Anticipate What’s Next: Combining Trend Forecasting and Synthetic Consumers for Smarter Decisions 

Hello, everyone, and welcome to today’s session, anticipating what’s next, combining trend forecasting and synthetic consumers for smarter decisions. It’s a lengthy title, but we will be breaking down exactly what that means in the next forty minutes or so. But in short, today, we want to look at how organizations can move beyond retrospective reporting to more predictive foresight driven intelligence and how AI generated personas can be used to test ideas, explore emerging audiences, and genuinely scale innovation. So whether you are in insights, marketing, product development, strategy, you’re in the right place, and there’s a lot for you to learn here. So a little bit of housekeeping up top. The agenda today, first, we’ll hear from Joe Rini, is the senior director of product management here at Market Logic, who will walk you through what we mean when we talk about AI personas. He will then hand it off to Mario Coletti, the managing partner at Next Atlas, who will introduce their platform. And then we will have a quick demo of both both the Next Atlas platform and how it integrates into our DeepSights environment. But first, who we are. So I just mentioned DeepSights. Market Logic is the provider of DeepSights, which is the leading special purpose AI solution for market intelligence and insights. Deepsights provides the market context layer of any enterprise AI AI stack, required to underpin high stakes investment decisions. So we are aiming to equip business leaders across teams with trusted market cons customer and competitor insights at scale. So that’s a little bit about us. And then we also have our partner, Nextatlas, here. Nextatlas is a pioneer in trend forecasting and predictions, with a focus on early adopters. Their platform evaluates information from over three hundred thousand users, extracting and filtering data for relevance, quality, and authenticity, revealing evolving consumer behaviors, and confidently predicting developments not yet visible on the horizon. So just with that, you might already see the synergy between our two, solutions, but we are, of course, excited to show that, show you guys what that looks like practically. So before I hand things over to Joe, I’ll just remind you to use the q and a throughout the next forty five minutes. We have two speakers here, so if you are directing a question to one of them, just put that in your question, and I will be back at the end of the session to go through the q and a portion. So, yeah, that’s my spiel done. Joe, you can take it away from here. Great. Thank you so much, Callie, and thank you for joining us today. Pleasure to be here. We wanted to kick off by diving a little bit into the customer research dilemma that we see the Market Logic platform, and in particular, our Persona’s offering as solving. And on the slide here, see a couple of the challenges that we see our customer base, our insights users, their stakeholders in marketing, r and d, etcetera, facing, which is this trade off between acting fast, trying to avoid missed opportunities, challenges like strengthening brand loyalty while avoiding reputation erosion, struggling with budget and investment optimization, but ultimately trying to put their customers front and center in all of the decisions that they make based on insights, which we know over the years has been a challenge and remains a challenge. The personas offering or what we’re starting to call the synthetic respondent space largely, I’ll talk a bit more why we’re shifting to that naming and so on, is or that approach, I should say, is really designed to provide forward looking audience intelligence at scale without sacrificing research rigor. So what our personas allow for, what they pull on is our customers’ current data, so not just a historical snapshot. And this is one part that Next Atlas plays a very important role in this for our customers who, who are working together with them. Our unique personas are based on our customer’s proprietary trusted data. They can be hyper personalized and built on the fly instantaneously based on that trusted repository of knowledge that we hold on our customers behalf. And what I was alluding to at the beginning here about moving towards this synthetic respondent naming or idea is we are now venturing into both quant and qual, synthetic offerings, and I’ll show a little bit of both of that, in a second, meaning that on the one hand, you can have these qualitative chat based interactions with one or more personas, but we’re starting to rolling out an offering where you can get more of a quantitative large end, so large number of respondent quantitative outputs on things like concepts, surveys and so on. So really broadening broadening the aperture in terms of what we were starting to be able do with synthetics. So the the one, let’s say, a little bit technical slide just to frame things. I just alluded to both our qual and our quant offerings. On the left hand side, and it’s actually where we’re more more focused today and where the, Next Atlas integration that we really wanna, focus on today sits, is these qualitative personas. So these are those chat based personas, which I’ll demo. You can go in and interact with, And they’re built in a couple of different flavors in the system. Either they’re and that’s number one on the slide. Either they’re based on our customers already built segmentation, persona work, and so on, and we then configure them in the system. They live in the system, and users can go and chat with them. And, again, these can be augmented with the Next Atlas content, but they may also be created on the fly. So perhaps customers don’t have finalized segmentation work or they have it in some regions, but not everywhere, and they want the freedom, the flexibility to leverage all of their insights, their Market Logic held insights, also the Next Atlas insights and build personas instantaneously based on user desire. We also support the ability to generate those personas. And there’s a couple of flavors here. That’s why you see two and three on the slide. We can do this from a knowledge repository. We can also do this from raw usage and attitude survey data. So really access that that rich raw data often underpinning, for instance, demand spaces work and so on, and spin that into personas on behalf half of our customers. And I think that’s something very novel that we offer that we don’t see many other providers offering. On the right hand side is our quant offering, so a bit of a newer offering for us. And as I said, we spin up panels of synthetics, and these are then able to reflect on concepts and so on, give you back quant outputs like top two box, the Likert distribution, and so on, and also allow you to chat with the results, if you will, chat with the brain of the panel to drive even further insights from that. So what makes us different? I mean, a couple of just key points, and we’ll kind of go through these as we as we go through them and we’ll continue here. But we are not based on just generic LLMs, large language models. Rather, as I’ve stressed, we’re really building our personas off our customers’ proprietary data sources as said, be those those in house data sources, Next Atlas, and so on. Rather than static and slow personas, ours are dynamic, updating, using the most recent research that our customers hold on the fly accessing relevant sources like Next Atlas to enrich the chats and so on and make the relevant insights available in that live experience. And rather than being based on these historical snapshots, they’re really always current. The other thing we do is we start to connect that qual quant piece. So I’ll go through this use case, but we see use of customers starting to do early exploration in the qual environment and then take that over in the same environment to get more quantitative outputs in the same system. We know it’s a challenge to speak with niche audiences. This is, of course, in in the kind of customer or b to c space, also in the B2B, the pharma, etcetera, spaces. And that’s where the ability to generate audiences based on our customer’s proprietary data can really unlock hard to reach customer segmentations and so on. And, of course, research is slow and so on. And as an augmentation to human research, we see this quantitative and qualitative persona offering as really unlocking and speeding up access to to still keep humans in the loop, and I’m talking about human response here, but also go to the synthetics earlier and more often in the workflow. So just a couple use cases, and I’ll actually take you through using the demo in a bit after Mario introduces the the next Atlas content. I’m gonna talk you through three different use cases that leverage aspects of our Personas offering. So imagine a problem. A CPG manager needs to accelerate some concept work. They’ve got some ideas, but they need to test them, and it’s expensive to go to customer segmentations out in the field. The solution is they can use our persona agents offering. They’re able to speak to these personas, chat with them, talk to them about flavor concepts, packaging, iterate on propositions, and so on and so on before ultimately, you know, taking that out with a higher quality output that they can move along the pipeline. We also see, for example, customers wanting to validate hypothesis about customer needs, and here are the use cases, and I’ll show this in the demo. Why not unload a lot of the qualitative research that could be done with the synthetics to our AI moderator offering? And I’ll show you how you can simply brief the system and now let the AI actually carry out the interview with the with one or more respondents based on our customers’ own guides around how research should be carried out, how interview should be carried out, and I’ll show you how we can take you through that and actually derive insights fully automated here by the AI system. And lastly, the the panel play that I spoke about, now it’s about how can we actually test messaging campaigns and so on out there. And the idea is go to one of the synthetic panels, which is niche and tailored to our customers’ particular segmentations, upload the concepts, and get back the purchase intent, the top two box scores, and so on directly in the system, iterate on those before probably finally taking that to to an actual paid human responding test, but having done all of that prework with the synthetics. So three exciting use cases I’m gonna highlight for you. And throughout, I’ll show you how the next Atlas content can additionally augment each of those expirations directly in the system. Great. And with that, over to you, Mario. Wonderful. Thanks, Joe, and thanks, Kelly, for the introduction. And very good afternoon, morning, wherever you are. It’s great to hear with you today and to work and be with Market Logic to develop this this great opportunity project for us. We’re very excited about the collaboration, the the synergy that we have because we are seeing great results out of that. And Kelly and and Joe, particularly Joe, already anticipated the way Next Atlas is being integrated in Market Logic and how it works. But I wanted to spend a few minutes to give you an overview about how Next Atlas work and why we are basically adding value to the kind of analysis, research and so on that you do with Market Logic. And I want to start as a beginning, as at the beginning with a problem. And what kind of problem are we trying to solve? And what kind of benefit are we trying to give to our clients? The key point and the key aim of Next Atlas is to solve an issue, which is that usually, by using the traditional data that you have collected through multiple sources, through your research and so on, the data that you got are already quite dated or obsolete. So in essence, by the time your usual radar is identified or spot a trend, usually that trend is already old or already quite well ahead, And this is an issue for innovation pipeline. This is an issue in order to be able to anticipate and come to the market with a solution which is the solution that is the real opportunity for you and really what the customer demand is. And the reason why Next Atlas has developed the technology that I’m going to explain to you is that by complementing the next status data with your legacy research, with your traditional social listening, and anything else that you use, you get a strong anticipation of what is going to happen in the market in the near future. And therefore, the benefit and the value coming from Next Atlas is basically to know what consumer will want next. And usually, we can actually create or give an anticipation that goes up to thirty months before they do. And this should actually give you the full time to go through your innovation pipeline, have the time to define what you want to bring in the market, have the time to plan and organize what you want to be be, whatever is a new product, a new SKU, a new campaign, whatever. Why and what is coming and why Next Atlas is able to do that and why is quite unique is because we have developed over more than ten years a technology that identifies what innovators and early adopters are going to do, and we constantly look into the social media space, the web space to identify those innovators and early adopters that are talking about new things, new conversation, new elements that may be important and relevant for your business. So we have this dynamic pool of over three hundred thousand verified early adopters that we constantly analyze, checking what are the conversations, what is the what is the response or the reaction from their followers. And all this is proprietary and closed in our system. So it’s not something that you get out of any other AI model, any other generative AI solution simply because the proprietary data stay within Next Atlas and basically are not available for other tools and other AI model to to understand and see. Therefore, we provide through through Market Logic, through the system of APIs, all the data that you are acquiring and answering to the questions that you are asking through the system of DeepSights, basically basically looking into and getting the answers from our three hundred thousand, over three hundred thousand early adopters. And this gives you that level of anticipation I was talking about, but also a track record which goes over ten years of history and therefore is not just only a repository of data of what is the projection for the future, but it’s also supporting the projection for the future with data that we have seen in the past. And as I was saying before, this is something that has not seen or has not been available to any other AI tool, and and therefore, this is why and how this proprietary system is coming to you through DeepSights as an exclusive data and exclusive information that you get through the questions that you are asking to to the system. We constantly check and verify that our predictions are properly measured. And in fact, when I was talking about thirty months early, this is not a claim. It is actually a measured record. Our benchmark, the usual benchmark is Google Trends, and we have over four hundred confirmation of our predictions on this level. You can see here some of them related to the anticipation of, you know, the trend related to avocado oil, pistachio cream, or hojica, but we have, as I said, more than four hundred of these demonstration that we are able to anticipate what is coming up. Therefore, whenever you are asking a question, you are being in a research, looking into your data through DeepSights, and asking for, you know, what is coming next, Next Atlas data are giving you the feedback that are coming from our early adopters and are giving you the projection for the next thirty months ahead. And as the evolution of agents have gone, and apologies for the wordy slide here, but we have worked with Market Logic, and we have worked in our system in order to translate that and change that or shape that not just in the form of traditional reports, but really making them arrive through synthetic respondents. We call them as well personas, but definitely they are synthetic dynamic respondents that we constantly refresh within our system. And our synthetic consumers, synthetic respondents are based profiles that are based entirely from our proprietary data and from the base of those three hundred thousand early adopters. They’re not this specific early adopter. They are a pool or they are basically the representation of the group of early adopters that belong and basically are applying or linked to the questions that you are asking. And so through the system, you can ask any consumer question in natural language, and I will show you through the demo how it works. And system will actually respond to you with basically a feedback that is leading into the respondents and the response from our synthetic respondents. Those profiles profiles that are grounded in the real signals that I was showing you earlier on and can be grouped so are perfectly aligned with what Market Logic do with their single personas and as well with the quantitative panels because you can actually ask individually to our significant respondent or you can also create panels. And this is another kind of key benefit that you will you will see from from my data. And and this basically is is it. So I think to conclude the introduction of Next Atlas, Next Atlas works as a as a platform, but works as a system that is transferred through APIs to Market Logic, and that you can have access in order to basically get answers to your key business question. And the idea and the promise is that through the use of that, you have the opportunity to have the first move versus your competitors as possible opportunities before the competitors do. You can have a better way to communicate and brief your teams because everything is substantiated by data. And also, you are more credible toward your stakeholders, so for your boardroom and so on because they will obviously, your your history, your story, your proposals, and so on will be credible and based on what basically our data are telling you. And and the beauty of that is that it is very dynamic, and therefore, it’s not that once you have set up and defined the the questions in a traditional market research, you cannot go back and check again. In this case, in our case, the the work is highly dynamic and you can actually keep asking questions and moving and exploring and expanding and going more deeply in the analysis of your research. And with that said, obviously, very open to your question. I will check into that, and we’ll answer later. But I would like to share my screen and take you through a quick demo of of Next Atlas, starting from the logic way that you would use Next Atlas and how the system works. So whenever you use Next Atlas, you always start from a business question, a problem, a challenge that you have in the market. You want to know more about about your business, about your consumer, and so on. And it’s very likely that you are going to start from a basic question, And then from that, you you will move forward getting more information and more detail out of that. So because obviously there is a slight time lag on the responses from the system, What I would have done is actually I’ve already started or already prepared some questions, but we will also ask them a live question, and we will see what the answer is. But just as a starting point, I actually asked to the system what are the hottest trends in hair care across Europe and Generation Z? One of the challenges with generative AI and with AI in general is that you need to be clear on the question you’re asking. And if the system is intelligent enough, rather than answering you straight away, we’ll actually come back to you with a question. And this is what actually the system does here. So it’s giving me already two options. You say, okay, you’re asking about hottest trend in hair care, but you’re asking about something related to a region, a market zone, Europe. But also you’re asking something about a consumer group or a generation, generation side. And so the system is already asking, you know, can you clarify? Do you want to have regional divergence and so understand how it works and what are the differences across different markets? Or do you want to know better your cluster in hair care trends? Well, I can say that I want to know both, but I would start with one, and therefore, I start saying I want regional divergence. And in that case, I get an early report about what are the key trends related to hairstyle. I also get a panel signal density, which is telling me that where are the hottest or, anyway, the most dynamic kind of conversation related to the topic. Topic is highly relevant in Asia Pacific, but also there is quite good relevance in Europe in general, And then we have also United States and Italy. And then I get the regional differences across the different markets. And all of that is supported by an Nextatlas outcome. And if I had a full kind of access to Next Atlas and the Next Atlas repository, I can actually click on any of those links and get into the full data and information related to to the topic that are coming from our whole Next Atlas repository. But then, obviously, the topic of today is talking about synthetic respondents. And therefore, having had the initial generic kind of answer from from the system related to health trends. Now I want to know exactly how this works with different consumers. And so I asked, tell me or create personas which match those kind of level of interest related to the topic. And the system create for me five personas in this case, four female and one male. And on those persona, I get the full profile of the persona, and I also have the Next link, so I get the direct information and direct data coming from Next Atlas. And I also get some kind of behavioral information. Plus, I also can chat with Zoe in this case and ask Zoe specific questions, or I can select Zoe and perhaps select the the next question, the next person, and I can create a panel and ask to do the panel, which is something that I obviously did, so I will show you in a moment. But as you can see here, I get a number of different synthetic respondents coming from different parts of the world, but I can actually narrow down and say, no, give me only personas related to Europe. And so everything is flexible here in terms of what you can get out. And if you create the panel, you can go straight to the panel and ask the panel, for instance, a question like, what is your primary choice when buying a product from your scalp? Health scalp health, a claim, an ingredient, and a brand? And then you will get the full answer from the panel you have created. And, obviously, you will see that the answer from each of the panel is going to be different depending on the profile of the people. Some of those answers can be expected, as you can imagine, but others obviously are certainly going to be surprising for you and for whoever is actually running a research. Now, obviously, this gives you a high level of flexibility in what you want to do. And in fact, you know, we started saying that we wanted something related to the different differences in region, but we didn’t mind also having the audience cluster. And in fact, I asked, okay. Now give me the clusters, and I can have already a view in terms of who are the key the key segments which are important for this kind of element of hair care, wellness, and so on. And so if you don’t have your research already done and run and you don’t have your segments, you can actually create or see the segments out of the system as well. So you can actually create them on the run. I just want to ensure you that the system works live. And so I ask a question right now, which is the main concern about scalp health during hot season and summer holidays? And I will ask the system to to work on on my question and and answer. But whilst we leave the system to work and think about the question I asked, one reminder is that everything that you can ask and get from the system can be downloaded. And therefore, here you see the PDF related to our original question, what are the hottest trend in healthcare across Europe and Generation Z? So everything that is actually an outcome from your query is actually being produced and can be downloaded and exported. While the system is still working on my question, I wanted to show you another question that I prepared earlier on, which is, is scalp health a concern for men or for women? And in this case, the system goes straight away in an audience cluster lens because clearly there is a cluster already created there. And the system is actually giving me the history about what are the key concern from a demographic perspective, and it’s giving me a number of different clusters. And within the different clusters, it’s giving me also the difference in terms of, you know, what is the penetration between and distribution between female and male. Here, this group here is fifteen percent male and so predominantly female. This is also even bigger for female. This one is almost exclusively female. This is almost as a very big part of male predominance or presence. And in each of these clusters, you see the full description, the age, the gender, the geographies, the kind of psychographic information, the need states, the behavioral cues, and also the link to to Next Atlas. Now I believe we should go back to the question that we were we were asking. And and so if we go back there in terms of let me see there where I was. Yeah. The sis the system will will actually give me that in a in a moment, but it’s not ready yet. Right. So basically, is this is the way the system works. The system has a high flexibility as you did see because you can actually ask questions. But in the meantime, you can deep dive in the question and split questions in different logic, like personas, opportunity, audience, regional divergence, as we have seen. And then you can actually create and get the information related to the different synthetic respondents. And from the synthetic respondents, you can actually get into panels and ask specifically to the panels all the different questions. All this works seamlessly together with Market Logic. And together with Market Logic, you have the combination and the benefit of combining the next status question with the questions that are coming from other sources that are present in DeepSights. So the additional kind of value of including Next Atlas into the system with DeepSights is basically to combine existing data, existing information with information that are coming from our innovators and early adopters and are giving you the projection of what future is, what is the future that is coming up and and therefore, what what are the projection that you can make related to your future innovation and your future systems. And I think having said that, it’s probably the right time to combine my demo with Joe’s demo and and see how the two things together work combining Next Atlas and DeepSights. Cool. Great. Thank you. I’m just sharing my screen. So I hope you see the personas environment. Thank you very much for that, Mario. I think it was really illuminating in terms of the type of content and and the value of what Nextelix’s output can provide. So I’m now I’ve now come into a Market Logic demo environment, and this is a quite typical setup that we see our customers have. So they’ll have a number of personas typically set up by maybe geography or category and so on. In this case, you could see, you know, this mocked up customer has some in a couple different key categories. And this would be a typical setup. Maybe in the personal care space, they’ve got three personas. And today, I’m gonna focus on the onset here on Martha Gomez. So really a prototypical example. She’s a thirty to forty five year old FMCG consumer. She would be fully built on this customer’s understanding of that segmentation and brought to life in our environment configured by us and brought to life there. And what would typically underpin a persona like this would be probably some usage and attitude survey data, which has been processed by our customers or by potentially you in the audience or an agency you’re working with, but also market research reports, interviews, and other data sources. And I think this is where one one area right at the onset where you can see how additionally having all of that valuable Next Atlas content directly in the persona as well in the back end, powering it can can help them strengthen the outputs. So let’s imagine in the first case that, you know, I’m an insights user, I’ve come to the platform, and I wanna just explore some of the pain points that this segmentation may have. So I’m gonna come into a chat that I had been carrying on a bit earlier here, and I was simply trying to explore a bit about who Martha is at the onset and everything that she’s answering here around her to answer my question, to describe her day, is based in that customer’s understanding of this segmentation. So pretty standard stuff, like just an understanding of what they do on the day to day. And then I wanna start honing in on a couple different areas. So the first thing is what’s your kid’s daily routine like. Right? So, again, the answer that I’m getting here, the routine for both herself and the children, why that routine works. This is gonna be largely grounded in their segmentation data and their persona understanding. I could continue to explore that path, but I now wanna start talking about some areas that are trending for her that are on Martha’s radar that might be an area we could go towards in terms of early product ideation. So the important thing to think about here is often that type of trend level data, which is quite fresh and and changing quite often, wouldn’t be in this persona description or or in the back end for the setup. It would simply not be an area that customers could could explore in too much depth. But with the next Atlas integration here, I’m gonna simply ask around some hot trends. And now Martha is gonna be able to provide me a really data backed answer here in her own words, so according to how she thinks and feels, but based on some of that Next content that Mario was demonstrating in the other environment. And it takes a second here as Martha’s reflecting on on the question and thinking about what will be relevant. Of course, in the back end, we’re actually communicating with Next Atlas here to bring that back. And what we’re seeing is kind of a condensed version of the more, let’s say, objective answer that was demonstrated by Mario, which also appears in our environment. I’ll show you in a second. But this is now being worded by Martha. So we can actually see for her which of the trends actually matter. So there’s an eco friendly angle that’s important to her. There’s something around, like, multiple use products and so on. There’s, a lack of resonance, if you will, with overly high-tech, super personalized solutions. And, of course, this is her words for this, but I can always dive directly into what the actual next Atlas report, looked like that powered that answer that was given here. And now I’m going to so you so first of you see there’s a couple additional pieces here that, appear, but they’re not overly relevant for Martha in this case, which is why they don’t appear in her word. Of course, we can always explore them additionally. Imagine I now want to hone a bit more in on scalp health. So I’m gonna ask what is trending in scalp health as I start to really hone this expiration with this persona into the key area that I want to potentially ideate around the product. Again, takes a second as Martha reflects about that. And, of course, there’s some aspects of this in the underlying data that’s gonna power Martha’s response. But, again, as we’re asking for something that’s trending, something that’s very new, we’re gonna ID actual actualize the, NEXT Atlas content here. So once again, some very hot trends that would be coming, you know, out in the past months, essentially identified by NEXT Atlas, are being surfaced here, and they are again coming to us in, in Martha’s word and so on. And, of course, I can always go and explore that further. So that might be one early kind of early stage just investigation into, one, who this persona, who this segment is, but also a bit, of what’s trending for them today that an insights person, maybe a marketer, would come into our platform, and explore. And one takeaway they might do to disseminate this internally and continue working on it, downstream would be they could download that. They could summarize the chat, which I’ll show you in a bit. Something that we’re seeing a lot of our customers do is leveraging our image generation capability. So I’ve just asked the system to generate an infographic here that’s going to take into account who Martha is, everything we just talked about, including all of that content, of course, from Next Atlas, and put that into something that’s powerful, that’s visual, that really communicates what we just discussed, and use that to additionally communicate internally around where we might wanna go with, for instance, the product direction that we’re working on. And, of course, these images are generated on the fly. We get a quick little summary who Martha is and the various challenges that she saw, the things that are trending that resonate with her, what she rejects, and so on. Remember I talked about the high-tech personalization that she didn’t like in that first answer and so on. So I might just download that and disseminate that internally. I wanna come back to so that’s that’s use case one. Let’s come back to the main platform of the personas environment, and I’m going to hop over. And you remember I spoke with the use case earlier on when I was presenting. Now I’m gonna leverage the AI capability to carry out an in-depth conversation with a couple of the personas. So I’ve actually brought in Sofia as well, a personal an actual haircare persona in addition with with Martha, and I’ve asked the system, hey. Figure out what their attitudes are towards two one conditioner and maybe discover some ingredients that would be relevant to them. I’m gonna kick off that AI moderated chat. And what happens is the system thinks for a second and puts together a package, if you will, of questions based on, one, what I prompted it, but two, the customer specific norms and guidelines around how to carry out research. And then you see this back and forth kicks off, like, right in front of our eyes here. The first set of questions have come back to us, and then there’s a series of questions posed by the moderator as it’s reflecting on what each of the personas is saying, and thinking about how it can derive or arrive at that final outcome, which was to really explore and identify key ingredients in this two to one product, right, if you remember what I prompted it. And just as you see, hands off the mouse, and, really, the system is carrying out this entire qualitative research, task, right in front of us. What’s great here is throughout this conversation, when relevant, Next Atlas content will also be being pulled in to help augment that understanding and that exploration. So let’s just give it a second. Typically, there’s about five questions as we arrive at the at the output here. And it’s almost too fast to follow with the mouse, but you can see the level of questions that the moderator is now posing back to the personas. So just get that area out here. And you can see we’ve really arrived at an actual, like, major mismatch, in, the perceptions and some some ingredients that would actually resonate here, including pricings that would probably resonate and so on. And now as that AI summarization comes to an end, we automatically summarize that for the end user. So I get this full briefing, if you will, on what the goal of the chat was, what the outcome of that chat was, and so on. I can then additionally click on compare with research, and here I can go to the larger repository of content and check if what was said in the chat by these two synthetics lines up with or aligns with the underlying market research repository of content, including the Next Atlas content again. So that’s really a powerful way to link up the overall knowledge repository with that chat that takes place. That actually takes a couple seconds to carry out, so I’m just gonna close that and come into a chat that I had previously done simply to show you what that output looks like. So I already fired the compare with research, and you see that there’s some partial alignment between what these two synthetics have said and what some of the, market research reports, themselves say that we know about that space. So, you know, kind of points me to potentially other areas I might wanna further investigate and so on. The next use case I wanted to show you, again, potentially leveraging Next Atlas content here is imagine, as I stated, right, your company or the company has several build personas, but perhaps they don’t have personas in every region. Perhaps some of them are out of date, or they simply haven’t done work in certain areas. That’s where they can leverage our create persona capability. And I’m simply gonna come over. I just had pre prompted the system. Create me a high income UK professional in the hair care space, premium consumer. So we had a bunch of hair care personas, but we didn’t have, any of this particular fit, and I wanna explore that with some concept testing in a second. So I’m simply gonna generate that, and now the system is gonna comb the repository of knowledge, potentially look to next Atlas for additional relevant information. Takes about a minute as it spins up with the persona. So I’ve just preloaded it here, and it’s generated for us, Victoria Ashworth. Again, a synthetic persona, but based on the underlying, data, I can go explore a little bit about who this person is, what sources underpinned how they were created, and so on, and then kick off a chat with them right away. So this is how you can ultimately go from a space where you do not have persona coverage to on the fly generating and being able to speak with one. And if I just, pull open that chat with, Victoria, And I’m simply going to I’ve preloaded a concept here, which I’ll show you in a second. Reflect on these two. So imagine that I have done that early expiration. Now I’ve generated a persona on the fly, and now I’m bringing two early stage concepts that I’ve got to that persona and asking, hey. Could you take a look at these and give me some early qualitative direction around which one maybe resonates better to you? And I can, of course, further explore that explore that, ask for areas of refinement, and so on. Takes a second as the persona views the image, dissects what it’s all about, and then it’s gonna give me their feedback on, what they think and so on. And while that’s loading, I’m just gonna switch over to the quantitative offering that we have just for the sake of speed here that is now the extension of that. So I’ve come into this other environment. Imagine I got the same concept, but rather than speaking to just one, two, three personas about the concept and getting back their qual opinion, I’m gonna, for instance, speak to twenty, fifty, one hundred of these agents and simply ask them to reflect on, and then we can score their outputs. So I ran that just before this call, and we can see that across those two concepts that were uploaded, we got relatively low top two box scores, a mean score, you know, in the the mid area. So this is really a a way to take that take concept testing that can be done in a call environment now and bring it into this more quantitative environment. I can then, of course, chat with those results or really deep dive into what those twenty agents, said about the concept to get a better, more nuanced understanding. And, of course, this panel can also be augmented, in the future with Next Atlas content potentially. So with that said, I’m on one screen, so let me just, come back to the main environment and stop sharing. I hope that was, illuminating, like to conclude there. Thanks, Kelly. Yeah. So I’m back. Thanks so much for, walking us through, both the Next Atlas environment and the DeepSights integration. There’s, of course, only so much we can show you in fifteen minutes of demo, but, hopefully, you start to see the, utility of these tools and the synergy between them. Of course, we looked at just two topics, mainly hair care and some ingredient topics in Next Atlas, but, these can be applied across different verticals and industries very, very easily. So, do get in touch if you want a more in-depth demo. So I know I said forty five minutes. We’re a little bit behind, but I think it was well worth our time to go through all that. And we will end just with, one or two follow-up questions, for Mario and Joe. If you have any other questions, get them in now, and we will respond via email. But for the time being, maybe I will start I will, I’ll send one question to Mario. And if you wanna comment on it, Joe, you could comment on it afterwards. But, Mario, real quick, when we talk about synthetic consumers, AI generated personas, some people might have hesitations around that. What can synthetic consumers tell me that I cannot learn from surveys or focus groups or social listening? If you want to comment on that. Yeah, that’s, that’s a great question. I think we need to get in our mind the, the, the, characteristic of a synthetic respondent. Synthetic respondent is not a single persona. It’s a panel, it’s combination of hundreds, if not thousands of personas, thousands of consumers from which we have actually gathered information, data, and so on and so forth. So a single persona is already in a way a panel, is already a way a group representative of similar consumers that will actually make choices around the product or a proposition. And so two benefits, in essence, from our synthetic respondents. The first is that we are talking and the personas represent those innovators and early adopters that are anticipating what then the early majority or the late majority will do in the future. Now, it is also almost impossible through traditional market research to spot and identify a high volume of those innovators and the doctors. You will always interrogate. You will always ask the people that are in the mass group and therefore that have already done the choice and are already mature versus some kind of proposition. They are not going the one that are telling you what the next need or the next product is. So the logic is a logic of changing the dynamics in terms of creating or getting data which are based on quantitative sources, quantitative values, and multiple kind of individuals rather than a small group. Because if we compare a focus group over a traditional market research of few hundred people, here we are actually talking about a significant bigger amount of data coming from multiple consumers, which are then represented by those synthetic respondents. Yeah, that’s that’s, I think, a good explanation. Anything from your end, Joe? Any perspective about synthetic consumers particularly useful? Yeah, I think it just, I mean, echoing a lot of what Mario said, these personas are typically, indeed they’re not to use some of the terminology, digital twins in the sense that they’re mirroring a single entity or person. They’re really built across an aggregation understanding of the segmentation, be that from Next Next Atlas is understanding or in other customer cases that we have other, for instance, survey results they’ve done and so on, where they’re doing some work or we’re doing some work to really boil down the key attributes there and make that then available to the end user. It’s it’s really not an end of one in that sense. Yeah. We have one question here about the panels that you mentioned towards the end, of your demo, Joe. I know we don’t have a ton of time to go into it, but can you explain how the panels work and how that’s different than just talking to a single persona. How do you get multiple agents in a panel? Yeah. That’s a good question. I I think, in principle, the panel is doing a lot of what the qual qualitative chat does know, which I demoed and which Mario also demoed, but it does a couple innovations, on that to derive these quantitative scores. So in fact, we are actually asking the the n equals fifty or n equals twenty respondents to reflect in plain English on what they’re seeing, like a human does in a quantitative test. But then we’re doing some back end techniques to convert those into robust scores and then aggregate them in such a way that they actually validate up against real human respondent tests that we’re able to compare to. Now, in terms of arriving at those panels in the first place, I guess there’s a lot of different ways one can do. What we have been focusing on in the past, quarter or two is really taking usage and attitude survey data from our customers. So really the raw survey results and using those to help power, the the large respondent, panels. So that’s a of course, that’s a bit of a different technique than, NEXT Atlas does, but they they work they could work in parallel and augment one another. And I’m happy to describe in more depth what we’re doing there if if of interest to to Nisha. Yeah. I mean, just a sneak peek. Joe is gonna be doing a dedicated webinar on synthetic panels, in July. I think the registration will go live tomorrow, so a little bit early on that. But it is a really interesting topic and and one of the newer, deployments that we’ve made available. So if you want to know more about that, you can check on our website tomorrow. A lot of questions coming in. I’m really sorry that we won’t be able to get to all of them. I think I will just end on maybe a more big picture question for both Mario and Joe, if that’s alright with you guys, looking at, how this is impacting, typical innovation processes. So just with the clients that we’re working with, if you can share just anecdotally the feedback you’ve received from them or what you’re seeing, while working with the clients, how are these synthetic consumers using this trend forecasting data improving a typical innovation process, or changing traditional concept testing? So, just any feedback from your end. Mario, let’s start with you, and then, Joe, you can wrap up. Sorry. Just unmute. Yeah. There you go. Okay. Sorry. Missed your point. Yeah. I just wanted to say, to ask to end here how you’re seeing trend forecasting and combining that with synthetic consumers, how you’re seeing that improve or change typical innovation processes or typical concept testing from your point of view? Sorry, I had a glitch on my connection. Yeah, so in essence, the significant change is related to two elements. One is anticipation and the use of synthetic consumer and the adoption of technologies like Nextata’s technology combined with Market Logic is moving from using static reporting and concept coming out of very highly expert and opinionated people into what really consumers are saying, and particularly the innovators and the early adopters are saying. And so you have on one side anticipation, and on the second side, you have tracking. So you can actually constantly see and check and monitor what is changing. And therefore, the combination is very powerful, particularly from the innovation perspective, because it is giving a significant advantage from a time element, but also from the ability to monitor and check how things are moving and changing and be sure that you’re actually moving forward and progressing in the right direction with your innovation ideas. Anything from your end, Joe? Yeah. I would again, echoing, I think we’re we’re we’re getting customer feedback at this point. I mean, the personas have been in the market now for a year plus that they certainly are starting to actually use the synthetics to fully, let’s say, validate or decide on concepts or product areas they want to go down without actually going to human respondents and or introducing the ability to speak to human synthetics in this case much earlier, more often, like earlier on in the ideation phase, and just getting, you know, much, much, much faster, feedback. So, like, anecdotal statements in trainings I’ve been in are three months for that, level of, output. Like, I literally get it in three minutes from the personas, which is, of course, great to hear that people are able to actually leverage them that fast, and that that often. Yeah. I think that, you know, speed is one of the most immediate ROIs you’ll see with something like this, just being able to ask a question and and get, get a persona’s take on it. But as with all these AI tools that are in the market, it is not replacing any, market research processes, but rather augmenting them, and, I hope that you can see the utility in that. In any case, I’m unfortunately gonna have to end it here. Again, a lot of questions coming in that we will follow-up with, either this evening or tomorrow morning. But I just wanted to say thank you on behalf of Next Atlas and Market Logic, and a big thank you to Joe and Mario for spending their time. We will continue having this discussion as the partnership builds and as we have more use cases and case studies to share with you. So please do look out for future events and collaborations between us. But for the meantime, Joe and Mara, let’s give a wave. I hope everyone is enjoying their summer weather, and we hope to see you at the next event. Thanks, guys. Wonderful. Thank you very much for having us. Bye. Bye. Bye bye.