18 Burst results for "Dr Duane"
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Yesterday we talked about AI and machine learning's innovation in mar tech. Today we're going to wrap up the conversation and talk about how science is changing the advertising industry. This interview is brought to you by higher logic. Are you ready to move beyond transactional relationships? Want to build more engaged, meaningful connections with your customers? Well, higher logic vanilla is the leading online community platform for growth minded companies. The vanilla platform combines the power of communities, Q&A, and knowledge bases to increase retention, reduce customer support, and maximize revenue, with higher logic, vanilla, your customers can seamlessly engage with their peers to get answers to their questions, provide feedback and become an advocate for your brand. So join more than 3000 companies representing 350,000 online communities with more than 2 million users using higher logic vanilla to get started today, go over to higher logic dot com, or you can use mar tech pod dot com slash vanilla. That's Marta pod dot com slash vanilla. And also since you're into marketing podcasts, check out marketing made simple, hosted by doctor JJ Peterson, marketing made simple, brings you practical tips to make marketing easy and more importantly make it work. And JJ is a brilliant marketer and a friend of the mar tech podcast. So to hear tips like how to write a captivating speech, or how to market yourself into a new job, search for marketing, made simple in your podcast app. We're going to HubSpot dot com slash podcast network. Okay, on to today's episode. All right, here's the last part of my conversation with doctor Duane veran, founder and CEO of media science. Doctor Duane, welcome back to the mar tech podcast. It's becoming a regular thing. It's becoming a regular gig then. It's kind of how we roll around here, you know? We get you once in the next thing to know is three days later, you're still talking. I appreciate you coming back and I'm excited to wrap up our conversation so far we've talked about artificial intelligence and machine learning. We talked about neuroscience and how it's changing, consumer research, how we can understand what some of the signals are and why people react to specific ads or media. And yesterday we talked about more artificial intelligence machine learning how that works and how we're applying it into more attack. Today I want to specifically focus on science and advertising. It seems like there are some fundamental changes for how people react to advertising. Also some of the formats of advertising. So as we start to think in a different capacity about how we should advertise, where our advertisements should go, talk to me about how some of the technologies that you have worked with are changing marketers thoughts on their ad buys and ad strategies. I think the biggest change that we've seen over the course of the past decade or so is that especially with the premium advertising like TV ads, et cetera. We've really gone from a one trick pony, which was the 32nd commercial to a universe of infinite possibilities. When it was one format, people had to fit whatever their needs were into that medium. So it was just a factor of these are the rules. You have a 32nd commercial, figure out how to take your marketing objectives and squeeze it within that template. Now we live in a universe with, like I say, infinite possibilities. So the question becomes, what should it ran to? And there's a lot of possibilities out there. So you need to have some way of really grappling with what your objectives are in the first place, what you're communication objectives are, and how to measure for success. And that's where science can really make a difference. One of the things that I've always said and I have not studied this, but I say it anyway because it's my job as a podcast host is that the formats of advertising are changes because people are exposed to so much more media. Now, I know that I have my phone in front of me and that seems to be enough information for me to say, well, I consume more media. I see social media. I still watch television. We got the radio on. There's podcasts that have ads inserted to them. There's all sorts of different formats. Do you have a sense of whether people are actually engaging with media more because the assumption is if there's more media than people are getting tired of being advertised to. And I actually haven't seen the science. I did not read the research. It is an inference, not any sort of study that I've conducted, try to understand if people are actually consuming more media and more advertising impressions. Oh, no, there's really good research, which shows that the consumption of media has grown exponentially. I mean, there's no doubt about that. And of course, if we're getting exposed to more media, we probably are getting exposed to more ads as well. So in that sense, the challenge for a brand to stand out is greater than it was before. But by the same token, there are a lot more ways that you can engage. And that's an opportunity for a brand, particularly relative to its specific communication objectives. So that's where it comes again. I say this thing about communication objectives. The ultimate job, of course, of marketing is to help promote sales. But that's too big an objective. You have to break that down into a strategy for how you're going to do that. And you want to know whether your strategy is effective. And that really comes down to a series of communication objectives. So now the question is, do you know whether or not you are delivering against each of those objectives and historically we don't know because we're just trading on these common currencies that we have like ratings or something. But now we have the ability to get much, much, much more precise. So I use the example a couple of days ago on our show about the Mars study that we did. A lot of Mars ads are about humor and you want the ads to be funny and you want them to be funny and specific moments in time during the ad. Well, now you have a very, very precise measure of whether it was funny or not funny. And I will tell you that our research says that whether it's funny or not does have a difference in terms of whether it's also going to be effective. So it does translate ultimately to sales. But now you can actually measure scientifically whether the ad was funny. And a brand needs best in class measures for each of their objectives. And they're not all going to be the same for different brands. So people are going to be trying to doing different things. And it's just about more accurately measuring that. So Mars is trying to measure funny with their advertising if it's funny. There's a positive association with the brand and hopefully that has an impact on underlying sales. When you think about B2B measurement, when you're thinking about reaching people in a professional setting, obviously humor is not something that is devoid, but it's also not something that we rely on from B2B. Should we all start making our ads into long running jokes or is there a time and a place where you need to think about other measurements other than obviously humor to get your ad across, get your message across? What are some of the advertising measurements that you've seen for the B2B market? The most important, of course, is attention. Or it's not actually attention that you should be interested in. It's inattention or the absence of inattention. It sounds like a subtle difference. More attention is not necessarily better, but no attention is definitely fatal. So you want to get through the threshold of getting enough attention for everything else to matter. And if you don't pass that threshold, nothing else matters. So attention is a great example whether you're talking about B2B or whether you're talking about B2C, everybody needs attention. And if you don't got it, you don't got an effect. Yeah, I mean, without any attention, no one's going to understand, have awareness, recalled. Nobody's going to think about your brand. You're dead on arrival, but there is good attention and bad attention as well. You know, it seems like maybe this is different in the political sphere where sometimes all attention is good attention. When you're a marketer and you're starting to think about creating good buzz as opposed to just getting out there, what are some of the ways that you can evaluate what works and what doesn't? It's not that there is good attention. There's a tension and there's no attention. And then there's something else that comes in after attention. And that's really my point. Attention is the threshold. If you don't have attention, you're not going to notice what goes on. But just because you have attention, doesn't mean that it's going to be good or bad. It's everything else now that starts to kick in about how people start to evaluate your message. So it could be that you're looking at whether that is positive or negatively oriented. So are you getting this attention, but then are you seeing, we call this valence, which is whether things are overall positive or overall negative. Are you getting positive or negative valence? Are people reacting in a favorable way towards the brand? Or are they acting in a negative way towards the ground? Again, that's a big threshold across that threshold. And now you're on to the next one. And so layer by layer, you just work to delivering against your objectives. Is someone paying attention to your message? Do they think of it positively or negative? What's your valence? What comes after valence? It depends by brand. For example, you could be the kind of brand that has a very emotive message, or you could be a brand that has something that requires cognitive function. You know, that requires you to think about what's being set very different scenarios. So really once you go past the attention pressure, everything else now is going to be brand specific rather than generic. So you're going to be talking about that brand. So for example, maybe this brand is looking for empathy. That's something very different to a brand that's looking for. You thinking about their claim. So you really start getting into specific metrics around what the objectives of a brand are rather generic measures about overall, whether something's got attention or didn't get attention. Time for a one minute break to hear from our presenting sponsor, HubSpot. As marketers were driven to solve problems, and there's problems that can be fun to solve like brainstorming a brilliant new ad campaign, and then there are problems that can be frustrating. Like when your ad tech doesn't work. No matter what the problem is you're trying to solve, the HubSpot CRM platform is easy to learn, use and love because it's designed for the ways marketers actually work. With customizable tools that you can add or subtract as you grow, HubSpot is ready to help you squash the endless problems in your business. To learn more about how HubSpot can help your business grow better, go to HubSpot, dot com. Now, when you think about measuring the effectiveness of your advertising, we talk about understanding attention, understanding valence, understanding, you know, some of the other emotions that you're generating, those seem like a very difficult thing to measure without a lab. And B, things that you can't necessarily dictate in terms of business results. My advertisement was successful because this many people more came to the website and bought something is how most advertisers think about ad performance as opposed to my ad was successful because people think of my brand as fun as light as beautiful as serious, the emotions that come out of it. How do you bifurcate the understanding of why someone is reacting to your ad as opposed to the underlying performance or should you actually think about them separately? You should think about them in a related sense, but you should think about both. They're both important. Whether you got sales or not very important. I mean, there's no point in delivering against your communication objectives. If those communication objectives don't translate into sales. So that's important. But also whether you're moving parts are kind of like getting you where you want to go is also important. The problem that we have is that you don't want a brand to have a strategy that is speculative. A branch should have a very deliberate strategy. If it's trying to trigger an emotion, it should have a reason why it's trying to trigger that emotion. It should measure whether it's successful in doing that, and ultimately that will help it know whether it's strategy is on target. It could be that it's strategies off base. That's not the right strategy that triggered that emotion doesn't translate to sales. But you won't know if you're not accurately measuring for those objectives. So the business of measuring these objectives is really about making sure not only that you're delivering against the objectives, but ultimately as a consequence that your strategy is actually on target. When you think about the science that is being applied across customer research across martech across the advertising industry, it seems like there is a practical way to measure business performance that most of us rely on. And then there is the understanding of why give me the pitch of why marketers should take a more scientific approach to understanding the why behind their marketing efforts as opposed to just focusing on the performance. So sales behavior is oftentimes a very tactical way of looking at what we're doing. You want to inform your larger strategy. You want to have a theory about not only a theory. You're also trying to build a brand and your value, whatever your product is, the portion of your value that is being delivered by your brand is far greater than the value that's actually being delivered by the physical product itself. Branding is the real estate of the mind, and it's where you sit in a person's psyche ultimately. And you're not going to get that by the sales behavior. You're not going to know where you are situated and how people's situate you in their minds relative to your competitors. So you have a very short term very short lived strategy if you're just living on sales alone. It's the classic problem that a brand has where they may discount their product in the store and get some sales. But they're not going to make profit that way, is that really where they want to position themselves. So in the science and in the art of really understanding your brand, who you are, what works for you, what connects with your consumers, you're not going to get there by sales data alum. You're going to need to understand why you got the sales. And you're not going to be able to get there unless you have some measures of those moving parts within that. So that's the reason why I think again, it's not enough to just have the sales data. It's not enough to just have the moving parts. You really need both of those to come together to get the full picture of what's going on. It's a little bit like a thermometer in a room. You're getting the overall temperature in the room. So you can say, oh, it's hot or it's cold or whatever, which is great. But you don't know why it's hot. You don't know why it's called, and you're not going to be able to adjust the temperature ultimately. If you don't know what it is that's contributing to it. So the moving parts really go together. You can't change the temperature of the room just by increasing the thermometer. You need both. You have to have the moving parts and the performance. My fundamental belief about marketing and our job as marketers is to understand who our customers are and try to deliver marketing messages, products, and services to those customers that make their lives better to solve problems and create value for them. And without a deep understanding of who your customers are and what motivates them, you can never really accomplish that, taking a scientific approach to understanding not only who your customers are, but what drives their behavior is something that will make you a better marketer. At the end of the day, we're all held to the end results, but our job is to understand who the customers are and using services like media science and technologies like machine learning and artificial intelligence to not only better understand our customers to better target them, is just going to make us better marketers. Dwayne and I appreciate you coming on the show and walking us through some of the more advanced technologies you use. Thanks for being our guest. Always the pleasure of that anytime. All right, that wraps up this episode of the mar tech podcast. Thanks for listening to my conversation with Dwayne veron, the cofounder and CEO of media science. If you'd like to get in touch with Dwayne, you can find a link to his LinkedIn profile in our show notes. You can contact him on Twitter where his handle is Dwayne veron. Or you could visit his company's website, which is media science dot com. And also a special thanks to HubSpot for sponsoring this podcast, no matter what problems you're trying to solve, the HubSpot CRM platform is easy to learn, use and love because it's designed for the ways marketers actually work. With customizable tools that you can add or subtract as you grow, HubSpot is ready to help you squash the endless problems in your business. To learn how HubSpot can help you grow better, go to HubSpot dot com. A special thanks to higher logic for sponsoring this podcast. If you're ready to move beyond transactional relationships to start building more engaged, meaningful connections with your customers, then join more than 3000 companies representing 350,000 online communities with more than 2 million users by using higher logic vanilla, the vanilla platform combines the power of communities, Q&A, and knowledge bases to increase retention, reduce customer support costs and maximize revenue. To get started, go to higher logic dot com, or you could visit martech pod dot com slash vanilla. That's mar tech pod dot com slash vanilla. Just one more link in our show notes I'd like to tell you about. If you didn't have a chance to take notes while you were listening to this podcast, head over to martek pod dot com where we have summaries of all of our episodes and contact information for our guests. You can also subscribe to our once a week newsletter and you can even test your topic suggestions or your marketing questions, which will answer live on our show. Of course you can always reach out on social media. Our handle is mar tech pod POD on LinkedIn, Twitter, Instagram, and Facebook, or you can contact me directly my handle is Ben J chap. And if you haven't subscribed yet and you want a daily stream of marketing and technology knowledge in your podcast feed, we're gonna publish an episode every day this year, so hit the subscribe button in your podcast app, and we'll be back in your feed tomorrow morning. All right, that's it for today, but until next time, my advice is to just focus on keeping your customers happy.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Work. No matter what the problem is you're trying to solve, the HubSpot CRM platform is easy to learn, use and love because it's designed for the ways marketers actually work. With customizable tools that you can add or subtract as you grow, HubSpot is ready to help you squash the endless problems in your business. To learn more about how HubSpot can help your business grow better, go to HubSpot dot com. So when we think about how machine learning is affecting the mar tech industry, I understand the idea that you need to find a dataset and point out what the true target that you're pointing at, I have some data about who's downloading my podcasts. I have a good sense of who I want to target. I can't just build machine learning to turn them into subscribers or turn them into revenue. I need to use a series of tools that are connected together. I've got to figure out who to target. We're using machine learning algorithms that Facebook and Google and all these other companies have put together to help us drive traffic, and then I'm getting someone to my site, and there's potentially more machine learning that goes into site optimization. Talk to me about where the actual mar teching is happening. Where are people using these advanced technologies to do marketing successfully? Well, everywhere, everywhere. I mean, you mentioned at the outset. The big players in the game and how they optimize the content. That's how machine learning. That's all algorithms. The idea of trying to optimize your content, all of that is machine learning. But having said that, this is very quirky as well because you don't understand exactly why the computer has developed the algorithms it has. And there are very weird artifacts in there. Let me again give you an example. So I talked about the blinking software. And our blinking software was particularly good because it was optimized far labs. So you sit in a cubicle and it's better than you would get in the real world because that algorithm is even more accurate because it's in the same environment consistently with even lighting and a whole bunch of things like that that would not happen necessarily if you were another environment. But we changed the furniture once. We changed the chairs in our labs and you see the back of a chair off in the distance. It's not prominent in the image at all. It's just as you're doing it, you see the back of somebody else's chair in the background. And that one difference through our algorithm, our algorithm was operating at 92% accuracy. Suddenly we changed the furniture and the accuracy dropped to 82%. So we lost ten points off of a quirky thing. Why would the back of another chair impact your algorithm? And this is the problem with this approach. It's a theoretical. You don't understand what happens. And in another study that we did, we had these very interesting results predicting market success, but we broke it down and we looked under the hood, and we saw that the order of the alphabet, you know, where the brand appeared in the alphabet was one of the factors, which was driving the machine learning algorithm. Clearly, that's not a factor. So it's a very dangerous art as well because if you're over reliant on it, you can be making some massive errors without any common sense to kind of like guide it. So when somebody is thinking about building a machine learning algorithm to be more efficient in their business. Where are the places that you're seeing them build them the most? Is it in advertising? Is it in site optimization getting someone through your funnel, understanding who your customers are? Where are some of the most common applications? The most common applications really are where you have the data. Already there's analysis occurring around the data. This is now another hand to help you. And again, I want to emphasize, it's important to not have the AI drive your car for you. That's a riskier proposition than it sounds. But to make sure that it's happening in concert with people who are looking at it rationally still and making sure that it makes sense. Because it can be a very dangerous and loaded. For us, though, it's actually about tools. Where we find the biggest benefit is not relying on it to draw our conclusions. But in relying on it to help us with the tools that we need to be able to draw the conclusions. So it helps us enormously, particularly dealing with very complex algorithms associated with the measures that we have, which are calibrated to be individual. For those kinds of applications, it's huge. When you using it as a tool. I guess what I'm confused about is what you mean by a tool. I hear, well, okay, where should I think about building machine learning to help me optimize my company's performance and the answer is it's where the data is. And I'm thinking, is it in my customer acquisition? Is it my retention? Is it figuring out my sales funnel in various parts of the marketing funnel? And what are the tools that I should be using? So when you say, well, you use it to evaluate your tools or to optimize your tools. What do you mean by tools? So most people just want to drive, they don't need to know all the parts that go into the engine, a mechanic needs to know that. You just want to drive the car. In the same way, there are people who are going to be working with the tools to build the car and maintain the car. But the benefits of that really are to the driver. So let me give you an example. We have a spin off, which is called park connect. And that software that we make it available to the market. It's for the qualitative research industry. So this is for people who do focus groups, in depth interviews, UX user design, kind of like research, user experience, research. So we make these tools available to them. And there's a lot of AI that goes in because what we've done with heart connect is we've made the process of doing that research a lot easier by having a virtual assistant. We call her Ava, having Ava basically there to help you the same way that you might have an assistant helping you with some of the tasks that you're doing along the way. So for example, you're doing a zoom session as a focus group and you're listening to these respondents. The type of things that hard connect is doing for you is one. It's transcribing everything that's being said so that at the end of the session, you have a transcript. It's translating it. It's translating into over 60 languages. Again, the translation feature, AI, the transcription feature, AI, then ask the session that's happening. It's automatically tagging key content. You might have given it some key words. It sees a key word. You have a flag pop up visually for you if you're in the client room. So that's helping draw your attention. Oh, they just said free. So I'm interested here. Are they just mentioned in competitors? So Ava is automating the tagging, which you might have had a person do before. Once the session is over, oh, it's tagging the sentiment. So is this a positive sentence? Are they saying something good or negative about the brand? It's tagging that. So you're saying, oh, that was something positive. They just said. And then when you go into the session at the end of the session to do your editing, you can automatically extract. You can say, give me all the negatives. And instantly, it gives you all that video, or you can say, automatically extract any time somebody talked about the competitor or any time somebody said the word free or anything you want. And it will automatically extract that. Now, normally with this type of insights research, it takes about 6 hours for every one hour of actual content for you to be able to extract the insights that you want. This now takes what was 6 hours of work for each hour, and there might be 30 hours of data that you collected, and it takes that and reduces it to almost instantaneously. So you're using the benefits of AI as a tool to give you better access to faster access, better access to the insights. I think the reality is that artificial intelligence and machine learning are all around us in marketing and specifically in the martech industry. When you think about how we're advertising, we're relying on the machine learning and artificial intelligence from some of the platforms when you think about optimizing your site and some of the tools that you use to help you understand what your throughput is there, even things like, hey, content creation. We're thinking about transcriptions, artificial intelligence machine learning is not only starting to help us understand tag flag and make meaning out of our content. It's also starting to have us produce it as well. So we're going to continue this conversation talking about other ways that artificial intelligence and machine learning are relevant for marketers in tomorrow's episode. And that wraps up this episode of the Marta podcast. Thanks for listening to my conversation with Dwayne veron, the founder and CEO of media science. And the third part of this interview, which we'll publish tomorrow, doctor Dwayne and I are going to continue the conversation talking about how science has changed the advertising industry. If you can't wait until our next episode and you'd like to learn more about doctor Duane, you can find a link to his LinkedIn profile on our show notes. You can contact him on Twitter where his handle is Dwayne baron that's DU, and you can visit his company's website, which is media science dot com. And also a special thanks to HubSpot for sponsoring this podcast, no matter what problems you're trying to solve, the HubSpot CRM platform is easy to learn, use and love because it's designed for the ways marketers actually work. With customizable tools that you can add or subtract as you grow, HubSpot is ready to help you squash the endless problems in your business. To learn how HubSpot can help you grow better, go to HubSpot dot com. A special thanks to higher logic for sponsoring this podcast. If you're ready to move beyond transactional relationships to start building more engaged, meaningful connections with your customers, then join more than 3000 companies representing 350,000 online communities with more than 2 million users by using higher logic vanilla, the vanilla platform combines the power of communities, Q&A, and knowledge bases to increase retention, reduce customer support, costs, and maximize revenue. To get started, go to higher logic dot com, or you could visit mar tech pod dot com slash vanilla. That's mar tech pod dot com slash vanilla. Just one more link in our show notes I'd like to tell you about. If you didn't have a chance to take notes while you were listening to this podcast, head over to martek pod dot com where we have summaries of all of our episodes and contact information for our guests. You can also subscribe to our once a week newsletter and you can even test your topic suggestions or your marketing questions, which will answer live on our show. Of course you can always reach out on social media, our handle is mar tech pod MAR TEC, POD on LinkedIn, Twitter, Instagram, and Facebook, or you can contact me directly my handle is Ben Jay shapp. And if you haven't subscribed yet and you want a daily stream of marketing and technology knowledge in your podcast feed, we're gonna publish an episode every day this year, so hit the subscribe button in your podcast app, and we'll be back in your feed tomorrow morning. All right, that's it for today, but until next time, my advice is to just focus on keeping your customers happy.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"These associations. It's also the same idea for what's building self-driving cars. All right, Tesla. I don't know how you're going to avoid it, but here the car crashes go avoid them. Good luck. So getting on to artificial intelligence and machine learning's impact on mar tech when we think about how the applications of being able to process data and allow computers to find solutions, how is that changing what's happening in mar tech where we're weaving multiple tools together and sort of building a smarter flytrap for marketing. So
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Welcome to the martek podcast. Today we're going to talk about the science of machine learning and marketing, joining us is Dwayne veron, who is the founder and CEO of media science, which is a lab based audience research provider, incorporating a range of neuro measures, including biometrics, facial expression analysis, eye tracking, EEG, and more. With state of the art labs in New York, Chicago and Austin, media sciences discovering actionable insights and advertising, technology, media, and consumer trends. Yesterday, doctor Dwayne and I talked about how artificial intelligence and machine learning are changing consumer research, and today we're going to continue the conversation talking about some of the innovations from AI and ML in the martech industry. This interview is brought to you by higher logic. Are you ready to move beyond transactional relationships? Want to build more engaged, meaningful connections with your customers? Well, higher logic vanilla is the leading online community platform for growth minded companies. The vanilla platform combines the power of communities, Q&A, and knowledge bases to increase retention, reduce customer support, and maximize revenue with higher logic vanilla, your customers can seamlessly engage with their peers to get answers for their questions, provide feedback and become an advocate for your brand. So join more than 3000 companies representing 350,000 online communities with more than 2 million users using higher logic vanilla to get started today, go over to higher logic dot com or you can use mar tech pod dot com slash vanilla. That's Marta pod dot com slash vanilla. And also since you're into marketing podcasts, check out marketing made simple, hosted by doctor JJ Peterson, marketing made simple, brings you practical tips to make marketing easy and more importantly, make it work. And JJ is a brilliant marketer and a friend of the mar tech podcast. So to hear tips like how to write a captivating speech or how to market yourself into a new job, search for marketing, made simple in your podcast app. We're going to HubSpot dot com slash podcast network. Okay, on to today's episode. All right, here's the second part of my conversation with doctor Duane veran, founder and CEO of media science. Doctor Duane, welcome back to the martek podcast. It's great to be back. Excited to have you back on the show and excited to continue our conversation yesterday. We talked about consumer research and specifically how media science is taking advantage of some of the more cutting edge technologies monitoring the signals that people give while they are exposed to media and some sort of stimulus. All right, do your hands get sweaty. Does your heart beat go down? Does your mouth go up when you see an ad? Those are all different signals that tell us different reactions to an ad, a political statement, somebody's presence. So consumer research is being changed pretty dramatically by some of the testing and science that we're able to do, there's also a fair amount of artificial intelligence and machine learning that is going into the mar tech industry. Let's start off by first off defining artificial intelligence and machine learning. What are the two of them? And then I'm also going to ask you, obviously, what's the impact they're having on the martech industry, but start off with giving me the definitions of AI and ML. Well, rather than a definition, I might just talk through because I think people hear these things and they don't know necessarily what it entails. And where the real revolution is, how it is that this kind of like turbo charges everything. I mean, it's really powerful. And it's implications. I hear artificial intelligence. And I think, oh, that stuff that Google, Facebook, Twitter, Microsoft, Apple, Netflix, are doing, and that everybody else says they're doing, but really they're talking about marketing automation and machine learning. Am I wrong? I feel like artificial intelligence, you need this giant dataset and people actually aren't using it. So artificial intelligence is a broader concept, just the idea that you have machines helping facilitate decisions. But machine learning is more specific, and much more accessible than people realize. Machine learning is not as difficult to do, any company could really start doing it. It's not that hard. Let me give you a little bit of background on what machine learning is and how it works. And I might use a real world example to help illustrate that. So that media science, one of the measures that we use based on software we developed ourselves is we analyze people's blinking behavior. So when you're paying attention, your eyes blink less frequently and your eyes are open for longer periods of time than they are when you're not paying attention or when something is not. I mean, if you think about it, if you're getting drowsy, your eyes are kind of keeping close longer periods of time, et cetera. That's at the extreme. So we want to measure people's blinking behavior. And we need software to do it. And there was not software in the market to do it. So we had to invent our own. Now, our team worked on that for many years. And in the initial years, it was a very, very, very hard challenge. Because you have to have algorithms to detect the face. Now you need algorithms that detect the eyes. I mean, you can imagine how incredibly complex it is to work out the math that you need to kind of figure out when an eye is open or close, given a natural video feed, for example. Now then machine learning comes along. And something which we had been working on for four or 5 years suddenly goes from being a task that requires years to do to something that literally requires days to do. Tell me how that works. How did it move from a yearlong project or multiple years to a couple of days? So machine learning, the way that it works is you have a reference. You need a reference library. So in this case, we're interested in people's blinking behavior. We need a library of videos for which we have humanly coded when the eye is open and when the eye is closed. So that's a task. We might farm that out to third world countries or something and get all this data back around when the eye was open when I was close. And then you give it to the computer. And you say to the computer, okay, computer, you go out there, I don't know how to tell you how to do it, but you figure out, here's the data. This is when the eyes open, this is when the eyes close, you figure it out. And that's what machine learning does. And the way that it does it is it literally takes every pixel on the screen, and it just creates math around running millions of trials over and over and over again, testing different possibilities to see how it can account for that. And then at the end of that process, these millions of data runs that it can do again very quickly, it then gives you an algorithm, which will detect whether the eyes open or close. And it's just as good, if not better than what we would have produced after 5 years of work trying to get there. So now you have this tool which revolutionizes in anything that you can reduce to a reference library. You can actually then start getting that kind of machine learning around. So very, very powerful, particularly in terms of the speed at which it can start to detect
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"The martech podcast is a proud member of the HubSpot podcast network to find great business podcasts like this one, visit HubSpot dot com slash podcast network. From advertising to software as
InnovaBuzz
"dr duane" Discussed on InnovaBuzz
"Trend you don't see authenticity as much perhaps as as we would like is coming back to this expectation. Isn't it so again. It's that self awareness that he's a kite. Mike that choice to be yourself. act talking about the superwoman. Versus the superb woman. Being yourself was actually a lot less draining and a lot less energy rate just any acta. Yeah and i do a training with puppet training and that's actually a course company about the authentic leader and how their studies that show you know the productivity that results when a leader is authentic. There's there's so many positive attributes that angered just somebody dating themselves being authentic in their interactions people. Yeah the other important thing i think. The the is that authentic regatta give ourselves permission. Decide that Some people are gonna look at us as people as characters inside not lock that person and that's perfectly okay opinion because those that say war really lot personnel up to work with person. Yeah the they know what they're getting and from from a business perspective you want to work with people who were view he because if you're working with somebody in it's a struggle. It's not worth your time. Heck i will thanks. Genesis has been fabulous wiccan. People find out more bad year. The work you do get a hold of your book and might even reach out so thanks for sharing. Now thank you for asking So if you i to find out more about how we were with emerging adults Helping kids get on the right path. Though to next steps navigation that palm he he wanted can find out more about what i do women in helping them to their offensive powerful south though to this woman that And i have two books on amazon. And he searched my name or the superb woman. You will find it. So one of them was written in paris. Way was the other unwritten gas. The other one was new jersey. Not quite as my okay. Do you have some hot advice. Rallies today janet. I would say the best thing you can do. As i've been saying is years how to take that time to valley yourself to know that you have neak hips and talents that the world needs to figure out what they are with anita's in the world and how greater boston so succinctly put two of her. I finally janet. Who should i get on this. Show wine Ihs met woman yesterday. Melissa layer and she has a company of cultivating sales and she has done amazing job of pulling together all these different communication platform into one. So that you can just look at one place and have have your of logs that you do do all of your unification's it's all in place. I was like she comes from the point that it's all about that businesses about relationships and in row technology in place so that you can fast health us. How great that's one of my favorite subjects. Because i'm i'm kind of technology geek. But i recognize. Nic a this annoys me. When people abdicate the human aspects to the technology and said well know either either while the technology doesn't allow it so bad luck or or. It's like i i don't want to talk to you. I'm just send send a help. Request to the system or whatever they. Yeah so one of my favorite topics to talk. Matt sal juices to melissa will reach out. Have that conversation asked her will. Thanks so much for sharing your time in your thoughts with Generously today agenda. Brilliant joy that conversation. I think there's a lot of value in what you've shed. So all of this to the future. Let's dine touch. Hanky you're gonna really appreciate. I hope you enjoyed that. Wonderfully informative engaging conversation with janet and took something away from her episode. Jennifer focus on curiosity about self and about our own thoughts. Were some of the big highlights for me. I'd love to know what you took away from janet's episode live comment below the blog. Post which you can find out inova biz dot. Co forward slash. janet. Neel that is john i in. At e. l. all lower case one word in over biz dot. Co forward slash. janet. Neel you'll also find contact information for getting in touch with janet as well as links to the superb woman website to the next steps navigation website to generate social media pages and the other resources. We spoke about ina conversation today. If you like this episode please to share it share it with at least two other people let it might help because you're doing them a messy fiber tag me on that share and i will reach out to you with a special thank you. Janet suggested that we have a conversation with melissa blair of cultivating sales on a future enough about podcast. Episode said melissa. Keep an eye on your inbox for an invitation from us to the another bus podcast. Courtesy of janet. Neel tune in again to the next episode of the inevitable's podcast. We have yet more fantastic guests lined up including dr duane. Veran of media science and stiffen hit a brand of dream data. Thanks for listening to this episode. Make sure you subscribe to the show to be reminded of new episodes. Describe legal review. If you like even that. I'm asking you to leave a review because pickups other people find shark go to inova biz dot. Co two joint an mocking transformation community and excess. A free gift. My team and i made for you. It's the marketing mazda mini class. We wanna give you everything you need to. Transform your marketing. Indoor human sainted relationship. Crocus were injured until next on on straps from penelope's remember be some and keep innovating..
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Them separately. You should think about them in a related sad but you should think about both. They're both important whether you got sales or not very important. I mean there's no point in delivering against your communication objectives if those communication objectives translated to sales. So that's important but also whether you're moving parts are kind of like getting you where you wanna go is also or the problem that we have is that you don't want a brand to have a strategy that a speculative a brand should have a delivered strategy if it's trying to trigger an emotion it should have a reason why it's trying to trigger that emotion it should measure whether it's successful in doing that and ultimately that will help it know whether it strategy is on target it could be that it strategies off base that that's not the right strategy that trading that emotion doesn't translate the but you won't know if you're not accurately measuring for those objectives so the business of measuring these objectives is really about making sure not only that you're delivering against the objectives but ultimately as a consequence that your strategy is actually on target doing when you think about the science that is being applied across customer research across martic across the advertising industry. It seems like there is a practical way to measure business performance that most of us rely on and then there is the understanding of why. Give me the pitch of why marketers should take a more scientific approach to understanding the why behind their marketing efforts as opposed to just focusing on the performance so sales behavior is oftentimes a very tactical way of looking at what we're doing. You want to inform your larger strategy. You wanna have a theory about not only a theory. You're also trying to build a brand and your value. Whatever your product is. The portion of your value that is being delivered by your brand is far greater than the value that section being delivered by the physical products itself. Branding is the real estate of the mind. And it's where you sit in a person's psyche ultimately and you're not gonna get that by the sales behavior. You're knocking know where you were situated. Now people situate you in their minds relative to your competitors so you have a very short term for short lived strategy if you're just living on sales alone is the classic problem that a brand house where they may discount their product in the store and get themselves but they're not gonna make cross it that way. Is that really where they want to position themselves. So in the science and the art of really understanding your brand who you are. What works for you. What connects with your consumers. You're not gonna get there by sales data alum. You're going to need to understand. Why do caught the sales and you're not gonna be able to get there unless you have some measures of those moving parts within that. So that's the reason why again. It's not enough to just have the sales data. It's not enough to just have the moving parts. You really need both of those to come together to get the full picture of what's going on. It's a little bit light a thermometer in a room. You're getting the overall temperature in the room so you can say. Oh that's hot or it's call whatever which is great. But you don't know why it's hot you don't know why it's called and you're not gonna be able to adjust the temperature ultimately if you don't know what it is that's contributing to it. So the moving parts really go together. You can't change the temperature of the room just by increasing the to monitor the you you need. Both you have to have the moving parts. And the performance my fundamental belief about marketing and our job as marketers is to understand who our customers are and tried to deliver marketing messages products and services to those customers that make their lives better to solve problems and create value for them without a deep understanding of who your customers are. And what motivates them. You can never really accomplish that. Taking a scientific approach to understanding not only who your customers are but what drives their behavior is something that will make you a better marketer. At the end of the day we're all held to the end results but our job is to understand who the customers are and using services like media science and technologies like machine learning and artificial intelligence to not only better understand our customers to better target them is just gonna make us better marketers. Dwayne i appreciate you coming on the show and walking through some of the more advanced technologies you use. Thanks for being our guest. Always a pleasure and anytime all right. That wraps up this episode of the mar tech podcast. Thanks for listening to my conversation with dwayne veron the co founder and ceo of media science. If you'd like to get in touch with dwayne you can find a link to his linked in profile on our show notes. You can contact him on twitter where his handle is dwayne. Veron d. u. a. n. e. v. a. r. a. n. or. You could visit his company's website which is media science dot com a special thanks to hub spot for sponsoring this podcast. Don't forget that the inbound twenty twenty one conferences coming on october twelfth through the fourteenth. So if you want inspiration if you'd like to grow your network or learn from global leaders in business and media like oprah winfrey spike lee you know me head over to inbound dot com. That's i n. b. o. u. n. d. dot com check out their lineup or register for free inbound dot com and also a special thanks to lincoln marketing solutions for sponsoring this podcast. It's time for you to do business. Where business gets done and you can get one hundred dollars of advertising credits towards your first linked in campaign. there's some terms and conditions apply. But you can get one hundred bucks off if you're launching that first campaign by going the lincoln dot com slash martic again. That's lincoln dot com slash marshek. Just one more lincoln. Our show notes. I'd like to tell you about if you didn't have a chance to take notes while you're listening to this podcast head of mar tech pod dot com or we of summaries of all episodes and contact information for our guests. You can also subscribe to our once a week newsletter and you can even send us your topic suggestions or your marketing questions which will answer live on our show. Of course you can always reach out on. Social media are handle is mar tech pod. Marta c. h. p. od on lincoln twitter instagram and facebook. Or you can contact me directly. My handle is benjy shop. B. e. n. j. s. h. a. p. and if you haven't subscribed yet new on a daily stream of marketing and technology knowledge in your podcast feed. We're.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Where science can really make a difference. One of the things that i've always said and i have not studied this but i say it anyway because it's my job as a podcast host. Is that the formats of advertising or changes because people are exposed to so much more media now. I know that. I have my phone in front of me and that seems to be enough information for me to say well. I consume more media. I see social media. I still watch television. We got the radio on. There's podcast that have ads inserted to him. There's all sorts of different formats. Do you have a sense of where the people are. Actually engaging with media more because the assumption is if there's more media than people are getting tired of being advertised to and i actually haven't seen the science. I did not read the research. It is an inference. Not any sort of study. That i've conducted try to understand if people are actually consuming more media and more advertising impressions. Oh no there's really good research. Which has that he. Consumption media has grown exponentially. I mean there's no doubt about that and of course if we're getting exposed to more media we probably are getting exposed to more ads as well so in that sense. The challenge for a brand to stand out is greater than it was before but by the same token there are a lot more ways that you can engage and that's an opportunity for brand particularly relative to its specific communication objectives. So that's where it comes again. I say this thing about communication objectives the ultimate job of course of marketing itself from sales. But that's too big an objective. You know you have to break that down into a strategy for how you going to do that. And you want to know whether you're strategies effective and that really comes down to a series of communication objectives so now the question is do you know whether or not you're delivering against each of those objectives and historically we don't know because we're just trading on these common currency that we have like ratings or something men now we have the ability to get much much much more precise so i i use the example a couple of days ago on our show about the mars study that we did a lot of mars. Ads are about humor. And you want the ads to be funny and you want them to be funny and specific moments in time during the ad. Well now you have a very very precise measure of whether it was funny or not. Funny and i will tell you that. Our research says that whether it's funny or not does have a difference in terms of whether it's ultimately going to be effective so this translate ultimately to cells but now you can actually measure scientifically whether the ad was funny and brand needs. Best in class measures for each of their objectives. And they're not gonna be the same for different brands so people are going to be trying to doing different things and it's just about more accurately measure that so mars is trying to measure funny with their advertising if it's funny there's a positive association with the brand and hopefully that has an impact on underlying sales when you think about bdb measurement when you're thinking about reaching people in a professional setting obviously humor is not something that is devoid but it's also not something that we rely on from b to be. Should we all start making our ads into you. Know long running jokes or is there a time and a place where you need to think about other measurements other than i've humor to get your ad across. Get your message across. What is the advertising measurements. That you've seen for the b. two b. market. The most important of course is attention or it's not actually attention that you should be interested in is inattention or the absence of inattention is sounds like a subtle difference. More attention is not necessarily better but no attention is definitely fatal so you wanna get through the threshold of getting enough attention for everything else to matter and if you don't pass that special nothing else matters so attention is a great example. Whether you're talking about beats be or whether you're talking about b. to see everybody needs attention and you don't got it you don't in fact. Yeah i mean without any attention. No one's going to understand. Have awareness recall. Nobody's going to think about your brand. You're dead on arrival but there is good attention and bad attention as well. You know it seems like maybe this different in the political sphere where sometimes all attention is good attention when you're a marketer and you're starting to think about creating good buzz as opposed to just getting out there. What are some of the ways that you can evaluate what works and what doesn't it's not a theorist good attention. There's attention and there's no attention and then there's something else that comes in after attention and that's really my point attention. Is that threshold. If you don't have attention you're not gonna notice what goes on but just because you have attention doesn't mean that it's going to be good or bad it's everything else. Now that starts again about how he will start to evaluate your message so it could be that you're looking at whether that is positive or negatively oriented. So are you getting this attention but then are you seeing. We call this valence. Which is you know whether things are overall positive or negative. Are you getting positive or negative. Valence are people reacting in a favorable way towards the brand or are the acting in a negative towards the brand again that's a dig threshold crossed that fresh. Oh and now you're onto the next one and so layer by layer. You just work to delivering against your objectives. Is someone paying attention your message. Do they think of it. Positively or negatively surveillance. What comes after balance. It depends by brand for example. You could be kind of brands. That has a very emotive message or you could be a brand that has something that requires cognitive function. You know that requires you to think about what's being set very different scenarios so really once you go past the attention pressurall. Everything else now is going to be brand specific rather than generic so. You're going to be talking about that brand so for example. Maybe this brand is looking for empathy. That's something very different to a brand. That's looking for you thinking about their claim. So you really start getting into specific metrics around what the objectives of brand are rather than generic measures about overall whether. Something's got attention or didn't get attention. When you think about measuring the effectiveness of your advertising we talk about understanding attention. Understanding valence understanding. You know some of the other emotions that you're generating those seem like a very difficult things to measure without a lab and be things that you can't necessarily dictate in terms of business results. My advertisement was successful because this many people more came to the website and bought something. How most advertisers think about ad performance as opposed to my ad was successful because people think of my brand as fun as light as beautiful as serious the emotions that come out of it. How do you buy for kate. The understanding of why someone is reacting to your ad as opposed to the underlying performance. Or should you actually.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Hello margaret is benjamin shapiro here and thanks for tuning into the mar tech podcast. Which is brought to you by the hub. Spot podcast network not spot. Podcast network includes great. Shows like the. I digress podcast. Which helps marketers streamline their businesses by sharing hacks mindsets and actionable tools. Listen learn and grow with the martic podcast. And i digress on the hub. Spot podcast network at hub spot dot com slash podcast network from advertising to software as a service to data across all of our programs and violence. We've seen a fifty five. To sixty five percent open rate getting brands authentically integrated the content performs better than tv advertising. Typical life span of an article about twenty four thirty six hours. We're reaching out to the right person with the right message and a clear. Call that action that it's just a matter of timing. Welcome to the mar tek. Podcast a benji shabby llc production. In this podcast you'll hear the stories of world class marketers use technology to drive business results and achieve career. Success will on earth the real world experiences of some of the brightest minds in the marketing and technology space. So you can learn the tools tips and tricks. They've learned along the way. Now here's a host of the bar tech podcast benjamin shapiro. Welcome to the mar tech podcast. Today we're gonna talk about the science of machine learning in marketing. Joining us is duane veron. Who is the founder and ceo of media science which is a lab based audience research provider incorporating a range of narrow measures including biometrics facial expression analysis eye tracking eeg and more with state of the art labs in new york chicago and austin media. Sciences discovering actionable insights advertising technology media and consumer trends. So far this week dr duane and i have talked about how to use artificial intelligence and machine learning in consumer research and yesterday we talked about a and machine learnings innovation in martic today. We're going to wrap up the conversation and talk about how science is changing the advertising industry but before we get started with today's interview i wanted to share some exciting news as part of our relationship with the hub. Spot podcast network. I've been invited to participate. in hub. Spots inbound twenty twenty one conference inbound is an amazing resource to find inspiration. Grow your network and to learn from global industry leaders across business marketing sales and customer success. And this year's lineup includes some amazing speakers. In business and media including oprah winfrey dave chang spike lee. You know just me and oprah hanging out talking about our media empires no big deal anyway. You can grab a free starter. Pass to access all spotlight conversations. Where upgrade to a powerhouse. Pass for full access to breakout sessions. Curated meetings on demand content and more the inbound twenty one conferences hosted with love by hub spot and it takes place on october twelfth through the fourteenth to learn more about the inbound. Twenty twenty one lineup or to register for the conference for free could inbound dot com. that's i n. b. o. u. n. d. dot com that's inbound dot com and this podcast is also sponsored by linked in marketing solutions. Let's pretend for a moment that you're about to launch a campaign. It tested well. Your entire team is happy and everything is going. According to plan except for the one thought in the back of your head. How do i ensure that the people i want to target are in the mindset to receive my message the answer is linked in because when you mark it on linked in your message reaches the people who are ready to do business and that means that your advertising campaigns will work as hard as it can as soon as you launch it there over sixty two billion decision makers on linked in. And that's why seventy eight percent of to marketers rate linked in as the most effective social media platform for their organization. I personally use linked in both as an organic growth channel to syndicate our content. And i use it as an advertising channel to retarget tech podcast listeners with some offers from our sponsors and hey the messaging platform is a great way to reach potential podcast guests as well so look whatever. Your marketing goals are from brand building to lead generation. Lincoln will help you reach your audience with the features that you know about like targeting by job title company name and location but you can also re target based on custom triggers from your website or even by uploading your crm lists and customize your campaign based on the action that you want the customers to take so look do business where business is done and you can get one hundred dollars of advertising credits towards your first lincoln campaign. there's some terms and conditions apply. But you can get it by going to lincoln dot com slash mark. Again it's one hundred dollars off of your first linked in campaign by going the lincoln dot com slash mar tech all right. Here's the last part of my conversation with dr duane. Veron founder and ceo of media. Science dr duane. Welcome back to the mark tech podcast. It's becoming a regular thing. It's becoming a regular gig. Then it's kind of how we roll around here. We get to once in the next thing you know three days later. You're still talking appreciate you coming back. And i'm excited to wrap up our conversation. So far we've talked about artificial intelligence and machine learning we talked about neuroscience and how it's changing consumer research how we can understand what some of the signals are in why people react to specific ads or media. Yesterday we talked about more artificial intelligence machine learning how that works. And how we're applying it into martin today want to specifically focus on science and advertising. It seems like there are some fundamental changes for how people react to advertising also some of the formats of advertising. So you know as we start to think in a different capacity about how we should advertise where our advertisements should go. Talk to me about how some of the technologies that you have worked with are changing marketers. Thoughts on their ad buys nad strategies. I think the biggest change that we've seen over the course of the past decade or so is that especially with premium advertising tv ads etcetera. We've really gone from a one trick. County which was the thirty second commercial to a universe of infinite possibilities when it was one format people had to fit whenever their needs were into that media. It was just a factor. These are the rules. You have a thirty second commercial figure out how to take your marketing objectives and squeeze it within that template now we live in a universe with like i say infinite possibilities so the question becomes push wrandell and there's a lot of possibilities out there so you need to have some way of really grappling with what your objectives are in the first vice. What you communication objectives are and had a measure for success and that's.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Without any common sense to kinda like guy so when somebody's thinking about building a machine learning algorithm to be more efficient in their business. Where are the places that you're seeing them build them. the most is it in advertising. Is it in sight optimization. Getting someone through your funnel understanding who your customers are where some of the most common applications the most common applications really are where you have the data already. There's analysis occurring around the data. This is now another hand to help you and again. I want to emphasize it's important. It's not have the drive your car for you. That's a riskier proposition than it. Sounds but to make sure that it's happening in concert with people who are looking at it rationally still on making sure that it makes sense because it can be very dangerous unloaded for us though. It's actually about tools where we find. The biggest benefit is not relying on it to jar conclusions but in relying on it to help with the tools that we need to be able to draw conclusions so it helps us enormously particularly dealing with very complex algorithms associated with the measures that we have which calibrates the individual for those kinds of applications. It's huge when you're using it as a tool against what i'm confused about is what you mean by tool. I hear well okay. Where should i think about building a machine learning to help me optimize my company's performance and the answer is it's where the data is and i'm thinking is it in my customer. Acquisition is at my retention. Is it you know figuring out my sales funnel in various parts of the marketing funnel and in a. What are the tools. That i should be using. So when you say well you use it to evaluate your tools to optimize your tools. What do you mean by tools. So most people just want to drive. They don't need to know all the parts that go into the engine. A mechanic needs to know that you just want to drive the car. In the same way there are people. Were going to be working with the tools to build. The car maintain the car but the benefits of that really archer the driver. So let me give you an example. We have a spinoff which is called heart connect and that software that we make available to the market. it's for the qualitative research industry so this is for hupa who do focus groups and interviews you s. user designed kind of like research user experiences research so we make these tools available to them and there's a lot of ai that goes in because what we've done with heart connect is. We've made the process of doing that research a lot easier by having a virtual assistant. We call her eva having eva. Basically there to help you the same way that you might have an assistant helping you with some of the tasks that you're doing along the way so for example you're doing a zoom session as a focus group and you're listening to these respondents. The type of things. That heart connect is doing for you. As one is transcribing. Everything is being said. So that at the end of the session. You haven't transcripts is translating. it is translating. Into over sixty languages. Again the translation feature ai. The transcription feature ai. Then asked the session this happening. It's automatically tagging key content. You might have given it some key. Words that season he word. You have a flag pop up visually for you if you're in the client room so that's helping draw your attention. Oh they just said free. So i'm interested here or they just mentioned a competitor so eva is automating the tagging which you might have had a person to you before whence the sessions over. Oh it's tagging the sentiment. So is this a positive sense of his saying something good or negative about the brand. It's tagging that so you're seeing. That was something positive just sad. And then when you go into the session at the end of the sessions do you're adding you can automatically extract you give me all the negatives and instantly it gives you all that video or you can say automatically extract any time. Somebody talked about the competitor war. Anytime somebody said the word free or anything you want and it will automatically extract that now normally with this type of insights research. It takes about six hours for every one hour of actual content for you to be able to extract the insights that you want this now takes. What was six hours of work fridge. Our and there might be you know thirty hours of data that you collect it and it takes that an reduces. It's almost instantaneously so you're using the benefits of a i as a tool to give you better access to faster access. Better access to insights. I think the reality is that artificial intelligence and machine learnings are all around us in marketing and specifically in the mar tech industry. When you think about how were advertising were relying on the machine learning and artificial intelligence from some of the platforms when you think about optimizing your site and some of the tools that you use to help you understand what your throughput as they're even things like hey content creation we're thinking about transcriptions. Artificial intelligence machine learning is not only starting to help us understand tag flag and make meaning out of our content also starting to have us produce it as well so we're going to continue this conversation talking about other ways that artificial intelligence and machine learning are relevant for marketers in tomorrow's episode. And that wraps up. This episode of the mar tech podcast. Thanks for listening to my conversation with dwayne veron the founder and ceo of media science in the third part of this interview. Which will publish tomorrow. Dr duane and i are going to continue the conversation talking about how science has changed the advertising industry. If you can't wait until our next episode. And you'd like to learn more about dr duane. You can find a link to his linked in profile on our show notes. You can contact him on twitter. Where his handle is duane baron. That's d. u. a. n. e. v. a. r. a. n. or you can visit his company's website which is media science dot com a special. Thanks to hub spot for sponsoring this. Podcast don't forget that the inbound twenty twenty one conferences coming on october twelfth through the fourteenth. So if you want inspiration. If you'd like to grow your network or learn from global leaders and business and media like oprah winfrey spike lee. You know me. Head over to inbound dot com. That's i n. b. o. u. n. d. dot com check out their lineup or register for free inbound dot com and also a special thanks to linked in marketing solutions. For sponsoring this podcast. It's time for you to do business. Where business gets done and you can get one hundred dollars of advertising credits towards your first lincoln campaign. there's some terms and conditions apply. But you can get one hundred bucks off if you're launching that first campaign by going to lincoln dot com slash martic. Again that's lincoln dot com slash martic. Just one more lincoln. Our show notes. I'd like to tell you about if you didn't have a chance to take notes while you're listening to this podcast. Hetero mar tech pod dot com or we have some revolver episodes and contact information for guests. You can also subscribe to our once a week newsletter and you can even send us your topic suggestions or your marketing questions which will answer live on our show. Of course you can always reach out on. Social media are handle is mar tech pod. Marta see hp od on lincoln twitter instagram and facebook. Or you can contact me directly. My handle is benjy shop. The nj s. h. a. p. And if you haven't subscribed yet and you wanna daily stream of marketing and technology knowledge in your podcast feed. We're going to publish episode every day this year so the subscribe button on your podcast app and we'll be back in your feet tomorrow morning. All right that's it for today but until next time my advice is to just focus on keeping your customers happy..
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"People actually aren't using it so artificial intelligence is broader concept. Just the idea that you have. Machines helping facilitate decisions but machine. Learning is more specific and much more accessible than people realize machine. Learning is not as difficult to do. Any company could really start doing it. It's not that hard. Let me give you a little bit of background on. What machine learning isn't how it works. And i might use a real world's out all taub illustrate that so at media science one of the measures that we use on software developed herself is we analyze people's blinking behavior. So when you're you're you're paying attention. Your eyes blink. Less frequently and your eyes are open for long periods of time than they are when you're not paying attention or when something is not i mean if you think about it if you drowsy. Your eyes are kind of keeping close longer periods of time etc. That's the extreme. So we want to measure people's leaking behavior and we need software to do it and there was not software in the market to do it so we have to advance our. How now archie marked on that for many years and in the initial years. It was a very very very hard challenge because you have to have algorithms to detect the face. Now you need algorithms that detects the is i mean you can imagine how incredibly complex it is to work out the math that you need to figure out what an eye opener close even a natural video feed for example now then machine learning comes along and something which we had been working on for four or five years suddenly goes from being tasks that requires years to do to something at literally requires days to do. Tell me how that works. How did it move from a year long project or multiple years to a couple of days so machine learning way that it works. Is you have a reference. You need a reference library so in this case for interested in people's blinking behavior we need a library of videos for which we have humanly coded when the is open and when the is close. So that's a task. We might farm that out to World countries or something and get all this data backs around. When the i was open i was and then you give it to the computer and you say to the computer okay. Computer you go out there. I don't know how to tell you how to do it but you figure out here's the data. This is when the is open this as those you figure it out. And that's what machine learning dust and the way that it does it is it literally takes every pixel on the screen and it just creates math around running millions of trials over and over and over again testing different possibilities to see how can account for that and then at the end of that process these millions of data runs that it do again very quickly and then gives you a algorithm which will detach whether the eyes of nur close. And it's just as good if not than what we would have produced after five years of work trying to get there so now you have the school which revolution as anything that you can reduce to a reference library. You can actually then start getting that machine learning around so very very powerful particularly in terms of the speed at which it can start to detect these associations myth also the same idea for what's building self driving cars. All right tesla. I don't know how you're going to avoid it. But here the car crashes go avoid them good luck so getting onto artificial intelligence and machine learnings impact on martic when we think about how the applications of being able to process data and allow computers to find solutions. How's that changing. What's happening in martic. Where were weaving multiple tools together and and sort of building a smarter fly trap for marketing. So it's the same idea if you have data which we have in abundance and if you have a voice of truth and that's indispensable to this working. If you don't have the whistle of god in the equation in other words. He doubt know what really happens. And if you put an approximation of what happens you've just built a bad system. You built a system that has noise it. That is likely to get wrong. So you have to have some point of reference which is accurate if you build a system to an inaccurate reference. you're going to amplify the inaccuracies so you have to have something for which you have some real knowledge about what actually happens. And then you're able to marry those two together yet your fact so when we think about how machine learning is affecting the marceca at understand the idea that you need to find a data set and point out. What the true target that you're pointing at. I've some data about who's downloading my podcast. I have a good sense of who. I wanna target. I can't just build machine learning to turn them into subscribers or turn them into revenue. I need to use a series of tools that are connected together. I've got to figure out who to target. We're using machine learning algorithms that facebook and google and all these other companies have put together to help us drive traffic. And then i'm getting someone to my site and there's potentially more machine learning that goes into site optimization talked to me about where the actual mar checking is happening where people using these advanced technologies to do marketing successfully. Well everywhere everywhere. I mean you mentioned at the the big players in the game and how the optimized the content that's machine learning that's all algorithms the idea of trying to optimize your contents. All of that is machine learning. Not having said that this is very quirky as well. Because you don't understand exactly why the computer has developed the algorithms it has and there are very weird artifacts in their lemme again give you an example. So i talked about the blinky software and our blinking. Software was particularly good because those optimize far labs. So you sit in a cubicle and it's better than you would get in the real world because that algorithms even accurate because it's in the same environment consistently with even lighting and a whole bunch of things like that that would not have necessarily if you another environments but we change the furniture. Once we changed the chairs and our labs and you see the back of a chair off in the distance it's not prominent in the image at all. It's just doing it. You see the back of somebody else's chair in the background and that one difference through our algorithm our algorithm was operating at ninety two percent accuracy. Suddenly we changed the furniture and the accuracy dropped to eighty two percent so we lost ten points of a quirky thing. Why would the back of another chair. Impact your algorithm and this is the problem with this approach is eight theoretical. You don't understand what happens in another study that we did. We have these very interesting results. Predicting market success but we broke down and we looked under the hood and we saw that the owner of the alphabet. You know where the brand appeared in the alphabet was one of the factors which was driving the machine learning algorithm clearly. That's not a factor so it's a very dangerous aren't as well because if you're over reliance on it you can be making some massive errors.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Hello margaret is benjamin shapiro here and thanks for tuning into the mar tech podcast. Which is brought to you by the hub. Spot podcast network not spot. Podcast network includes great. Shows like the. I digress podcast. Which helps marketers streamline their businesses by sharing hacks mindsets and actionable tools. Listen learn and grow with the martic podcast. And i digress on the hub. Spot podcast network at hub spot dot com slash podcast network from advertising to software as a service to data across all of our programs and violence. We've seen a fifty five. To sixty five percent open rate getting brands authentically integrated the content performs better than tv advertising. Typical life span of an article about twenty four thirty six hours. We're reaching out to the right person with the right message and a clear. Call that action that it's just a matter of timing. Welcome to the mar tek. Podcast a benji shabby llc production. In this podcast you'll hear the stories of world class marketers use technology to drive business results and achieve career. Success will on earth the real world experiences of some of the brightest minds in the marketing and technology space. So you can learn the tools tips and tricks. They've learned along the way. Now here's a host of the bar tech podcast benjamin shapiro. Welcome to the mar tech podcast. Today we're gonna talk about the science of machine learning in marketing. Joining us is duane veron. Who is the founder and ceo of media science which is a lab based audience research provider incorporating a range of narrow measures including biometrics facial expression analysis eye tracking eeg and more with state of the art labs in new york chicago and austin media. Sciences discovering actionable insights in advertising technology media and consumer trends yesterday. Dr duane and i talked about how artificial intelligence and machine learning are changing consumer research. And today. we're gonna continue the conversation talking about some of the innovations from a. and m. l. in the mar tech industry but before we get started with today's interview. I wanted to share some exciting news as part of our relationship with the hub. Spot podcast network. I've been invited to participate. in hub. Spots inbound twenty twenty one conference inbound is an amazing resource to find inspiration. Grow your network and to learn from global industry leaders across business marketing sales and customer success. And this year's lineup includes some amazing speakers. In business and media including oprah winfrey dave chang spike lee. You know just me and oprah hanging out talking about our media empires no big deal anyway. You can grab a free starter. Pass to access all spotlight conversations or upgrade to a powerhouse. Pass for full access to breakout sessions. Curated meetings on demand content and more the inbound. Twenty twenty one conferences hosted with love by hub spot and it takes place on october twelfth through the fourteenth to learn more about the inbound. Twenty twenty one lineup or to register for the conference for free could inbound dot com. that's i n. b. o. u. n. d. dot com that's inbound dot com and this podcast is also sponsored by lincoln marketing solutions. Let's pretend for a moment that you're about to launch a campaign. It tested well. Your entire team is happy and everything is going. According to plan except for the one thought in the back of your head. How do i ensure that the people i want to target are in the mindset to receive my message the answer is linked in because when you market on linked in your message reaches the people who are ready to do business and that means that your advertising campaigns will work as hard as it can as soon as you launch it there over sixty two billion decision makers on linked in and that's why seventy eight percent of to marketers rate linked in as the most effective social media platform for their organization. I personally use linked in both as an organic growth channel to syndicate our content. And i use it as an advertising channel to retarget tech podcast listeners with some offers from our sponsors and hey the messaging platform is a great way to reach potential podcast guests as well so look whatever. Your marketing goals are from brand building to lead generation. Lincoln will help you reach your audience with the features that you know about like targeting by job title company name and location but you can also re target based on custom triggers from your website or even by uploading your crm lists and customize your campaign based on the action that you want the customers to take so look do business where business is done and you can get one hundred dollars of advertising credits towards your first lincoln campaign. there's some terms and conditions apply. But you can get by going to lincoln dot com slash mark again. It's one hundred dollars off of your first linked in campaign by going the lincoln dot com slash. Mar tech all right. Here's the second part of my conversation with dr duane. Veron founder and ceo of media. Science dr duane. Welcome back to the tech podcast. It's great sitting back excited to have you back on the show and excited to continue our conversation. Yesterday we talked about consumer research and specifically how media science is taking advantage of some of more cutting edge technologies monitoring the signals that people give while they are exposed to media and some sort of stimulus. Our dear hands get sweaty. Does your heartbeat go down. Does your mouth go up. When you see an ad those are all different signals. That tell us different reactions to an add. A political statement Somebody's presence so consumer research is being changed pretty dramatically by some of the testing and science. That were able to do. There's also a fair amount of artificial intelligence and machine learning that is going into the mar tech industry. Let's start off by first off defining artificial intelligence and machine learning. What are the two of them. And then i'm also going to ask you. Obviously what's the impact. They're having on the martic industry but start off with giving me the definitions of a nfl well rather than definition. I might just talk through. Because i think people here these things that they don't know necessarily what it entails and where the real revolution it's how it is that this kind of like turbo charges. everything. I mean it's really powerful implications. I hear artificial intelligence. And i think oh that stuff that google facebook twitter microsoft apple net flicks are doing and that everybody else says they're doing but really they're talking about marketing automation machine learning. Am i wrong. I feel like artificial intelligence. You need this. Giant data set and.
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"A little bit more about whether you think that association is formed so the speed of response houses law so these are the kinds of measures that we get fascinating when you think about the consumer research aspect. You mentioned that mars and eminem. This is a large scaled. Multibillion assuming a multibillion dollar company obviously great brand recognition and so they're testing individual adds to understand what's going to perform better in market. What are some of the other types of customer research that marketers have been using media science for well pretty much. Every major tv network group is there has been a client of arts and a lot of where we have really specialized than developed expertise. Is anytime somebody comes up with. Something new and innovative in fact over the past decade pretty much every major advance. The tv advertising industry was first tested at media science. So this is things like. If you have a picture in picture at that's inserted say you're watching a football game and it's the end of the game if you push shaking hands or there's a rough call and there's a short That might've purifiction picture. You know inside the game experience itself. We've tested that. We know that that is actually more effective than traditional ad. Pause adds anything you can think of. That's a major innovation. That's happened in the past decade. That was first tested new science. And that's a big part of what we do particularly because in many cases were testing something that doesn't even exist in the world so we have the software engineers to kind of like make it happen so to speak and we toss it so. I think anytime people encounter a problem where they're looking at something thinking. Oh my god. How do we deal with this. Where the people that they call. So that's where we particularly specialize you know..
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"And you could see that in the data so there are things we don't have to go but it is really fascinating. Is you do stops that. You can't see in the world said hear what you're saying. And i do wanna take us away from politics but i understand how understanding how people are feeling and some of their physical reactions to media whether it be advertising whether it be to politics and measuring that i could see as being a very powerful tool now. I guess the question that i have is you're talking about measuring somebody's heart rate in a lab setting. That is not meant to be comfortable but clearly you have to have some sort of machinery hooked up to a person to gauge. What their responses. So look if you come in you tape and ekg onto me in. You're putting measures on two of my fingers and telling me to scroll through my facebook feed. I might feel differently about that. And i might be. Let's just have a different level of comfort in the lab knowing that. I'm being an lies than i would on my couch when i'm being my true and authentic self. How much do you think about or try to gauge. What the variant is based on the lab setting a knowing that you're having somebody that is clearly being tested as opposed to experiencing the media in the field. What's the difference between lab and field data. it's a great question and it is valid issue. It's one of the reasons why media science philosophically we want to make the measures as noninvasive as possible so even though we measure he e. g which require sensors on your head even though we have methods that are a little bit more aggressive in terms of making up these measures. Our philosophy is to use the measures which are least invasive specifically because of the very reasons that you talked about we wanna make the experiences naturalist possible thus for example the reason why we go to any sense create like a mock living room experience for people to get them absorbed in their normal kind of like encounter..
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Here's the first part of my conversation with dr duane. Veron co founder and ceo of media. Science dr duane walk the martic tech podcast. It's bigger thanks for having me excited heavy. As our cast excited to talk a little you know of the more technical side of marketing. This is the mark tech podcast. Normally kind of focus on the mar part. And you're going to bring some tech influence here in the sense of machine. Learning artificial intelligence the more sophisticated technologies we use..
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Dr duane walk the martic tech podcast. It's bigger thanks for having me excited heavy. As our cast excited to talk a little you know of the more technical side of marketing. This is the mark tech podcast. Normally kind of focus on the mar part. And you're going to bring some tech influence here in the sense of machine. Learning artificial intelligence the more sophisticated technologies we use. Let's start off talking a little bit about media. Science in the description of this podcast. I mentioned biometrics facial expression i tracking. Eg g. those sound like really complicated technologies. how are they actually being used in marketing. I mean they are complicated. Of course the issue that we address in our research is that when you're talking about marketing above all your taxes back human emotion but the tools that we use to get to human emotion usually depend on self report in other words whether it's a focus group for a survey or interview were relying on woke people. Tell us about their most journey. The problem is people lack an understanding of own motionless journey. So when you ask a person a question about how. They feel about something what they're giving. You is the rash on reputation of what they think. They must be feeling. And that's actually far removed from their actual emotional encounter so what we do at media. Science is we want to measure that emotional response directly rather than being just depended upon what people tell about it. So the tools that you mentioned are all tools that are designed to get at measuring that emotion directly. I mean they are fairly complex. One of the reason they're complex is because they very pressing the person so you can't do this against a generic set of measures you have to actually calibrate for the individual. And then you have to actually let the that individual's response relative to their universe so to speak so that you can situate them in terms of what it means for them against the data but very exciting because it just exposes layers of data that we don't see otherwise
MarTech Podcast
How Technological Advancements Are Changing Consumer Research With MediaScience CEO Dr. Duane Varan
"Dr duane walk the martic tech podcast. It's bigger thanks for having me excited heavy. As our cast excited to talk a little you know of the more technical side of marketing. This is the mark tech podcast. Normally kind of focus on the mar part. And you're going to bring some tech influence here in the sense of machine. Learning artificial intelligence the more sophisticated technologies we use. Let's start off talking a little bit about media. Science in the description of this podcast. I mentioned biometrics facial expression i tracking. Eg g. those sound like really complicated technologies. how are they actually being used in marketing. I mean they are complicated. Of course the issue that we address in our research is that when you're talking about marketing above all your taxes back human emotion but the tools that we use to get to human emotion usually depend on self report in other words whether it's a focus group for a survey or interview were relying on woke people. Tell us about their most journey. The problem is people lack an understanding of own motionless journey. So when you ask a person a question about how. They feel about something what they're giving. You is the rash on reputation of what they think. They must be feeling. And that's actually far removed from their actual emotional encounter so what we do at media. Science is we want to measure that emotional response directly rather than being just depended upon what people tell about it. So the tools that you mentioned are all tools that are designed to get at measuring that emotion directly. I mean they are fairly complex. One of the reason they're complex is because they very pressing the person so you can't do this against a generic set of measures you have to actually calibrate for the individual. And then you have to actually let the that individual's response relative to their universe so to speak so that you can situate them in terms of what it means for them against the data but very exciting because it just exposes layers of data that we don't see otherwise
MarTech Podcast
"dr duane" Discussed on MarTech Podcast
"Talk about the science of machine learning and marketing joining us is duane baron. Who is the founder and ceo of media science which is a lab based audience research provider incorporating range of narrow measures including biometrics facial expression analysis eye tracking eeg and more with state of the art labs in new york chicago and austin media. Sciences discovering actionable insights and advertising technology media and consumer trends and today. Dr duane and i are going to discuss how machine learning and artificial intelligence are changing consumer research but before we get started with today's interview. I wanted to share some exciting news as part of our relationship with the hub..