparison piece where I had implemented in the back end a regimented criteria methodology for evaluating the concepts. We are able to add all sorts of tweaks to the personas in the back end in terms of how they speak, the length of feedback that they’ll give, the type of feedback, the extent to which they’ll go outside the, let’s say, core segmentation understanding that’s been provided and wager wait or, yeah, wait wager opinions on areas that may or may not be within their typical realm. These are all configurable aspects that we work with our customers to ensure that they’re getting what they want out of the personas. Super easy to access, so they live and breathe in the platform. Anyone who has access to the Market Logic platform on the customer side can simply access and speak to them, but we also, of course, offer as an enterprise SaaS provider user rights around access to personas. So, ultimately, this can also be controlled so that only certain teams can access if that’s the case. Plenty of advanced capabilities like the chat history, the project grouping, the group chat capabilities, exporting, LLM based summarization, image generation, as I mentioned, is coming very soon. In a couple slides, I’ll speak all about our road map and what’s coming and and and exciting for us. So I think that’s one of the benefits of being with a insights SaaS provider like Market Logic. We have an aggressive road map, and we are constantly bringing out new features with this persona’s offering and kind of across the suite, if you will. Absolutely scalable, so it’s effortlessly the case that customers can simply scale up the number of personas that they have into new markets, new geographies, new age, cuts, and so on. So with, you know, various customers, we’re expanding and bringing the personas into new markets where there is either existing segmentation work already done on the customer side or, in many cases, not yet done, but that’s something that our expert consultancies teams are able to to do based on the existing personas that are there. And, actually, the road map, I’ll talk about other approaches to creating personas that go way beyond the existence of segmentation work. And, of course, it’s all part of the trusted Market Logic DeepSights environment. So how do our customers get started building personas? And there’s a couple of different ways that customers are able to to do this, and we’ve been adding new novel ways to create personas over the past months as this offer really expands for us. The first is, of course, and that’s what I’ve mentioned at the beginning, taking those predefined persona assets, so PowerPoint presentations, other such materials around already existing personas and segmentations in the organization and simply taking those and bringing them to life in the back end through our expert services team, giving them a real life spin that reflects all of the work that you’ve done around, your segmentations. We can also augment that and or actually build the personas and segmentations completely from scratch from transcripts, survey data, and other types of, let’s say, raw research or raw data rather around these personas. So a really classic case is to take, for instance, again, in the pharma space, for instance, groups of transcripts from particular pharmacists, health care professionals, etcetera, with no other segmentation work or PowerPoint presentations having previously been done, simply taking those transcripts and boiling them down into persona descriptions that then live in the system and letting you then go chat with them in the same way that I demoed. We’re also able to work with, survey data, like usage and attitude surveys and so on, and use that to better understand the personas and build them out. Something exciting for us is the ability to leverage trusted partners. So a couple partners we have who are already building personas themselves and working with one of our customers are Mindline and Next Atlas. Both of these are able to plug their personas into our platform so that our customers can have the the data from these trusted partners brought to life in our platform via an integration. So a really exciting area for us there, and I would love to speak more about those partners and other other potential partners in the question and answer period. And I have a big bore here because a really new hot topic for us, what we’re calling the persona builder. This is different than taking these, let’s say, static or fixed persona segmentations and bring them to life. This is really a feature that enables users to come to the platform and type some type in a, you know, plain language prompt or filter in the in the front end environment. Hey. Give me a persona who’s based in Germany, a male aged thirty five to forty five, who eats chocolate once a week. And we can then configure the system to go to either the market research reports stored in DeepSights environment or to onboarded structured data like usage and attitude surveys or to a third party provider, like a Next Atlas, and on the fly, spin up a persona that the user can then immediately begin to speak with. So that’s a really exciting area for us, and the technology is finally coming on board that we’re able to leverage with very good speed. So, you know, a matter of ten to fifteen seconds, the ability to create these personas on the fly, really unlocking true personalization in terms of building personas for very particular, let’s say, ad campaigns or exploration of very particular segmentation cuts that goes beyond what can be done with the predefined personas that a lot of our other customers have. So super exciting area. Again, happy to talk more about this in the question and answer period. It’s something that we are currently piloting, but it’s certainly coming into the the overall platform for our customer base. And all of this is underpinned by our approach of consulting, creating, testing, refining, and finally launching the persona. So whether it’s bringing to life your predefined persona assets or setting up a persona builder where the machine is then creating the personas for the end user on the fly, if you will. It’s all about our expert consultation to have the personas coming out and and chatting in a way that really fits, the way the customer wants. So before closing out and going into the q and a period, which I’m excited to hear all the questions about, there’s a couple of areas we want to highlight that are coming on our road map. If I just skip to the bottom, actually, the API support piece is something that’s huge for us and really differentiates us in the market. What this means is that you, customers are able to leverage our persona’s engine. So everything you saw in the demo there from the front end, from me chatting with the persona via API calls. We don’t need to go into technical details of what that really means, but you can then have the personas showing up in other internal in house environments, have concept and idea pipelines that are happening in other platforms, nonetheless, getting feedback from the personas as they’re configured in the Market Logic platform via API calls all behind the scenes. Again, exciting piece. Happy to speak about that in the question and answer period. If we come back to the top, image generation. So those couple of examples I showed you, in the demo, really leveraging the best in class Google, image generation models is a very exciting area that’s coming in a matter of weeks. This quarter, we’ll also be introducing an AI moderation feature. So imagine rather than having to lead the chat, with the personas, you could simply prompt the system to interview the persona on topic x, y, zed, or, upload a study guide and ask the system to perform the interview, and the AI would then carry out all the heavy lifting of actually interacting with the personas for you. I already spoke with the on the flight piece. And beyond that pilot and the ability to create new personas on the fly, we’re also looking into the ability to update the existing predefined personas for our customers via this on the fly model so that you could come to the system and say, okay. Rework one of the persistent personas into, for instance, another geography, so another country or another age cut, and the system would then look at all of the data that’s held and create a, let’s say, remixed version of that persona for the end user. And finally, I I spoke about a couple of them, but we’re really excited to be able to leverage our strong ecosystem and integration play that we have across the board at Market Logic with the personas as well. I already mentioned, Next Atlas as a partner there. We’re also in talks with a few other providers. So, I think it’s a very exciting area to be able to leverage all of the other disparate, types of data that our customers are using in order to power the personas and, have their outputs also showing up when speaking to them. And with that, I’d like to conclude. I hope you enjoyed seeing a quick look at our, AI personas and also hearing a little bit about how you can get set up with them and what we have planned for this exciting offering over coming quarters. And I would love to, continue speaking further in the q and a. Thank you. Alright. We have got some great questions coming in, which is always helpful. Had one question here there. Jody’s asking about when when you do the group chats, do the personas or the participants interact with each other and sort of build off each other’s input, or does is it just like you’re having two IDIs going on at the same time? Yeah. Great question. So first of all, hi, everybody. Great question. They do indeed, build off one another, meaning that the, yeah. Right? Like, the the the feedback from one of the personas may be reflected upon by one of the other personas downstream, and you’ll get this dynamic where one might say, you know, I agree with what Maya said, but to build on that, I would, you know, add this and that. And I didn’t show this, but you also have the ability to with an app functionality in the chat, at just a particular persona. So you can also then, even within the group chat, nonetheless, do some follow-up questions with just a specific persona. So it actually supports both directions of what, the question is getting at. This group dynamic as well as the ability to speak to just one of them, nonetheless, within the group setting. Sure. Question here too. Just asking if the original consumer research doesn’t have information on what a segment might do or think about some features, does the model infer a response, or will you get a message saying, like, insufficient data or something like that? Yep. Great question. So, I mean, this is top of mind for our customers, prospects, and so on. And I think to some extent, I tried to show in the configurability of how we set the personas up some of the ways or dimensions that we can let the persona either, you know, be verbose or reflect in a more regimented way versus a more subjective way. But part of that is also, whether or not we allow the persona or we try to encourage the persona, I should say, to weigh in on topics and so on that aren’t in that underlying I don’t wanna say training data because we’re not really training an LM, but in that underlying description that’s coming from the customer. Some of our customers really actually wanna leverage that, so they’re perfectly happy for the, the persona to be set up, but then to be able to answer on concepts, ideas, questions that obviously never existed in the, the previous research, and and just give its opinion using the power of the large language model. Others prefer it to be much more, fixed to the, the underlying description. And in those cases, the persona will, like, neglect to really give an opinion and so on. So it’s configurable, but it depends on sort of the customer’s app appetite for that and the use cases they want to, to carry out. So if they wanna be evaluating many novel concepts, they typically let the let the persona go a little bit outside of just the description because that’s kind of a feature of its ability to reflect on those concepts. Sure. Question there, dude. Just go ahead. No. Yeah. Sorry. The the the second part of that question, answer to that question is a a couple features that we’re working on now, to check the responses against our customers’ knowledge repository. So we hold all this market research on behalf of our customers in the platform, and this means that we are able to or in incoming features, able to go, okay. The persona said this about the concept or the persona has stated this, about it the way it thinks and feels. Let’s check if that’s backed up partially fully in the knowledge repository, and, the power of Gen AI makes it very quick and allows us to then do that comparison, put that back in front of the end user. So I think that’s another part of the answer to that question. Sure. Question here too. They’re just asking about how many clients are sort of using this to make decisions in place of traditional consumer research versus just, you know, kinda dipping their toes in the water while still can do it conducting other traditional consumer research? Yep. I think yeah. It’s a good question. So first of all, even the customers who’ve been using it for, let’s say, ten months at this point are on a trajectory themselves. I wouldn’t say any customer has yet replaced the other, more regimented going in front of, actual human respondents in in a panel setting on, you know, for instance, concept ideas, down the road. What we’re hearing is they’re using the personas earlier on, so it’s earlier and more exposure to customers, if you will, to help finesse, rework, and really, like, fine point these these concept before taking them forward to the other more traditional process that they’re used to. So it’s not a replacement. It’s rather, like, augmenting it and making it, more common to get this exposure early on. Sure. Great question here. They’re asking about, best practices for sort of building the personas. You know, if you observe differences in the quality of input persona input based on the inputs, you know, how they’re set up, which is like a robust qual input, you know, quant profile, social data, purchase panel information, that kind of thing. What are some of the good inputs? A great question. I was just in a customer so we were cross customer monthly session with existing for some of these customers get together to speak, and this came up today. I think let’s say it’s against what I originally maybe thought when getting involved in in building this qualitative stuff like raw transcripts, interviews, social quotes, and so on is actually very strong because the large language models really take to this type of input and can then leverage that and flip that around and really use that to power the outputs. And I think it gives it a very nice touch. That’s not to say so some some of our strongest users of of the system, it’s largely based on quant data. But if that’s gonna be the case, so let’s say, boiled down findings from u and a surveys and so on, I think in those cases, it’s best when there’s been real insights driven work on top of that. So they’ve taken that research, and they’re they’ve already turned it into persona or segmentation descriptions and added their expertise to it, added some plain language text descriptions around all that data. So it’s definitely a subjective component or, I should say, a qualitative component that helps, bring the personas to life for us. Yeah. Sure. Question you’re asking about, is a persona meant to be representative of an entire segment, or are they literally more like individuals from a segment that might or might not rep be representative of the whole segment? Yeah. That’s again, another good question. Look. I would say, like, generally speaking now, they’re meant to be representative. That said, in almost all of the cases, we actually but so either the customer’s already done this or we then do this on their behalf, which is give these personas actual names, locations, numbers of children, right, etcetera. So we we give them a human touch that then is obviously no longer representative. It’s actually a specific person, if you will, within that, segmentation. But I think, it’s still meant to be representative. Now, of course, sometimes there’s the the average is not the correct thing to have in in a case. So if it’s something like a preference for certain brands and so on, there can’t really be an average position there. So we actually do give it attributes that add some meat to it. And that all said, I think I I didn’t mention this in the road map, talk, but there’s another direction we’re going with the personas of this very implication that’s underpinning the question, which is to spin up more of, like, a panel of personas, if you will. So this probably wouldn’t be a chat interface now, but imagine the back end, there’s a hundred of such personas for a given segmentation, and now you can expose those to look at a concept and so on. So now you really do have a bunch of individuals, but with a large large large numbers, they’re then representing segmentation very well. Sure. Question here just to about, you know, some of the examples that were shown where, you know, CPG or consumer based. Have you done are there other personas in the b to b or health care or other sort of niche or niche audiences? Yeah. Great question. So definitely pharma and health care. In fact, I think of, like, of all the verticals we have, the one for which the highest percentage of that given vertical is using them is probably pharma and health care. So I don’t know. What say we have ten, like, seven out of ten are using it. So it’s definitely a really strong play for for them. And I think I listed some of the examples in the video. It’s like, could be HCPs or doctor groupings, could be, you know, hospital ministers, could be nurses and so on. So there’s a lot of different b to b or expert segmentations that it’s really a strong play to bring them to life. And I think those are some of the most challenging real humans to get ahold of and and and most costly. So in order to get ahold of a bunch of expert doctors in, I don’t know, a sub area of particular country can be very challenging for our customer base. So I think that’s somewhere we’re having this synthetic, plays is a very strong offering. Besides FMCG, of course, retail, we’re also seeing automotive, really taking to the personas. Yeah. Question here too just about asking about validation test and to compare what the personas say versus, you know, quant research with, quote, unquote, real consumers, you know, any sort of work you’ve done on that line. Yeah. I I would say to date, most of the I mean, of course, we do we we do our own, like, testing. So when we’re setting up the individual personas, our customers then typically do their own testing. So, for instance, exposing the personas that we set up now to the exact same ad test that they put in front of comparable real humans and seeing feedbacks like. We also use that to then further fine tune the personas, but that we’ve we’ve gotten very good feedback from the customers who’ve done that. They’re actually giving very similar responses. As we move into this, as I mentioned, this kind of panel, place, so quanted scale kind of, we will also be introducing more regimented obtaining past concepts and so on that have been put through the regiment testing process along with the scoring and then really using that to compare to what our panel personas are able to produce. So I think we’ve we’ve been doing some of it. And as we move into this more large end kind of persona space, we’ll be able to get a little more quantitative as well in the comparisons. Sure. Can the personas be trained, you know, not to answer questions that, I guess, are not appropriate questions to ask or, you know, any other thoughts on on how they’re they’re sort of trained to respond? Yeah. So I I think as I kinda said also with the with the degrees which the customer like us to shut them down, if you will, they can certainly be configured to stay away from certain topics or really just say, know, it’s out of my area. I wouldn’t like to speak about this and so on. At the same time, most customers kinda trust their user base to to, to some extent, do that and and don’t wanna lock the personas down so that, I don’t know. If you’re talking to a, FMCG persona who’s been set up for, I don’t know, personal care in the UK, but then you wanna ask them, like, what they eat in the morning, I think you’d still like a response there because it very well may feed into the campaign or the the product idea that you have. Inappropriate comments and so on, though, I think is is, you know, definitely something that’s, we can figure it nonetheless. Sure. And then question here too. They’re asking about how the persona agents connect to kind of the broader Market Logic, you know, DeepSights platform Yep. Integration is like. Yep. So currently, the the personas are so, of course, they they live in the platform. That’s all part of a single login, and then you’re in this safe secure platform and able to interact with the personas. The connections with other offerings is something that we’re bringing onboard this quarter, and there’s a couple different places that that will happen. The first is from our more objective knowledge based querying capabilities, which we call DeepSights, from the chat experience to be able to interact with the persona. So imagine you’re chatting away and getting trusted answers from your repository of knowledge. You could then say, hey. What does persona x say about that and so on, get that feedback directly in this other chat experience? That’s one place. The other side of the coin is in our innovation studio and always on agent suite. So this is a suite of Gen AI based agents that are surfacing trends, surfacing new ideas, and so on for customers, or facilitating innovation pipeline, the ability for them to go to the personas and get feedback kind of natively without the end user needing to actually do anything there. That’s just an automatic built in integration. So really ensuring that the the loop is sort of closed. So these personas have been set up are providing feedback on innovation ideas, trends, etcetera, all in a kinda closed loop. Let me see. I think there’s just a they’re just asking sort of if if you they’ve you’ve answered this already, but asking again about if a persona is supposed to be representative of a segment. Are they literally more like individuals? I don’t if you can just cover that real quick again. Yeah. I think I think there’s they are well, kind of rewording what I said previously. They are probably strongest when they actually are a single person with they live in a certain place. They have, like, this car, this number of kids, what they do. They have this job even though that’s representative of the segment. Because then they’re they’re giving real feedback that’s, one, sounds realistic and also is food for thought for the, the insight person, innovation person, the marketer who’s interacting with them. But I think we do a lot of work to ensure that they reflect the underlying, the core segmentation. And and often, the customer’s already done a lot of that work, and we’re just transposing that into the system. But if they haven’t done the work, we ourselves will do it. And in the I’d say in the most edge cases, we’ve gone to the customer and said, look. The segment that you have is actually quite broad or the, you know, these thirty transcripts you provided, they’re they’re too broad here to be one segment. So let’s split this into four, for instance. Right? So then we’re narrowing it down so they have a little more meat on the bones to help support the the use cases that they’re meant to. Sure. Do you have any thoughts on we got about five minutes left here. Just for people who have to take this kind of technology or the use of the test technology to their internal clients who might be like, AI, what? You know? They’re concerned about that. Any sort of best practices for how to not grease the skids, but just sort of, like, tell people how this data was created and, you know, to have confidence in it and just any, you know, things to sort of make that process a little bit easier for skeptical internal people. Yep. Sure. And I must say, like, this has probably been the one offering. I’ve been at Market Logic for many years now that’s met with the most skepticism because on the one hand, it maybe comes across at the beginning as like, okay. So first of all, this is Gen AI. Also, this is now you’re talking about synthetic customers, right, where the buy in then comes very, very quickly. And when when the kind of end results are seen like those comparisons between real human responses and what the personas can do. And the way that we can facilitate that very quickly is the ease with which we can set up personas. So most of our customers who’ve gone on to purchase and now use them, like, you know, production across the organization started by a quick touch point providing the PowerPoint presentation that contains the already created personas that are known in the organization or the the materials, which maybe will become one, and we simply spin those up in a couple days and let them actually test it and then share that with the with the internal stakeholders and let them test it. And I think the buy in is comes best when people actually use it and see the just the ease and the the real the realism of the feedback. Yep. Well, I think we’ve got everybody’s questions here. And anybody that we didn’t get to, I know Market Logic can connect with you afterwards to answer your questions directly. But thanks everybody for the great interaction. It’s always great to have great questions coming in. It means the material was very interesting. So thank you, Joseph, and thanks Market Logic and everybody out there for joining us today. Have a great rest of your day. Take care. Thank you. Thanks, Joe. Bye bye.