 Hi everyone, I am Ranjan, thanks a lot for warm introduction. I run a company called as Entropic which is primarily an AI company unlike Shubham I heard. It was a beautiful presentation about the whole architecture and design but I come from a different lens. So I am an AI guy who is trying to build solution into user research as a space and as user research has evolved and it has been contributing into the mainstream of the whole UX arena, user experience arena. What is the role of AI and how it has come together over a period of time and what's latest out there and then how it is disrupting this whole way of understanding consumer and as I think Shubham said, the role of a user researcher is actually to look beyond what seems very obvious so I will just take a continuity from there. So coming to how the user research as a process has evolved right. Historically it started with back in the days where user research as means is the core focus of user research is to get the user perspective and bring them at the center of all things that you are doing as part of your design or user experience design or any sort of journey mapping. Early days it was more about anecdotal research so I will get in people, ask questions, take manual notes, understand what they are saying. That was the means to understand consumer. And then came the internet era where you have emails and all that form factors where you can write surveys, you can roll out what you call as quantitative interviews, essentially to get the user feedback. And then from there it further evolved to something like social media evolution that happened at the start of this millennial wherein basically the focus was using more and more form factors to collect data, right? That's where you have had chatbots, you have got centralized feedback tools, some of this stuff. And something that has taken over in last a decade or so, right? AI powered user feedback, that's what we call it. And then I'll come to the sense of it as in how it is creating value lot more than how it has been in past. So there are two parts to it. One is data collection as in how I can be more efficient, fast, agile about understanding and gathering data about the consumer than the traditional way of doing things and making it efficient. And the second one is how you can assimilate all this source of information to make sense for you, right? So AI is supposed to enable you in ways where you can look at things which were not so obvious earlier, right? And that's where all the good technologies of AI comes in. And early in the decade there was a lot of chatbot, virtual assistant, automated user comments, those are the things that started getting used. And now is the era where there's a lot of mixed method research that is coming in where it's not about the boundary conditions of whether you are doing task-based journey mapping, unmoderated research, moderated research, but it's about answering a central theme of a consumer, what really his experience is like. And looking at a consumer, and this is a very interesting thing. Half of the industry calls consumer a user and half of the industry calls it a consumer. So if you go to a PNG, you will hear always consumer. If you go to a consumer internet company, you will always hear a user. And there's a difference between a user and a consumer. Consumer has a lot of life in it, so I always call consumer. So mixed method research is primarily looking at a user and its behavior for any particular experience from a lens which comprises of things like emotion, things like behavior, things like beyond what they're saying, it's not about, you do moderated research testing, you do user interviews, you ask questions. It's not about what people are saying, it's also about how they're saying, right? If you give me a pack of lays, I can say in two ways whether I like it or not. I can say, hey, I kind of like it. And I can say, I really love this particular product, right? And these are very two different statements. When you're looking at large set of user interviews, massive collection of interview database and trying to mine insights at scale, I think the problems become complex. So that's where AI is bringing in a lot more mixed method research to the industry. And what essentially it does is that, so let's say you have one of the core role of AI is analyze unstructured data and I'll not bore you with the text. But in simple terms, you have user interviews, you have thousands of things that a customer is saying. Essentially, there's an unstructured data where there's a lot of is are or, right? What is really meaningful juice out of it that you want to have as a takeaway? And can you have it evangelized within the organization is where AI comes and plays a role? The second one is categorization, the good old quantitative and qualitative, in other words, moderated and unmoderated and task based journey. Today, if you ask a researcher, I have done a screen recording map the whole user journey on a website or a mobile app and I have known what kind of timestamp is there, what is the clickstream behavior and all that stuff. And there's a user interview that he does separately, which is moderated. Now, are you these two data points talking? Is it coming together to create a comprehensive story? How about the time what you are seeing as a user action when they are not being watched versus when they are asked a question and proved? The answers are very different. How as a user researcher, you can come up with a UX strategy if you don't have a comprehensive view of what really a user is saying. And there's a lot of science and art to both of it. So that's where another domain where AI plays that how you can summarize and comprehend in a way which makes sense for you and at scale. Again, at scale is where AI becomes really powerful. And the last component of this whole AI breakthrough into user research is about can you go past the language barriers, right? Various culture, various religions, various regions do express things in various different ways. The same way in which in South part of the country we express things with whole not is not the how we express it in the North, right? So things are very different. Do we understand those language barriers? Do we understand the emotion evoked at the time of whatever we are articulating about our experience as a consumer, essentially? So that is again another play field where AI is playing a pivotal role. These are the three things major, analyzing unstructured data, bringing the world of quantitative, qualitative, unmoderated, moderated, task-based journey, user captures, screen recording. What set of information that you have trying to make sense out of it is another area where AI is playing big time. And last is a new kind of data point, right? Which is behavioral and cognitive science. And all these things taken together is where the magic is happening in terms of looking at customer like onion and you're peeling surface after surface. And can you try to understand the non-obvious subconscious behavior of a user? I ask a user, have you ever seen a survey where somebody has given 6.75 as a rating? It never happens. I ask a question, how do you feel about a product? Everyone writes eight or nine and net CSAT score always looks like 8.75. Any organization, any product feedback that you take. What does it give you as a researcher? It gives you probably a directional AB decision. Can you look beyond that to say that, hey, this is precisely a point where I was feeling so frustrated feeling in this ugly form. Where I'm feeling 20 fields, half of which I've filled in the previous page. Simple examples, and I'm not a user researcher. I'm a sales guy. So I would give that a front warning to you. So taking some simple exams on where UX sort of plays out. This is a quick example of how some of the cognitive behavior mapping sort of works where in things like eye tracking, right? Simple, you have always click tracking, time spent, all those technologies in play where you know what a user acted on, right? I went and make a decision of purchase now, pay now. But what triggered to it is not known enough. So click is like the eye movement is followed by click, right? So as a user experience researcher, if you're looking at things, you need to know what preceded the click or action. And that's where something like eye tracking comes in beautifully. What you see on the right side is fair amount of, let's say you're going through a user journey on a digital interaction. I've taken simple example for convenience. But if you're going through that journey, is it known to you what is the second by second attention, engagement, emotion, happy, sad, excited, relaxed, bored, as a user is going through the journey. What if you go through experience and second by second, I'm able to map that, fetch that. And that's what we bring in some technologies like facial coding where we analyze the face expression of the user as you go through the journey, comes into picture. How about you're narrating your experience as part of user interview and I understand your voice tonality and the confidence with which you are making claims. I like this product is a different claim than I kind of like this product, right? So can we read into those layers to really know what is real feedback versus what is kind of feedback? So those are the new age technology of the third block, cognitive science where it is coming out and bringing a lot of value in terms of how user research has been done historically. Overall, what it brings on the table is AI brings complexities, probably a lot of complexities. It brings a lot of advantages at scale also. We all know a bunch of stuff, greater accuracy, ability to look into biases, right? We all have our biases and preconceived notion as users and what we believe in and what we want to believe in. Likewise as a UX guy, user experience designer, I have a lot of biases of what I think is right and wrong. Can as an artist be, I hold back my perspective and really keep consumer at the center that requires tools like this. Where you can look at not just what customer is stating but what is the subconscious response of it? We call it system one and system two. System two is the stated response. I really like something, surveys, articulated response. And system one is something like my face behavior, my expression, am I happy, excited, relaxed, bored, sad? Am I zoned out? Am I attentive to what I'm seeing as a experience? And then eye tracking, you cannot beat the eye tracking. I cannot control a user to focus and respond in a certain way on a given experience. Every user behaves very differently. But then there emerges a pattern across 20 of your personas that this is the clear trend in which people look and navigate through your stuff, right? And your UX strategies can be very smart if you get into those kind of depths of understanding a user. So bringing in new tools with richer, deeper granular insights is where some of the AI plays there. Real time insights, obviously making things fast. You did a consumer interview if you're getting insights like 20 days later in a PPT format agency, your life is done, right? Like you cannot make decisions that delayed. Also you cannot come together to convince your bosses that I'm gonna make a decision like 20 days later when our agency is gonna throw me a PPT. So the need is how you can get all these things real time so that you can be a lot more agile and iterative. Time and cost efficiency, bottom line, top line, whatever we want to look at. Support for mixed research. Again, layman question, I want to answer and understand the consumer at the center. Method is all we as intellects have defined. Quant, call, moderated, task-based, all that is good. But at the end, you want a clear consumer understanding or user understanding. And hence, the ability to comprehend it together is gonna be massive. So that's one of the major advantages that AI brings today, particularly with some of the JNAI stuff that's happening around where you have now large language models that can mine through your conversation like last five year, 5,000 interviews. One of the customers that we work with has about 6,000 interviews over the last five years, consumer interviews. That is sitting in silos and nobody has ever looked at. Every next year, they do the same consumer research that they have done two years back. Every three-point-o-version is the same as one-point-o-version was. So point is that you have a historical friend of being able to look at it wisely and be able to understand whether you're going this way or you're going flat or you're going down. So the ability to collect the trend out of this interview, what if I tell you out of 6,000 interview every time somebody talked about a checkout page experience, AI will go and summarize all the high moments of those and those aha moments of those and give you into a stitched one-minute kind of a real you plate and you know everything that has happened for last five years about users' expectation of a checkout page. Simple example, but very powerful, very powerful in terms of how it impacts UX strategies in terms of conversion, in terms of whatever example you might want to take. It brings in complexities as well. There is questions on combining data from various sources. Is it a science? Is it art? I had user journeys which said something. I have user interview which is saying something else. I have eye tracking data which is saying something altogether different. How do I blend this together to create an understanding? AI is very capable these days on doing that. And obviously it is an enabler. It is no way going to give a recommendation that essentially it's an enabler for you to look at things in ways which make sense. And then for you to make decision. User trust and acceptance, I think with all the data source, the proof-based inciting is something that is coming more and more. Gone are the days that from a user interview agency will go and pick a small verbat and basis that come and present you final insights. People want to look things lot more holistically. Nuance feedback is something that it can interpret. And these are some of the complexity that is getting answered by AI. Context is something with LLMs that's coming in together. Quick one, summarization of some Jaganish tech that is out there. Content analysis, NLP, national language processing that is at effect. Big time task, that is big time tech that is helping out user research as a process. AB testing and predictive analytics. How many times you would have thought that, hey, I have researched so much of prototypes over last years. Is there a way a system that throws me, hey, this is the best possible prototype for your particular context that can be there? So tech is, AI is pretty much there. It can throw that kind of insight to you. It can be in a beautiful way. A very really powerful tool to go and ask questions and get a lot more insights. Gone are the days that you conduct research. If you would come, you would read, you would present. You want to ask questions like you ask Google Now and you want to get answered now. That's what it is about. Some key difference. Mix method, UX research, 60 million panel across 120 countries. Facial coding, eye tracking voice here. We hold about 17 patents on that. 58 language transcription, translation, all that stuff. And the last is semantic search. The difference between past and now in terms of natural language capabilities, earlier you ask a question. It gives you links as results. Now you ask a question. It gives you contextual, semantic summary like a human is answering. And that's where things, that's a massive breakthrough what large language model or LLM or generative AI has brought to the table in terms of how easy and intuitive it can be for you to summarize a whole user journey, user research is a process. So that's what broadly AI is bringing to the table in the world of user research. And thanks a lot for being a patient audience. Thanks a lot for your time.