Kirill Eremenko: This is episode number 281 with Founder and Director of Digital Strategy at Webfor, Kevin Getch.
Kirill Eremenko: Welcome to the SuperDataScience podcast. My name is Kirill Eremenko, Data Science Coach and Lifestyle Entrepreneur. Each week we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today and now let’s make the complex simple.
Hadelin: This podcast is brought to you by Bluelife AI. Bluelife AI is a company that empowers businesses to make massive profits by leveraging artificial intelligence at no upfront cost.
Kirill Eremenko: That’s correct. You heard it right, we are so sure about artificial intelligence that we will create a customized AI solution for you and you won’t need to pay unless it actually adds massive value to your business.
Hadelin: If you’re interested to try out artificial intelligence in your business, go to www.bluelife.ai, fill in the form and we’ll get back to you as quick as possible.
Kirill Eremenko: Once again, that’s www.bluelife.ai and Hadelin and I both look forward to working together with you.
Kirill Eremenko: Welcome back to the SuperDataScience podcast, ladies and gentlemen, super excited to have you back here on the show today. Today we’re talking about marketing. We’ve got the Founder and Director of Digital Strategy at Webfor which is a marketing agency in the US. Kevin is an expert in the space of marketing. He’s also a published author. His book, Future Proof Your Marketing just came out a few days ago.
Kirill Eremenko: In this podcast, we won’t be talking about data science algorithms or artificial intelligence methodologies and frameworks and things like that. If you’re after that, then this is not the podcast for that. This podcast is specifically about marketing and where this field is going. The reason why I wanted to invite Kevin onto the show is because marketing is one of the fields where data science is most heavily applied, where artificial intelligence is making massive impacts and it’s important to know, if you’re in this field or if you want to get into this field, it’s important to know where it’s going.
Kirill Eremenko: Kevin has some fantastic ideas about that. You will find out about what the future of marketing looks like and of course we’ll touch on the surface about some of the technologies, but overall you will need to make your own conclusions on what you will need to study or what you will need to incorporate in your data science career in order to break into this field.
Kirill Eremenko: However, in terms of the marketing space itself, it’s a very valuable conversation. Here are a couple of things that you will learn from this podcast. What digital assistants are and where they’re going with the help of people like Ray Kurzweil at Google. Kevin’s philosophy on what gets measured gets managed and what it means for marketing and data science. Why websites are less and less important, how segmentation is slowly transitioning to personalization, creating amazing customer experiences, disk profiles, natural language processing and computer vision and their role in the future of marketing.
Kirill Eremenko: That’s what this podcast is about. If you’re into marketing or you want to dive into this space and apply your digital skills there, this is going to be a very valuable podcast for you. On that note, without further ado, I bring to you Founder and Director of Digital Strategy at Webfor Kevin Getch.
Kirill Eremenko: Welcome back to the SuperDataScience podcast ladies and gentlemen, super excited to have you on the show and on the other line I’ve got Kevin Getch calling in from Vancouver. Kevin, how you doing today?
Kevin Getch: I’m doing very good.
Kirill Eremenko: It was so cool talking just now that you specifically told me you’re from Vancouver, Washington, not Canada, and I still managed to make the mistake. How often do people make that mistake that Vancouver, Washington and not Vancouver, Canada?
Kevin Getch: Honestly, more often than not, and even in the States when I’m talking to people in the states that are outside of the local area here they always think it’s Vancouver, Canada. You’re not alone.
Kirill Eremenko: Do you know the story behind this? Which city came first?
Kevin Getch: Honestly, I don’t know. I just know that we’re the better Vancouver. I love Vancouver BC and I travel up there quite a bit but I do think that they came before us. I want to say we’re about 100 and… We’re only about 114 years old or something like that. I think they came before us.
Kirill Eremenko: Okay, that’s very cool. There’s a city in Western Australia that recently celebrated its 70th birthday. So young. When you say 114 that is huge in my mind. Because simply Australia is such a young country.
Kevin Getch: US is pretty young too. That’s one of the things when we were over there and just the European side of things. I was really amazed at the history and how old some of those both buildings and cities were.
Kirill Eremenko: Yes, it’s crazy. I’m in Slovenia with my business partner we were traveling through some places in… Was it east of France? There are cities there couple hundred, 500 years old and so on. There’s cities in more like the Middle East zone that are thousands of years old. How crazy is that?
Kevin Getch: Yes it’s pretty insane. When we were over there in London, we stopped by Stonehenge and I’m looking at this and they’re saying… Basically it’s supposed to be about 4000 years old. I’m just like, man, it’s insane. I really love that history.
Kirill Eremenko: The thing that also gets me every time is paintings. Some of these paintings like Leonardo, Michelangelo painted hundreds of years ago. They’re still there and they still look the same as they did hundreds of years ago and they’ll be there when we’re gone. They’re paintings. It’s crazy.
Kevin Getch: It’s really cool. Who knows, someday we’ll be like that.
Kirill Eremenko: Maybe. Anyway, Kevin, we met at Tony Robbins Business Mastery 2.0, well not 2.0 but part two in Amsterdam, a couple weeks ago. How did you find the event?
Kevin Getch: Man, it was amazing. I had been to the first business mastery. There was some repetition, but there was a lot of new content as well. The repetition is really important. I brought my wife with me, she also owns her own business. It was really… For her it was a game changer which I knew it was going to be. I was really excited to have her there. Her business has been growing and I wanted to make sure she had the skills and tools to hit that next level.
Kevin Getch: Entrepreneurs were always being pushed to grow more and that was really eye opening for her. It was great for me too. The group I was with was amazing and just the speakers and the content. All the Tony Robbins events I’ve been to have been just top notch.
Kirill Eremenko: Absolutely agree. I think it is five days, every single one of those days was worth the price paid for the whole event. Totally awesome. What business is your wife in?
Kevin Getch: She owns a speech therapy practice.
Kirill Eremenko: Oh wow.
Kevin Getch: She mainly works with pediatrics, so children [inaudible 00:08:25] speech therapy.
Kirill Eremenko: Gotcha. Okay, very nice noble area to be in. You were one of… It was so fun meeting you. I was just standing there to get a tea, you were I think in line before me and you are one of the most nonconforming people. You were barefoot. You were walking around barefoot what’s up with that?
Kevin Getch: You’re like, “Hey, you from Austria?” It’s just funny because I think you’re standing up a lot during those events, and you’re jumping around and stuff like that. I just find it more comfortable to take off the… I had flip flops on that day so it was really easy. I just feel it’s more comfortable to take off the shoes. It seems it’s better for your body and posture overall. I just took it off and was walking around barefoot. I’m like, hey, why not?
Kirill Eremenko: That’s so cool. Well, it was great to catch up. I’m glad you could make it onto the show. So, Kevin, you’ve got a lot of experience and you’ve done a lot of different things and your book just come out, actually. Congratulations on that. Tell us a bit about yourself. You’re now a published author, you’re business owner, you’re in the space of digital strategy and marketing. If somebody off the street were to ask you who is Kevin Getch what would you say?
Kevin Getch: Well, if somebody off the street asked me I’d probably talk more a lot about my personal life and some about my professional life but I’d say, I’m a husband, I’m a father, I’m very passionate about helping people, especially focusing on my local community. I do a lot of work with local teams in the area. I speak at different events to help local teams like the Boys and Girls Club or we have another rock solid community teen center in the local area that I went out and spoke at recently. Then I also own a creative and digital marketing agency where we basically help small businesses and medium sized businesses make more money and really develop strategies that help them grow their business by more than 30% on average.
Kevin Getch: That for me has been a huge part of just enjoyment because my biggest driver is having a positive impact in other people’s lives. I’m able to do that and we have now a team of 14 people where I’m not only able to have a positive impact in our clients lives but I’m able to have a positive impact in their lives and in turn, they are able to have a positive impact in our clients lives and in the community.
Kevin Getch: Beyond blessed to be where I’m at, I have two amazing kids and a beautiful wife. If you ever hear me complaining, slap me.
Kirill Eremenko: Love it. Thank you for [inaudible 00:11:22]. Your book that just came out two days ago, well, as of when this podcast is going to be released, it will be two days. Future proof your marketing. Tell us a bit about that. What is it about?
Kevin Getch: It was funny because I actually want to thank the previous business mastery in Tony Robbins event for me doing this book because when I was at business mastery in Las Vegas, it was like, set some lofty goals for yourself and it was like take away all the limitations, all the excuses, all the different things that are in your way because being a business owner and a father and a husband and kids in sports and all that kind of stuff, there’s a lot of time there.
Kevin Getch: I’ve been making excuses. One of the things I wrote down was, launch my book in the next eight months finish and write my book. I knew it was a big task. I didn’t realize quite how big it was. I wanted to do that for a few reasons. One, obviously, there’s some self interest in just growing our brand and all that but the other part was really, I’ve seen this happen before. I’ve seen when we went… When I say we, when the world went from more of a print, marketing and it evolved into more digital marketing, there is literally hundreds and hundreds of businesses that just couldn’t keep up.
Kevin Getch: They didn’t know how to adapt, they were still doing the same old thing. The market was changing and it was changing fairly quickly but not as quickly as what’s going to happen. My concern is, then a lot of businesses that didn’t evolve ended up going extinct. Right now we’re about to go through what I consider, at least in my lifetime, the largest dramatic shift in our lifetime. In the next five to 10 years, we’re going to see so much change, it’s going to make the last century… In comparison, it’s going to make the last century look small as far as advancements both in technology and changes in consumer behavior and things like that.
Kevin Getch: It’s going to happen very quickly. I wanted to write a book that prepared business owners, marketing executives, so that they understood what is currently working, what’s going to change in the future, probably. Obviously looking at consumer behavior, looking at some of the large tech companies where are the consumers and tech companies pushing this marketplace, and how do we create a strategy that’s not only successful now but creates the foundation for what’s to come. That’s what really prompted me to write this book. I’m excited to share it. I’m hoping that a lot of people get a lot of value out of it.
Kirill Eremenko: Fantastic. Well, thank you very much. You said something is going to happen in the next couple of years? What’s going to happen exactly?
Kevin Getch: Telling you exactly is hard, but I don’t think predicting the future is all that hard when it comes to consumer behavior and tech and things like that. For me, it’s just is pretty easy to connect the dots, so to speak. What’s driving this change… I guess I’ll first lay the foundation of what I call the macro trends that help… Because I realized there’s so many different micro things are going to happen and there’s no way to really look at all those different things and really predict all those different things.
Kevin Getch: You can at a macro level look at what trends are going to drive the market, and then understand what’s driving those trends. In the book, I talk about what I call the three P’s that are these macro trends. Three P’s are personalized, predictive, and proactive. In the next five to 10 years, and I actually… In 2015, I originally came up with these three P’s as far as driving the trends, and they have definitely held true. I just think that we’re going to continue seeing an exponential growth in these areas.
Kevin Getch: Personalization obviously has grown substantially in the last four or five years and for the next five to 10 years is going to increase significantly. What’s driving all of this, to a large extent is advancements in technology like AI, machine learning, and some of the underlying currents that we’re going to see catch up even more the next five to 10 years is going to be advancements and computing power.
Kevin Getch: Right now there’s certain things that just from a processing standpoint, the amount of information, whether it be audio or video, it’s starting to happen more. In the next five to 10 years, computing power is going to see with quantum computing and a number of other things, we’re going to be able to see such a higher level of computing power, that it’s going to make processing everyday information and being able to take and record audio, just all day long and video all day long make it possible. By doing that, it’s going to allow for a complete change.
Kevin Getch: I can definitely dive into more details there as far as the specifics, but I think for me, I think it’s important to understand that the largest or the most fundamental shift for people is the proactive part. We’ve been in a reactive marketing phase for most of our lives. I mean if we want to go look for something, if we want to go search for a restaurant or something like that, we get online and we start searching for restaurants.
Kevin Getch: If I want to learn how to fly a plane, I have to go in and start typing in flight lessons, learning about flying, all that kind of stuff. It’s reacting to me, but in the future… We’re already starting to see this at a smaller scale. In the future, it will be much more proactive. We might be having a conversation and I have my digital assistant that’s in always on mode, it’s there to listen. It can take notes and be of use and value to me.
Kevin Getch: As it’s always on and listening. It’s basically processing all of that data and because we’re having a conversation about saying, hey, one of my goals is to learn how to fly a plane in the next eight years. It takes that information and it processes and later it will pop up and say hey, here’s some information on whether learning how to fly is right for you, and of course, here’s some ads for flight schools.
Kevin Getch: That’s just a small portion of what I think, to me, I think the large tech companies know that the Holy Grail for them is basically proactive assistance. If they have… If that’s set up in place, it will provide so much value obviously to the person who’s using it that, obviously, especially if they get the majority of it for free there may be advanced features, there may be the always on mode, maybe there’s other things you have to pay for some of this stuff.
Kevin Getch: As we go down the road, we’ll see how that plays out. Even just from a free standpoint, which I think will be the majority of it, all of a sudden that digital assistant becomes basically an executive who has an assistant. They become the gateway to all the other things. Now as you build that marketplace, you end up developing something that could be the largest… What I call a gateway channel to all other marketing channels, whether it be social media, or search or email or all these other things, you now have this gateway channel between those things.
Kirill Eremenko: We’re already seeing that I think in it’s very basic stages. For instance, I have Foursquare on my phone which helps me like find vegan restaurants when I go to a new city. It’s really very useful tool. It’s all user generated data and people rate these places and [inaudible 00:19:43]. Sometimes when I’m just even driving through a new city, even before I decide I want to have lunch, I want to find a place where to eat it, I guess it knows through my GPS location where I am, it knows about lunchtime and it pops up this notification, “Hey Kirill, based on your previous… place that you’ve eaten previously, you might like this place in this city.” I’m like, wow, that’s so cool. I didn’t have to do anything. I just click that place and I go there and I have lunch.
Kirill Eremenko: The recommendations are pretty outstanding there. I think, would you agree that we’re really seeing these predictive or proactive marketing in basic forms?
Kevin Getch: Yes definitely. Actually I use almost that same example in the book talking about some of the current ways that we’re already seeing proactive. Whether it’s your calendar, whether it’s your digital assistant popping up and saying, “Hey, based on the traffic and your next appointment, you’re going to have to leave in 10 minutes to get to your meeting on time.” That’s one of the ways we’re starting to see that.
Kevin Getch: The same thing with restaurants… The analogy I often give is, if you are going to actually hire… You’re going to hire a personal assistant that was going to work for you. It’s like, would you want that personal assistant to not really get to know you and just sit there and wait for when you ask them to do something and always just be waiting and then say, “Okay, yes, I’ll go get that for you.” No, you’re going to want them to actually really get to understand you, understand your needs, be proactive about serving those and taking care of you. If they’re really good, they’re going to predict your needs and offer up solutions even before you need them.
Kevin Getch: If they do that, then they become so valuable to you. That’s what these companies want to do. They want to create the best digital assistant you ever had. If they provide that much value… I’ve been speaking to audiences where I have a couple hundred people in the audience and about 94% of the people, 95% of the people would say, yes I’d definitely take that. Obviously, if it’s a free digital assistant, no question. They’re like, yes hook me up.
Kirill Eremenko: There’s quite a quite a few around, right? There’s Siri, there’s Google Assistant, there’s Cortana I think from Microsoft, there’s Alexa, which one? Which one do you choose?
Kevin Getch: Honestly, those are the top ones. Will.i.am, the singer, writer, artist whatever, will.i.am is actually creating a digital assistant called… I think it’s omega. He has 300 employees at least last time I checked, which was probably three, four months ago. He was working on launching this digital assistant. There’s not just going to be one and honestly, they’re definitely companies that have, I guess what I’d say, they have big strategic advantages in this marketplace.
Kevin Getch: Search is definitely the foundation of these digital assistance. Companies with a really strong base foundation of search can obviously come in better and provide better results because they’ve already have that huge amount of obviously data and experience in pulling up and providing useful information and useful answers. That is definitely a foundational element that gives them a strategic advantage, as well as size as well as other things.
Kevin Getch: We’re truly in a position where anyone could come in and potentially disrupt this marketplace. When I see something like will.i.am I’m like you never know. If it gets in the right market and finds a niche, honestly, even Bank of America in the US has its own digital system called Erica. Many companies are going to have digital assistance in their different ways.
Kevin Getch: As far as the standout ones, it’s funny because I look at Apple, and Apple has so many strategic advantages, but I feel like they’ve stopped innovating in the last couple years, or at least aren’t positioning themselves as much to take advantage of the things that they could and maybe that’s purposeful. Maybe that’s just part of the direction they’re going. They have a strategic advantage in a lot of way.
Kevin Getch: In the US, iPhone has a large percentage of the market. Whereas Google and Android operating system globally has a larger percentage of the market. Then obviously, Amazon has a huge advantage as well, for people shopping patterns. In home digital assistance, they actually have a larger footprint than a lot of companies do with Alexa in home. Each one has its advantages. For someone like Google who basically is just… Part of their strategic advantage is they’re the most used search engine, and pretty much soon the digital assistant will just be baked into everything there. They’ll have the largest, I guess, user base of people using digital assistance.
Kirill Eremenko: Very interesting. Basically, at this point, doesn’t really matter which one you choose. Just pick one and go with it.
Kevin Getch: I mean, everyone has their own preferences. If you want to get serious and be like, which one is the best, I think people have their preferences. From an accuracy and helpfulness standpoint, Google’s digital assistant beats all the others pretty hands down. Siri has normally been pretty bad. Whereas Alexa is in the middle. Cortana is even looking… It seems to me there may be getting… They’re trying to find their own niche outside of just being a general digital assistant. We’ll see what path that they go in.
Kevin Getch: When I talked to some of my friends, people at Google or people at Microsoft that are in the know in these areas. It’s really interesting, because they are very much so looking at not only like digital systems, but digital assistants that are, I guess, have emotional awareness based on the tone of your voice. At some point based on biometric feedback by being able to see facial expressions and things like that.
Kevin Getch: Google actually has a patent since 2012, actually, a patent on biometric feedback for changing their search results. They’re not using it at this point, but if at some point, they have a visual capability to see your face while you’re doing a search, and they see, your facial expression isn’t satisfied with the result. They could use that as feedback to determine maybe these weren’t the best results.
Kirill Eremenko: Wow. That’s crazy.
Kevin Getch: They’ve been thinking about this for a while. It’s not something that’s just came up. It’s just a matter of I think… Actually, one of the things I find really interesting is Google hired… What’s his name? Ray Kurzweil. Are you familiar with Ray Kurzweil?
Kirill Eremenko: Yes. The predictor of the future.
Kevin Getch: Yes, he’s got an 86% accuracy rate. Well, they hired him back in… Was it 2013 I think? It was around that time. The big thing that he was bringing to Google was one that was obviously his… A little backstory, Ray Kurzweil has never worked for another company. He’s always had his own business. Obviously, he doesn’t need to work for anyone.
Kirill Eremenko: Yes he created the synthesizer back in the day I think.
Kevin Getch: He’s actually created so many different things and been part of it. He’s a genius. He’s a futurist. He’s a genius. Why would he go to work for Google? Honestly, it’s because they have the, I guess, the most of the foundational elements, in order for him to create what he wants to create. In order to create a digital assistant that really understands the information, the people.
Kevin Getch: What I thought was really interesting was, everybody was like, “Why is Ray Kurzweil being assigned to the auto reply in Gmail?” He was in charge of that. It’s because of data science. Data scientists understand that. They look at that and they go, “Okay, what is the biggest data spot where people are having conversations, where I could actually train a machine learning algorithm to understand these communications that are going on?” Then I can also start providing responses and based on how people select those responses, I can create reinforcement around what was accurate and what wasn’t accurate.
Kevin Getch: That’s how they’ve been training the digital assistant throughout this time, is looking at conversations in Gmail and teaching. Basically learning how to have a conversation and how to provide responses. Those are some of the foundational elements as I look forward, and I see how far we’ve come already and then what the next five to 10 years holds. It’s really… To me, it’s really exciting. I’m excited to see where we’re going to go.
Kirill Eremenko: Those predictions are getting so good. Now they’re predicting the topic of the email. I was writing up an email earlier today. I click the subject line, and the topic just appeared and there was exactly what I wanted to put in there. Then I actually deleted it, tried to come up with something better just for the sake of my ego. I couldn’t. I just left whatever the assistance… It’s crazy.
Kirill Eremenko: You mentioned data science and I think this is a good segue to that. What’s the whole role of data science, artificial intelligence analytics in this new future of digital systems that’s coming?
Kevin Getch: Oh, man, it’s… That’s a really interesting question. I think there’s actually a lot of different aspects. That’s a big question because one, AI, machine learning is driving… is going to be driving this. It’s going to be one of the key drivers for these macro trends, pushing them forward. As far as the future, our ability to process massive amounts of data, and be able to intelligently at scale, provide solutions and good answers and good assistance throughout this process.
Kevin Getch: Then it’s all going to be done by obviously data scientists, machine learning, people who are professionals in AI and machine learning on that side of things. That’s all going to be driven by them and there’s obviously a shortage in that area. There definitely needs to… I keep telling people if you want to get in a growing field, a really in demand field where people will basically pay you to go to school to some extent, get into data science. Get into just AI, machine learning programming area things.
Kevin Getch: To me, that’s the backbone that’s going to drive things. I also think it’s important to understand analytics is going to have to change to some extent as we grow. Our organization is very analytic driven company. We want to make sure… We have a saying. It’s actually what gets measured gets managed. Something you hear throughout our organization all the time.
Kevin Getch: Whatever we’re doing from our marketing campaigns, we want to have measurement in place to be able to track what’s effective and what actions are the users taking and understanding what we want them to do and what they want to do and finding that sweet spot in the middle. As we move towards much more audio, as we move towards different modes of interaction, we’re going to have to capture different things to some extent.
Kevin Getch: Some people… We’re already to the point where if you’re looking for a restaurant, someone may never visit your website, because they might be searching on Google or Yelp, or Foursquare, like you mentioned. They might be looking at these different sources. They may never end up visiting the website. It’s important to understand and have the right analytics data and be able to pull all that information in so that you understand what’s actually effective for marketing. What’s getting us attention, what’s getting… Where are people taking the actions we want them to take. Whether it’s directions to our office, whether it’s phone calls, calling into our office, whether it’s someone filling out a form or making a purchase online.
Kevin Getch: All those things, to some extent, are trackable now. One of the areas that is really interesting to me in the future, and we’re seeing it right now with some of our clients is retail locations where we can actually get data around how many people are actually driving into the office based on anonymized data.
Kevin Getch: There’s a lot of different things we’ll see I think going forward that analytics is going to need to change to some extent to keep up with the changes in both consumer behavior and how they act and where they go, and all that kind of stuff. In 10 years… I definitely think the website is a critical component and it’s one that businesses own. They need to be very… They definitely need to be very focused on that.
Kevin Getch: But there are some businesses, and some of these are probably smaller businesses that may not need a website as much in the next 10 years. They’ll have these other platforms, they’ll have ways of doing it. It’s interesting, and it’s becoming fragmented and that’s part of the challenge, but I don’t know. To me, that’s exciting. Obviously, there’s a lot of opportunity for us to help businesses because it is getting so fragmented, but that fragmentation offers opportunity, because it allows us to target more effectively in some of those areas to.
Kirill Eremenko: Got you. You mentioned in your company, what gets measured, gets managed. Can you give us a couple of examples? What do you measure in your company about your clients and how does that help you manage those relationships?
Kevin Getch: Definitely. One of the clients we just brought on recently, they’re big in the education space. They have 148 locations around in the United States. When we first started working with them, we had seen that there had been a downward trend in their traffic. A downward trend in the number of conversions. People taking actions, in this case, filling out a form on the website.
Kevin Getch: They weren’t tracking phone calls really by channel. They didn’t know which channels were driving phone calls. They had all of their local listings. Their business name, address, phone number for all those different locations. They did have what are called UTM tracking parameters on those sites, but the way they were set up was wrong. It was actually putting everything in the wrong bucket.
Kevin Getch: I’ll give you an idea of what we did was we actually went in there and we switched over how they’re call tracking setup so that we can see… When someone comes to their website, we use dynamic number insertion so it changes the phone number based on the channel that they’re coming from. Whether it’s social, or search, which search engine so we can track and actually get data around where that information is coming from.
Kevin Getch: Then we also know from those calls, it’s integrated with their CRM data. We know ultimately, we can track to some extent, did those calls turn into a good lead? Then ultimately, did they end up becoming a client. We can close that attribution gap between leads and conversions and actual revenue, with a good amount of accuracy. That was one portion.
Kevin Getch: The other portion was they’re setting up their on site analytics correctly, moving them over to Google Tag Manager. Getting just better data, and then fixing all of their local listings, because the UTM tags… There’s little things and people don’t realize that sometimes in a UTM tracking parameter if for source, if they put in… The traffic in this case should be coming from a search engine, because it was all Google My Business listings.
Kevin Getch: They had something like GMB listing in there or Google My Business or something like that. What that does is it just throws it in other category when you actually look at analytics. This becomes a problem when you’re trying to understand and go up say 90% of your conversions are coming from other. We don’t really know what that is. Well, as we looked into it more we discovered what it was.
Kevin Getch: As we started to crack that UTM tracking parameters on their actual Google My Business listings, now they’re able to actually get real data. They see how many conversions are coming from it. We set it up so that they know that it was coming from each individual location. Then that individual location pulls into their CRM data as well. Then it’s about closing that attribution gap so they understand where are we getting our inquiries from?
Kevin Getch: Then are those inquiries turning into customers? How much revenue do those customers equate to so that they can look at what’s their actual return on investment? Because most businesses unfortunately don’t know. They know half my marketing is working, they just don’t know which half, right?
Kirill Eremenko: Yes. I guess what you said about driving those numbers all the way to the revenue and understanding where that revenue is coming from, and how it’s distributed across possibly different challenge channels, different types of customers, and so on. That’s a very important thing for data scientists to consider. It’s in the best interest and should be a part of a role of data scientist.
Kirill Eremenko: I think it is it, whether it’s explicitly stated or implicitly, if a data scientist is in the marketing space, what the business cares about is the revenue. Is the final bottom line. Nothing else. Good intentions or just good work and things like that. That’s not good enough. Results and specifically revenue that’s what really matters.
Kevin Getch: Well, ultimately, that supports their clients, and that supports their team. Because if they’re putting… Will say for a company, they’re putting $100,000 behind a marketing tactic that isn’t working. That’s not effective. It’s actually creating more costs for their customers, it’s bloated, they can’t pay their employees as much. If they can be more effective in their actual marketing tactics, they can better serve their customers, both the external ones and the internal ones.
Kirill Eremenko: Yep. What role does segmentation play in all of this? That’s a big part or one of the core value adds that data science can offer in space through machine learning or through other techniques on segmentation of customers. What kind of role does segmentation play in Modern Marketing?
Kevin Getch: Well, the role it plays right now is already really cool. The role it’s going to play in the future is going to be even cooler. Segmentation… Right now, you can segment by different customer types. There’s different ways to set that up, whether you have customers that are actually coming in and logging in based off their previous, I guess, experience with you and your website. You can start segmenting and changing the messaging for different audiences based on their previous experience with you in different ways.
Kevin Getch: Maybe they visited your website, and now they come back, you could say welcome back. If they put in, signed up for an email form in the past and put in their name, you could say, welcome back, Kevin, or you could start personalizing it a little bit more to them because you’ve gathered this data on them and you now know that, hey, this person is coming back and it’s Kevin.
Kevin Getch: When you start to personalize those things, it’s like if you send out an email to someone and just say, “Hey, you.” Or don’t even address them by name, the effectiveness rate of that drop significantly, but if you actually address it to them, personalize it to them, so it’s being sent to them, the effectiveness is much higher.
Kevin Getch: We’re already starting to see that with our current audiences. I still look at it as being a little… in its infancy, because you might have customers who are logged in or have accounts with you. Maybe it’s an E-commerce site, or maybe it’s a educational based site, whatever it may be. You can personalize based off that information. Knowing they have certain experience level with you, maybe they’ve reached a certain point with you, they’ve been a customer for so long. You can change your messaging. You can change your offers to this.
Kevin Getch: I find this to be awesome but part of me is also torn because I know what’s going to happen in the future and I already want it to be there. What I’d love to be able to do is where segmentation is going to go is, by utilizing machine learning is the only way I see this possibly happening, is at scale being able to communicate with large number of customers in a personalized fashion that is going to be more effective for every single individual as opposed to audience segments.
Kevin Getch: Right now, for instance, Google, will use this audience segments but they also personalize down to the individual level to some extent based off of maybe location or based off of search history and things like that. For us, for business owners with a website, we can start to personalize more effectively based of location and based of previous history with the website experience with you, and the business and the website and things like that.
Kevin Getch: That’s great but I think as Google’s algorithm or search engines algorithms become more down to the individual level where they get to know you, we also as businesses need to think about those three P’s and how can we be more personalized? How can we predict the customers needs and how can we be proactive, right?
Kevin Getch: Segmentation is going to become even more critical in the future because it’s going to have to be done at a algorithmic level and be able to communicate effectively. Right now, one way we do that is with chatbots. You can create chatbots but to me, at least, the experiences I’ve seen, are not as personalizes as I would like them. They’re not very…
Kevin Getch: Sometimes they’re very ineffective, honestly, then sometimes they’re effective because it’… It’s more of an if this, then that kind of communication, as opposed to a learning communication where it’s had 10s, or hundreds of thousands of conversations and it’s learned different patterns around people and how to communicate effectively with those people based on the words they use. We’ll be able to look and understand based on words people use, and maybe when they start communicating with us understanding what their disc profile is.
Kevin Getch: We might know what their personality profile’s like to some extent. We know that they prefer short, concise communications and they like these words, or we have someone else who might be on the far end of the spectrum where they’re very detail oriented, and they want a lot of detail. There’s going to be different ways that we can communicate with them in the future. That kind of segmentation is only going to be done, at least at scale, it’s only going to be done by machine learning which is going to require data scientist to help with that segmentation.
Kirill Eremenko: Very interesting. Basically, you’re saying that segmentation, as we see it today, where you have these cohorts of customers, is going to be replaced with personalization rather, where you don’t try to fit somebody into… Approximately, they fit into this group of people, or approximately they fit into this category of customers that we have. You find out more and more information about them and personalize your approach, your marketing, everything you can towards that person.
Kevin Getch: That’s exactly it. The only thing I’d say is you basically would be drilling down farther. You’d still have the categorization of this person is male and maybe in this age range, or female and they tend to be… There shopping for these general categories. You’re still going to have the affinity category that they’re in and you’re still going to have what they’re in market for. [inaudible 00:45:43] those larger buckets but then, with machine learning, with that capability, you’ll be able to drill down at the individual level, much like you and I would be able to and actually talking to an individual and really find out their specific needs.
Kevin Getch: What’s so powerful about that is, when you find out that individual’s specific needs, what’s actually important to them, and then you deliver that, then you’re creating this amazing customer experience rather than trying to just categorize them in this general category, which is what we have to do now and is more effective than not personalizing it, obviously. That’s the right way to do it. But again, we could fall into assumptions. Not every male who’s 32 years old is going to have these preferences, right?
Kirill Eremenko: Yes. Very good. That’s very [inaudible 00:46:36]. You said that, already now, you’re able to by the words people use, create their disc profile, understand whether there more detail or less detail, short phrases, and so on. Are you able to disclose what kind of technology, what kind of machine learning you use in order to accomplish that?
Kevin Getch: Yes, there’s a tool called Crystal Knows. It’s C-R-Y-S-T-A-L-K-N-O-W-S. Crystalknows.com and actually… I utilize this and it’s actually integrated with my calendar. I also have it set up through… There’s a extension in Chrome. If I’m on someone’s LinkedIn profile, I can click the button and it will estimate their disc profile based off of the information that’s in their social or their LinkedIn profile.
Kevin Getch: Actually, what’s funny is I’ve done a lot… I use it for interviews, I use it when we meet with clients.
Kirill Eremenko: Are you using it now?
Kevin Getch: Yes. I haven’t used it with you, but I should have. What’s really cool about it is, it’s actually really helped me in communication because I am… In the disc profile, I’m a very high D very high I. If I meet sit down with someone and they’re very high C, or high S, my communication style may totally conflict with their communication style. If I go in, and I know this person is, say, a high C, I know they’re going to be a little more cautious. They’re going to want a lot of details to make a decision. Whereas I’m just like, “Hey, let’s go, let’s move.”
Kevin Getch: That’s just my communication style. That’s how I am. That’s fine but if I want to help them, I’m going to need to communicate in the style that is effective for them. I’ve actually asked a lot of the people, both that I’ve interviewed as well as clients I’ve sat down with, and I’ve showed them, “Hey, this is what I came up with, is this correct?” And they’re like, “Oh, wow.” They’re reading it. They’re like, you’re not… It’s pretty accurate actually. I’d give it maybe an 80% accuracy rating which is pretty good for just going off of LinkedIn.
Kirill Eremenko: Interesting. Basically to… This is interesting podcast because you are obviously an expert in the space of marketing. I’m in my head trying to convert all these things that you’re saying about the future, which is very insightful about where we’re going. I’m trying to convert this into, alright, what does that mean for data scientists who are listening to this. What kind of technology should they be looking in for their future careers, if they want to get into the space of marketing or if they’re already there and they want to be successful?
Kirill Eremenko: Sounds like to me that a massive role is going to be played by natural language processing. Not only just a massive role, but in order to accomplish this segmentation to personalization transition, we are going to have to move away from algorithms like sessions. K means clustering, K nearest neighbors and other ways of creating these segmentations to something that creates personalization. That is predominantly natural language processing.
Kevin Getch: Computer vision, natural language processing, I think are probably two areas that are going to grow a lot. Obviously, natural language processing is more of the foundation because everything starts from that language. Actually part of Ray Kurzweil coming to Google was understanding the… Coming up with semantic search. He was big with semantic search and understanding word vectors and creating an actual framework for understanding the semantics, understanding the intent behind a user’s query.
Kevin Getch: There’s a lot that he came up with. That’s the foundation. What I think is interesting is I think, we’ll see a lot more in computer vision as well because part of the problem there is just processing power. It’s grown a lot so we’ve seen a lot. Again, you take something like Google Photos, gives you unlimited ability to upload your photos. It’s the one of the things I really like about what Google does is they’re like, “Hey, let me give you this free tool that’s just massive value, because now my phone’s not clogged up with all these videos and all these photos.”
Kevin Getch: Then they’re like, “Do you want the ability to be able to tag and sort and view and search all your images.” I’m like, yes, I want that ability. I’m sitting here looking to sell this elliptical machine. It’s in the garage behind a bunch of stuff. I’m like, I think I might have a picture of it somewhere. I just go on Google Photos and I search for elliptical and a photo pops up with the elliptical machine that’s in the garage only in half the picture but it still found it.
Kevin Getch: That’s the power of computer vision. That’s why they have such… They’re getting this large data set of everyone’s photos and then they’re using that to train it, and then based off your interactions with it, if it’s actually working, all that kind of stuff. As processing power improves, computer vision is going to come in a lot more. I think, always on video will be something that we’ll see in the future. Maybe farther down the road, but, and the ability to process all that information and see everything as we go. Tie that in with augmented reality and some of the other things and the ability to process and record that.
Kevin Getch: There’s definitely going to be a lot in both those fields. I agree 100% that natural language processing is foundational for all of this.
Kirill Eremenko: Fantastic. Natural language processing, and computer vision technologies. If you want to get into this space of marketing in the way that is going to be in the future. Any other trends in marketing that involve technology or data science that you’re witnessing currently that you’re anticipating?
Kevin Getch: It coincides but I think there’s still some delineation between chatbots and digital systems. I do think chatbots are definitely something that is going to continue growing significantly. Again, it’s one of those things where there’s definitely a close relation between chatbots and assistance to some extent. They’re still very clearly delineated in a lot of ways as well. I think right now, as I been utilizing chatbots a little bit more and working with them and testing them out for clients and things like that, I think there’s a lot of different methods.
Kevin Getch: One of my friends and colleagues, Larry Kim, created a company called Mobile Monkey. One of the things that they’ve done is basically created a chatbot that is integrated with Facebook Messenger. While that doesn’t seem… Okay, there’s a lot of chatbots, Facebook Messenger and all that, what I found really interesting is the ability to then tie it in Facebook itself and when you have a post… You might post on Facebook, and you put in something called a comment guard.
Kevin Getch: When someone goes in and comments on that post, whether it’s being pushed out, boosted or by an ad, or whether it’s just an organic post, a comment, you can automatically have your chatbot, send them a quick message saying, “Hey, thanks for commenting.” You can actually start developing that conversation automated. Then it has a whole funnel that you can bring them down and say, “Hey…” And start gathering more information. Are you a CEO? Are you an executive, marketing manager or a business owner maybe? You might ask that. They say that, then you can take them down this path.
Kevin Getch: To me, I honestly think that there’s so much value in that right now. My concern is it can come across intrusive in some areas. You have to be really careful as a business to not just go out there and be like oh, let me do this. Let me actually test this. If you go out to all your customers all of a sudden, and you have a chatbot that’s messaging them all these different things and it’s going through Facebook Messenger, some people may feel like Facebook Messenger is not the place to do that.
Kevin Getch: That’s the problem when you have segmentation. All those fragmentation, you have all these different marketing channels and all these different tactics. You have to not only understand what’s appropriate there and what’s going to be effective but how is my customer going to respond to this? Possibly starting out with a smaller test and starting to see how that works. Because I’ve had people message me through Facebook Messenger and I’m like, “Why are they…” I felt a little violated.
Kevin Getch: I’m like why are you calling me through Facebook Messenger on the weekend. I don’t get it. I barely know you. Why are you calling on Facebook Messenger? I think that’s big. There’s a lot of things that I’m really excited about in the area of… I guess, just overall. One of my main areas is just search. Organic search. Search engine optimization as well as paid search.
Kevin Getch: We have a team of specialists that do all this but for me, I think it’s really exciting to see where that is going. Machine learning is being used in search algorithms for quite a while. The technology, the ability to have better answers and questions. Some of the stuff is fundamental. When I say fundamental, it’s more I think in its infancy. In the past, if you would ask… You would go online, and you would search for the Statue of Liberty, say, “Where’s the Statue of Liberty?” Then you would do another search afterwards and you would say, “How tall is it?” It wouldn’t know what you were talking about.
Kevin Getch: It would just look for the words, how tall is it in a bunch of documents. This is where semantics changes it because you have this history. You have their search history, you know, they just searched for the Statue of Liberty. When you say how tall is it, it knows that you’re referencing the Statue of Liberty and will actually tell you how tall the Statue of Liberty is. That’s just a small example of some of the changes.
Kevin Getch: In the US at least, you can search for a hotel. If you’re using Google Assistant search for a hotel and it will ask you, “Do you know what dates you are going?” You can say I’m going these dates. Then it will say, “Would you like to narrow the things by amenities or pricing or ratings. Things like that.” You can start to narrow it. It’s all voice. This is all done by voice or prompts on your screen like a text message almost. Then it will provide you personalized results.
Kevin Getch: That’s already happening. I mean, to me, a lot of the stuff I’m talking about is already here, it’s just to the level. It’s growing at a different extent. To me, I think some of the areas in search that I think businesses can take advantage of is a lot of the searches. What Google’s trying to do is they move to at one point becoming more of what they call an answer engine where they come up with answers.
Kevin Getch: I think they’re moving away from answer engine and becoming more of a… They’re moving towards becoming an assistant. That’s the direction that they’re going. You can still take advantage of that by making sure that you’re creating the right kind of content. If you know what questions your customers are asking, you know what they’re interested in. If you really understand your customer really deeply, then you can provide better content and better solutions for those customers.
Kevin Getch: One of the, I guess, assertions that I make in my book is that customer experience will become one of the largest ranking factors in the next 10 years. The simple analogy I use is, if you were referring your friends to a business and they weren’t having a good experience, how long would you do that for? Probably… Not very long, right?
Kirill Eremenko: Yes, of course.
Kevin Getch: This is the same with a search engine, their business is dependent on how well they do at referring people to your business. They’re looking, understanding the customer understanding who they’re referring to and is this going to provide value for them? Is this going to meet their needs, the intent of what they’re searching for? The better you can do at providing that the better. Customer experience, whether it be when someone comes to your website, does it load quickly? Do they find what they’re looking for? Is the website laid out in a mobile friendly fashion?
Kevin Getch: All those different factors come into play, as well as how’s the content laid out and does it answer the question specifically or is it buried in a bunch of other information on the site, there’s so much that comes into play there.
Kirill Eremenko: Very interesting that you mentioned where Google is going because this actually echoes something that one of my previous guests, Khai Pham, on two podcasts before this one mentioned and he was talking about, in the 20th century, we lived in a world of information and hence that’s where the search engines came from. The 21st century, we live in an age of knowledge and therefore the next step from search engine is a reasoning engine.
Kirill Eremenko: Something that you ask a question… You can ask a question not just get like you said one short answer, but something that will do the reasoning for you and do the research and give you the appropriate descriptions, experiences, and whatever it is that you’re after to understand you much better. Sounds like the Google Assistant that it’s moving towards might be exactly that.
Kevin Getch: Yes I would agree with that 100%. The assistant is basically an advanced version that is, again, not even just reasoning, but personalized. They’re going to know if my… When we look for a restaurant, my wife likes to go to gluten free places, food preferences. If you’ve seen anything about Google duplex, it can call up and make reservations.
Kevin Getch: I could say, “Hey, find me some good restaurants. Some place I haven’t been to that you think we’re going to like.” It could go out there and search and say, “Based on people who have reviewed restaurants like you have in the past that like those, have also reviewed this highly.” They could use that information. They could say they have gluten free information and items on their menu for your wife. Would you like me to call make reservations? The system calls up and actually makes the reservation for you as well.
Kevin Getch: It’s little beyond reasoning and that’s why I like to call it assistance because it’s going to actually do actions on your behalf as well. You can ask it to go purchase something. You can tell it to call and make you a reservation. There’s a lot of things that’s actually going to do much like an assistant would.
Kirill Eremenko: Got you. Well, thank you very much for sharing that. This actually brings us to the close of this podcast. Just to recap, you shared the three main piece of marketing, where it’s going, personalized, predictive, proactive, and that’s all going to be enabled by technology. That’s where data scientists come in and they need to look out for trends that will facilitate this transition and will facilitate the future marketing.
Kirill Eremenko: That’s the best bet to be successful in the world of marketing that’s coming. Is to work on these skills, work on these technologies, such as we identified computer vision and natural language processing as a couple in order to be most prepared for the future that’s coming. Is that all right?
Kevin Getch: That sounds great.
Kirill Eremenko: Awesome. Well, once again, thank you so much, Kevin, for coming. If anybody wants to grab your book it’s called Future Proof Your Marketing. Just came out. Apart from that, where else can people find you, connect with you?
Kevin Getch: I love to connect with people on LinkedIn, our company website, webfor.com. We have a blog on there consistently and put up new information and resources and stories. As well as my own website, Kevingetch.com.
Kirill Eremenko: Awesome. Fantastic. Thank you so much, Kevin, appreciate you coming on the show today and talk to you later.
Kevin Getch: Thank you, Kirill, I appreciate you having me on.
Kirill Eremenko: Thank you, ladies and gentlemen, for tuning in to the podcast today and spending this hour with us. I hope you enjoyed our conversation with Kevin and got some valuable insights from what we discussed. My favorite takeaway was the whole idea behind digital assistants. Very interesting where the world is going. Possibly very soon our search will actually be… Not just search but a conversation with a digital assistant.
Kirill Eremenko: I think Kevin did a great job portraying what that will look like in terms of how we’re going to want our personal assistance or digital assistance to be personalized and therefore understand us better. Even though we all have these privacy considerations and we don’t want our data to be leaked or anything like that. At the same time, we will want our digital assistants to know us better so they can help serve our needs better.
Kirill Eremenko: That’ll be a very interesting future we’re heading into. As you can imagine, data scientists and artificial intelligence experts, machine learning engineers are all going to have a massive role in that. Hopefully, this podcast outlined what to look out for. Kevin’s book is called Future Proof Your Marketing. It just came out recently. Two days ago as of when this podcast is launched, two days before that. You can go and pick it up and learn more about Kevin’s perspective on marketing and where this whole field is going.
Kirill Eremenko: Of course, as always, you can get the show notes for this episode at www.www.superdatascience.com/281 where you will also get any of the materials mentioned on this episode plus the transcript. On that note, thank you so much for being here. I look forward to seeing you next time. Until then, happy analyzing.
