75. Unlocking the Power of Amazon Marketing Cloud with Sreenath from Intentwise

75. Unlocking the Power of Amazon Marketing Cloud with Sreenath from Intentwise Speaker 1: Hey, welcome everyone to another podcast episode with Clear Ads. I am so excited today because we have an absolute legend that's joined us on the podcast and it's none other than Sreenath and he has basically been a tidal wave when it comes to Amazon Marketing Cloud within the e-commerce space. He's been kind of a thought leader from the very beginning and yeah, it's just so good to have you join us. Speaker 2: Thanks so much, George. Excited to be here. Speaker 1: Brilliant. Awesome. I just want to let the audience know how I came to know Intentwise and figure you guys out. You had your Ad Optimizer tool for many years, but what made me familiar with Intentwise was I went into a rabbit hole of Amazon Marketing Cloud and just the incredible bridge that I was going to create between DSP and PPC over a year ago. And when I went onto YouTube to try and find any content on Amazon Marketing Cloud, Sreenath, you were the only person that put stuff out, right? And your content was so helpful because I need things done down for me. I need things really simplified and you did just that. You've also interviewed a bunch of people who were in the game and do an Amazon Marketing Cloud and actually, well, really sealed the deal with me and Intentwise. What I found was when I wanted to just learn SQL, because I kept failing the AMC exam, and I had no SQL experience. And you have this incredible feature on the website where someone can just go in and learn how to do basic SQL to get the queries for Amazon Marketing Cloud. So your team were just phenomenal as a, hey, I'm not a client. I'm not paying you anything. Can I please have free access to this tool? And they were like, absolutely. They responded quickly. They answered my questions. And I was like, okay. At that moment, I had like a positive taste in my mouth of Intentwise. I said, okay, I obviously need to talk to these guys at some point. And then we kind of obviously met at conferences, etc. And we kind of got to where we are. Before we start at the beginning, I'd love everyone to know Sreenath, for those who are in the e-commerce space and don't know who Intentwise is, don't know who you are, give us just a background of where the company is now and what it is you do. Speaker 2: Awesome. First of all, I love that story about getting to our site to learn SQL. You know, we just took this path of education as our means to build our brand and brand awareness, and we enjoyed doing it. So I'm so glad You benefited from it. Just to tell you a little bit about Intentwise, I think, broadly speaking, the way we frame the problem we're solving is, if you're a brand or an agency, you know, and operating in the Amazon, Walmart space, there's just this incredible amount of data coming at us. However, it often tends to be fragmented. You still need to stitch it together, make sense out of it to make smart decisions. So our goal as a company as we have framed it is how do we bridge that gap between this fragmented but abundance of data to real actionable insights and actions, right? So that's our goal. We truly at the core are an analytics company. We have three solutions with that as the kind of core DNA. One is an ad optimization platform. That was our first solution. That's where we came to the market several years back. The other is a data foundation layer where we completely automate the collection of all the data you need from Amazon and Walmart. You have full ownership of the data, all the way to helping you build visualization and reporting, including for agencies, a white label solution. So a full-blown data and infrastructure solution which forms the foundation. And then lastly, as you talked about Amazon Marketing Cloud, Amazing asset or solution that Amazon had launched, but extracting value from it is hard, right? Like you have to know SQL, you have to know a lot of like nuances. So we built a solution. We actually co-developed it with Amazon that sits on top of AMC, really designed to help the brands as well as the agencies. We extract value from Amazon Marketing Cloud a whole lot quicker. And then, of course, we wrap that with, you know, services like custom queries and audience creation and a ton of education around how to get the most value from it. But those are the three pieces that make up our platform. They can all work together or they can work independently. So that's Intentwise. We work with brands, we work with agencies, agency partners like you are so critical for us. Great relationship with our agency client partners. So that's on the Intentwise side. Happy to share with you my personal background, but you know, if you have any questions, I'll take that. Speaker 1: Yeah. So I want to kind of step back Sreenath. I want to go back to like, okay, as a young child growing up, like what was it that you wanted to be when you were older? Speaker 2: Oh, great question. Look, I think just a little bit of background on me. You know, I grew up in a typical kind of lower middle class family in South India. My dad had a government job for his entire career. Entrepreneurship was the last thing anyone would have thought about in my extended family circles, too risky. So that's how I grew up. But I'd say, you know, always interested in maths and sciences. I thought I'd be a physicist or a scientist somewhere, you know. But. As it happens with everybody, like things change, and I did get an engineering undergrad degree, an engineering master's degree. Always had been fascinated with data. But, you know, I never thought I'd be here. It's like, you know, how Steve Jobs at Apple says, like, no matter what you do, the dots connect. So it's a few different things I did that led to this particular moment where, you know, I had the opportunity to be a founder of this business. But, yeah, so, you know, I got my engineering degree, but really worked in the software space all my life. Prior to starting Intentwise for 10 years almost, I was working in a public company where I had the opportunity to run a data team and an online marketing team with sizable spend, close to $100 million of online spend. But that's the DNA, that's the background in which I learned a lot of stuff, a lot of battle scars around data and online marketing. Which then led to the formation of Intentwise. Had I looked back 30 years and guessed this is what would happen? Absolutely not. Speaker 1: Okay, so let's step back here. So let's go to the first stages of Intentwise. How did the company come to be? How did those conversations, those first early thoughts come? Speaker 2: Great question. So I had a day job at Orbitz, the online travel company. And this entrepreneurial bug had bitten me a long time ago and I had tried to start two businesses very early after grad school. Failed, right? One of them was in the dot-com boom time and then the bust. I was part of the bust. But then, you know, things happened, life happened, you know, I got married, I had kids, you know, was in steady jobs for a while. And then this whole entrepreneurial itch was always there and there came a time where like, you know what, this is my moment. I'm going to do this. I talked to my wife. It was like, why not? And I had this brilliant idea that, oh, I've been in the travel space. I'm going to go help a whole bunch of hotels with digital marketing. I quit my job, started to pursue that opportunity. Very quickly I realized super commodity market. I have no idea how I'll be different. I don't think I can get anywhere. I probably talked to a hundred hoteliers. And I was just sitting around trying to figure out what the heck do I do. Back then I was in Chicago and I had friends who were either advisors or new companies that had big Google AdWords spend. And they all knew that I had worked on Google AdWords spend at my job. They're like, why don't you come help these guys? And I remember distinctly going to one of these clients and I said, okay, give me a dump of your search term report of Google Ads. Okay, let me figure out what is going on. They did that. And I'm looking at all the data and there's a column in the end that says monthly agency cost. Okay. Okay. Consistently 10% are spent, 10% are spent, 10% are spent. I had no, literally, I'd never spoken to an agency before. I did not know the agency model. And I'm looking at that number and like, I went back and asking, hey, what is that last column? Like, what is that cost? Oh, that's the money we pay an agency. That's the money you pay an agency? Really? Speaker 1: Okay. Speaker 2: How about I get 10 of these clients and that's a business? So I actually started a Google AdWords agency. Speaker 1: No way. Speaker 2: Yeah. And we quickly got up to about 5, 7, 8 clients. Business was good and all that. But I knew, me and my co-founder Raghu knew like, we were product people, right? We are not an agency services business kind of people. We are constantly looking for opportunities. One thing we found was that there's always gaps between the intent in the search term data and the relevance of the ads and the landing page that is delivered to people. For example, in the travel space, if you searched I'll give you a Chicago example. If you searched hotels near Wrigley Field in Chicago, you will get a Google ad on Google when you search. But it doesn't matter. Most clicks, if you go to the websites, they show you Chicago results. They're not showing you results next to Wrigley Field. So that intent gets lost. Our big idea was we will mine the intent, ensure that the entire experience is relevant, and with that, we'll optimize spend. That was our big thesis, which is why we call ourselves Intentwise. My best example of this is we did the analysis of search term data. There's a restaurant equipment parts company in Chicago. And we realized that they're spending a ridiculous amount of money on a phrase match term called Jackson Control. Jackson is an OEM that makes controllers for, I think it's ovens or something like that in restaurants. They did not know that a bunch of their money was going to people searching for Janet Jackson Control music album. So when we wrote this little algorithm, it automatically detected that and showed it to them, right? But we still were trying to figure out, okay, how do we make a business out of this? But accidentally at that time, coincidentally, I was talking to a toy brand in Chicago and the CMO, trying to get their business and the CMO goes, I'd love to work with you guys, but guess what? I'm moving all my money to Amazon Ads. And my reaction was, what? Amazon does ads? Speaker 1: It's so interesting because the same thing happened to me. I was like, you have a what account? An ads account? You're getting what kind of returns? That's not true. Looked into that account and I was like, this can't be real. Their return is ridiculous compared to what we're getting through Google Shopping. Speaker 2: Same, same thing. So I was like, okay, can I shadow you guys for a couple of months? So I was in their offices and trying to understand what this is. And really that's when the aha moment happened for us saying, oh my God, like there is a real platform product opportunity. We went back, we literally completely pivoted, told all our Google AdWords client, we're moving on to a different model. Raise some friends and family's money, double down on building out the product. And that's how our journey started back in mid 2018. And by then we did not even have Amazon API connection. So we had to find a way to get API connectivity to build anything. But anyway, so that's where our journey started. But we are still Intentwise. Speaker 1: It's fascinating. I've really enjoyed just hearing the kind of the kind of the first baby steps of Intentwise. And I kind of want to move a little forward now and just figure out There's obviously been so many advancements with the data that we can get from Amazon over the last two years. I mean, our minds were blown with the search query data, but now it's like the integration of AMC into just having access to it, but now also coming into sponsored product ads. 2025 is going to be a huge year with all of this information, but I'm sure there's going to be so many people listening to this. Right, they're asking the question, how on earth can I use this data to my advantage to make better decisions with all of the campaigns that I run? So someone who is in the day in day out data, how would you answer that question? Speaker 2: A great question. I think to your point, if you look at Amazon's own journey, if you dial back three years, right, and you ask the question, what data do you get? You get data from sponsored ads, DSP, and what at that time was called MWS. I mean, you could download data from different places, but from a systems perspective, those are the only three things you could get. Dial forward to now, You've got real-time data coming from Marketing Stream and Rapid Detail Analytics. You've got granular shopper and event-level data coming from Amazon Marketing Cloud. And then the amount of, basically, Amazon Marketing Cloud has now become the center of all ad measurement for Amazon. Every time they add a new ad inventory, the measurement happens for Amazon Marketing Cloud. So I think what that has done is now you can actually answer questions like you couldn't answer before. Like, hey, how is my DSP spend influencing my self-sponsored ad spend performance? Or how can I go find those people with a very specific action that they have taken that I can go after? Or how can I exclude my direct-to-consumer customers that I already have and I don't want to be spending my Amazon Ads money trying to educate them about me? So what Amazon Marketing Cloud has done is you can now answer questions you couldn't answer before and find and create fine-tuned audiences that you couldn't target before, right? To the question of how can you use it, I always say this about Amazon Marketing Cloud as a simple way to think about how you get more and more value is the value you're going to get from Amazon Marketing Cloud is a function of how good a question you're asking. But how do you ask questions unless you know what is inside Amazon Marketing Cloud? The number one thing I suggest to hands-on doers is get certified so you know what data sets exist. And number two is, what I say is like a list of questions that would be great for you to answer that you can activate and do something about in your ads. And I call that a roadmap. Consistently have like a top five list. What do you want to know that you didn't know before? How many of my brand searches actually had a non-brand search before? Because we think about brand searches often as defensive and that's it. Maybe you want to change that. What is my real lifetime value on these customers? Because soon you're going to get five years worth of data on every shopper from Amazon in terms of purchase history. So there are a number of things you can answer that I couldn't answer before. It is a matter of knowing what those are. And being disciplined about probing and experimenting through that. I mean, there's a number of use cases that someone like us can share with you where clients have had success. But I also know that one size doesn't fit all. Right. So you're going to have to go through a bunch of questions and constantly probe this data set to get what you're looking for and also experiment. Speaker 1: Okay, can I just stop you there Sreenath, because you stopped it really interesting earlier. You said the first point is get accredited. Can you just expand on, okay, where on earth do you go to get accredited and find this information about Amazon Marketing Cloud? Speaker 2: Yeah, honestly, if you just Google AMC Certification Amazon Learning Console, you'll find the link. It's pretty straightforward. Happy to share the link separately as well. That's where I would start. Speaker 1: Okay, perfect. And because you... You are just in the trenches of all of this and you see so many ways, different ways that AMC is implemented on different accounts. Can you give me some examples, kind of a range of examples of how that data is being taken, audiences have been created and then being utilized to allow these brands to flourish? Speaker 2: Yeah, I'll give you absolutely. So here's a simple example. In fact, we have published a case study with Amazon. We have a seller in the publishing industry as a client. They have about 2400 products. And their big interest was how do we actually get new to brand, more new to brand customers. And when we looked at all the products, it turns out that about 12 of them, for whatever reason, drive much better new to brand acquisition than the rest. Those products, you know, they isolated those products, reallocated budgets for those products, even if the ACoS was higher than overall account ACoS that they were trying to shoot for and to drive up much more new-to-brand acquisition, right? So that's a classic example. That's one. There's a number of others, like for example, The high value customers and audience is a classic one. Now, you can define high value whichever way you want. Like, hey, I want to create a group of people that have spent X amount of dollars with me over the last 12 months or purchased X number of times with me over the last 12 months. And I want to create that audience and perhaps a big boost in my sponsor product or sponsor brand campaigns or target them with my sponsor display campaigns or run very targeted specific ads on DSP. So that is a classic example of where a number of our clients are benefiting from. I'll give you another slightly more complex example. So we have someone in the vitamins and supplements space who has a reasonable D2C presence. What they have done is they have uploaded their first party data, which is their customer data. So that they can see the matching of, okay, where is the overlap between my direct customers versus Amazon? They also uploaded it with a few attributes such as, okay, what is the last product they bought? So that they can create a pool of people to whom they can upsell a particular product, right? They've also used that list or an audience pool to say, I want to exclude them from DSP exposure because I want to go after net new people. I mean, I can go on and on, honestly, but these are all and if you notice some of these you can take and just try to execute and see if there's an opportunity across all your accounts. Right. But some of these will be nuanced and very specific to the business. If subscribe and save is important to somebody, you could go create a pool of audience of People who are buying from you but are not opted in to subscribe and save and run very targeted ads towards their audience. So it'll depend client by client. There's of course some standard queries like path to conversion, time to conversion and a few other metrics. But it's a matter of sitting down and based on the client's business, figuring out Where are the opportunities? And it's good to make a list of those potential opportunities because at the end of the day, you're starting with a hypothesis a lot of times. And then go after them one after the other and see what works for you. Speaker 1: Amazing. So many examples. This is great. I'm going to ask you a question now that gets thrown around quite a bit, which is, I don't see incremental growth in my sales of Amazon DSP when I scale spend. How does AMC come into the equation to help solve this problem? Speaker 2: Great question. I mean, there is I look at it two ways, right? So there is opportunities to do things differently in the execution of the campaigns. There's opportunity to measure differently after the campaigns are run. Because of AMC. Okay. Let me talk about execution, right? There's a concept called in-market audiences as you know in DSP right or lifestyle audiences. What you did not know before is what is my penetration with my DSP ads in this in-market audience. Okay, if there's a pool of 100,000 people, how many of my DSP ads are reaching? In fact, here's what I want to do. I want to actually exclude people who have already reached and go after net new people in my in-market audience. That is a great way to get your brand in front of people that don't know your brand. So that's an example of things you can do from an execution perspective that AMC allows. Another very simple example. You can do a histogram of number of exposures of an ad versus performance of that audience and say, you know what, I want to set my frequency cap to 8 or 10 or 12. And by the way, without AMC, you had no basis to figure out what to set the cap as. People generally had a, you know, oh, it should be 8, right? But there's no basis for it. AMC gives you the basis to determine what frequency cap should be. So, there is things you can do from an execution standpoint. And of course, from an analysis standpoint, there's a lot of things you can do. Like, okay, these DSP exposed users, are they driving organic sales? Is there an overlap with my sponsored ads? And when DSP combined with sponsored ads, is that better than just sponsored ads? There's a metric we are looking at for a brand where we are saying, like, how many Shoppers who have never bought before are searching my brand's name. And they've only been exposed to DSP because that's another way to measure, is my brand awareness growing because of DSP, right? So I think both from an execution and measurement standpoint, there are things AMC allows you to do. And then we didn't even talk about custom audiences, right? There's a bunch of standard audiences in DSP, but you can craft audiences in a very fine-tuned way through AMC that you cannot do on DSP. I want to go after people who showed interest in this particular product or added this product to a wish list or a gift list. This is where I think getting certified and knowing all the signals that exist will help you determine what kinds of audience pools can you actually create with AMC. Speaker 1: This is great, thank you Sreenath. Another question I have for you is, there's going to be a bunch of people who listen to this that are not doing DSP. They've just got sponsored product ads running and now with this rollout of AMC within the ad console, what are some practical things that they can do without DSP to leverage this data? Speaker 2: Look, I think, I mean, the two, just for the benefit of the audience, there is two features that got launched around audiences in sponsored ads, right? So one is You could do bid boosting on audiences that you can create in AMC and push into sponsored ads. So you can do boosting of bids and on sponsored display, you can target very specific audience. So one of the examples I suggest is, think about it this way, right? Let's take your top performing non-brand keyword in your account. Okay. My guess for anyone listening is that your click-through rate on those ads will not be higher than 2% at best. Right? Which means 98% of people who are searching your super high intent keyword are not clicking on your ad. How about we create a pool of those audience and go after them in Sponsored Display because they have super high intent right now. That's an example. So you can create different types of audiences and of course target them through Sponsored Product, Sponsored Brand and Sponsored Display. So that's one example of many where you can create audiences and target them. And also on the measurement side, If you look at sponsored ads today, you have new-to-brand metrics on sponsored brands and sponsored displays. You don't have them in sponsored products. But AMC allows you to get new-to-brand metrics on sponsored products also. More importantly, With AMC, you can de-dupe new to brand across all these campaign types and say, new to brand from an account perspective, I don't care what campaign, right? How many new to brand users is your sponsored ad spend acquiring? Is a metric you can track and see if that's going up or down, what is going on, right? So from a measurement perspective too, you can do things that you couldn't do before. Speaker 1: Amazing, that's incredible. Thank you Sreenath. I want to kind of pick your brain as well because I think there's a lot of talk happening around AI and there's been so much going on in the news recently. I remember we briefly discussed this last week when we spoke but I would love to understand where you think this is going and some of your predictions as how this is going to be successfully adopted by companies in this industry. Speaker 2: Great question. Honestly like it's an area that we are deeply involved in doubling down on but let me just begin by perhaps clarifying some things that get lost in the marketing and the noise around this right. So number one it's important to understand what fundamentally changed when ChatGPT got launched a couple of years ago. See, AI as we define it has always existed, right? So, AI or machine learning around structured numeric data. The weather forecasting people use it in our space. When everyone says AI when it comes to bidding, it's really machine learning. There's a known set of algorithms that get used for bid management and we have AI driven bidding algorithms too and we have had them for a while. That is not new. What is new is that AI start to become extremely valuable on super unstructured data also. Texts, images, videos, audios, Understanding it, generation of it, translation of it is where the advancement happened with GenAI. So let's start there. The other thing I want to clarify is it doesn't matter which GenAI model you look at. We're fascinated when you ask ChatGPT to write a poem, it writes a poem. You ask it to create an image, it creates an image. It does things that we never thought were possible. So I think our human mind is just completely enamored by this sudden shift. But in the process, there is one more leap we seem to be making, which we shouldn't, which is somehow we've lot of us in our, we've consciously or subconsciously started to think that these models have reasoning and logical ability. That's a mistake. They don't. What these models have is an ability to find patterns in historic data that has been fed to them. And so it's important to recognize that which also means that the output has to be vetted and verified if you're seeking accuracy. If you're not seeking accuracy, it is amazing. Ask it to write a recipe, it'll give you a recipe. You have no idea whether it's a right or wrong recipe. Ask it to give you a plan for a city, it'll give you a city, a plan. You can't sit and judge whether it's accurate, right? So for those use cases, a content generation too. You ask it to summarize a paragraph, like there's no concept of right, wrong, right? So those use cases are perfect, right? When you start seeking accuracy in the output, that gets more and more difficult, which is why if you look at 90% of applications, they don't center around accuracy. LLMs, when you ask it a question, it takes time to respond. So a lot of use cases also are not real-time. They are async. Like something is taking notes as we talk. That's an asynchronous application. It's going to take some notes and send you a transcript later, right? So it's great for, I think LLMs fundamentally, the innovation is in the fact that search has gotten better, translation has gotten better, and summarizing and generation has gotten better. That's what has happened. What does it mean for, I mean, I view myself as a, you know, analytics and software company. Here is a fundamental shift I think will happen. What LLMs have done is software can now understand language a whole lot better. And that has fundamental implications. The way you, the user, has interacted with software for the last 30 years is about to fundamentally shift. Because I don't have to click around a bunch of buttons to accomplish a task. I can just speak. I can just type. That changes the game. Let me just paint the picture a year from now. You use the Intentwise platform. Maybe you have a Slack channel internally. Imagine that Intentwise is a bot on your Slack channel and you ask it all kinds of questions and it responds back to you and you never even have to log into Intentwise. That is the future of software. There'll be a lot of use cases where you can extract value. You can just write AI at intentwise.com. You can just write emails. Hey, Intentwise, do an audit on this account for me. An hour later, you get a PDF with a very detailed audit on an account. And then you ask a question, you know what, like dig deeper into sponsored product, product targeting campaigns and tell me what's going on, what changed. He'll come back with a what changed analysis. Like again, I'm talking about features that exist today, but your interaction with them will be different. Point being, Software, as we know it, is about to shift completely, and the way users interact with it is about to fundamentally shift. Now, it's not easy to make that transition. That's where we spend a lot of time. We can talk about that. But I would frame it as the way we use software will fundamentally shift, and a lot of it is based on the foundation that software understands language very well. Speaker 1: Amazing. Okay, that's great. And you can move this next question outside of kind of AI and ChatGPT and all the advancements that's been made in that area. But what is it that you see in the coming months and maybe next few years that scares you? Speaker 2: I was at Jensen Huang's keynote speech at CES, right? I walked out of that session saying I'm half scared and half excited. He showed he picked up this box and showed us this thing is maybe five inch by five inch by two inch and he said that's the that's a supercomputer. Okay. We are going to a place where the amount of processing power and storage ability is going to explode. And so and the large language models LLMs are going to get better and better and better. So I look at this wave of Gen AI especially for software companies. It's a tidal wave that is on its way. Either you get hit sideways by it or you better you be ready and ride it. Yeah, we don't want to get hit sideways. That's my number one concern, but also an opportunity to be honest. So that's number one. And then the second thing I would say, see, we operate in a place space where we are the layer right in between the underlying Amazon and Walmart platforms and the end users, right? Brands, agencies. The constant battle for us is trying to figure out how do we add incremental value on top of these platforms in a way that that platform is not going to offer the same capability. So that is the other big risk that always exists in our space. This is also why while we started as an ad platform, we've evolved to serve other needs of agencies such as yours, such as like you've got reporting needs, you've got data needs, and so constantly evolving in a way that we stay complementary to these core platforms because otherwise Everyone talks about competition in the ad space. This will sound weird, but I personally rarely looked at the known named software companies as my competitors. That's not who I worry about. I worry a lot more about the actual platforms like Amazon and Walmart. But that also has kept us on a path where we continue to offer differentiated capabilities from the platform and we're staying complimentary. So that would be the other risk we have to keep front and center. Speaker 1: Fascinating. Okay. Is there anything that I should have asked you that I didn't ask you that you think would be very important for people listening to know? Speaker 2: Great question. I'm thinking about this. I think perhaps the one area we have not talked about is how commerce itself is evolving, right? I will share two different anecdotes and perhaps bring those together and, you know, who knows what the future is, but it'll at least help us think about where this might be going. And it has implications for Everyone offers in this ecosystem. On one hand, my parents live in India. So I go there and a big chunk of my team is there. So I go there every December. it's 4 a.m my uncle is doing some ritual at his new house 4 a.m because that's the right time and they were shot on lime for the event that they are doing lime right and his son clicks a couple of buttons and the lime are delivered within seven minutes to the door Okay. People sit down for dinner and realize, oh, we don't have yogurt. Someone presses a button, yogurt is delivered before you get to the point of eating yogurt on the dinner table. That is quick commerce in places like India. Okay. Speaker 1: That's unbelievable. Speaker 2: US and Europe are yet to see that. Okay. So there's that. The other anecdote, which is actually from last week, is OpenAI launched an agent called Operator. You have to pay to get a subscription. I went there and I said, hey, I'm looking for cordless vacuum cleaners. Go find me, get me three recommendations for products priced under $300 and four-star rating or above. And I let it go. It did its search. It created a video of everything it did and sent me three recommendations. This is an agent working for me and shopping. And then you combine that with Amazon's Rufus, where you can interact with and figure out what you want to buy. So you put all these together. I have a sense that in about 12 to 24 months from now, the way we think about shopping, the way we think about ads, like what does a sponsored ad mean if I have an agent shopping for me and getting me what I want? So we are going to see some pretty transformative things and we haven't even talked about social commerce now. There's a TikTok ban and there's stuff happening there, but social commerce is here to stay as well. And as there's generational shift, The way people shop and products get made and delivered is absolutely going to change the pace. I do not know yet. But, you know, like we just had to keep one eye on on on that future state. And generally, from what we have seen over the last couple of decades, the new innovations come through much, much faster than before. Just keeping an eye on where commerce might go is a super important thing. Otherwise, it's easy to get blindsided. Speaker 1: Amazing. Sreenath, this has been incredible. I've really, really enjoyed not only hearing about your kind of starting point with Intentwise, but also Some of those things that excite you about the future and how we can actually take advantage of this incredible data that's been gift wrapped for us. I really appreciate your time coming on and doing this. I really appreciate just how open you are. You share so much across the different platforms. I'm sure many of the listeners here have heard you talk about kind of AMC, AdOptimizer and all of the ways it could benefit their businesses. If anyone wants to reach out to Sreenath, he's found that Intentwise, there's going to be links included into the profile. Thank you so much for joining us and I hope to see you again not too long from now to share some more insights of any updates that you've come across. Speaker 2: George, thank you so much. I appreciate all the questions. I totally enjoyed the conversation. I hope to see you at the next conference, perhaps. Speaker 1: Yeah, sounds good.