Episode 261

podcast photo thumbnail
1x
-15
+60

00:00

00:00

A.I. As agency owners and leaders, we understand what it is and what it does, but we haven’t quite figured out the best way to use it in our own agencies for ours or our clients’ benefit. Many still believe it’s something inaccessible or unaffordable. But, A.I. can transform our deliverables, change the ROI we deliver to clients, and make us even better at what we do. Today. At an affordable price.

My guest for this week’s episode is Neil Sahota, a professor at UC-Irvine in California. Neil is a specialist in artificial intelligence marketing. He was an IBM Master Inventor and is currently doing some astounding work as an A.I. Advisor for the United Nations. Neil authored the book, Own the A.I. Revolution, and is one of the foremost expert on emerging technologies as they relate to business strategy.

In this episode of Build a Better Agency, Neil joins us to share his journey and technical expertise. He explains what agency owners can do to harness the power of artificial intelligence marketing so we can use it in ours and our clients’ favor. He also walks us through some of the most progressive A.I. tools marketers are using to communicate with their audiences more effectively.

A big thank you to our podcast’s presenting sponsor, White Label IQ. They’re an amazing resource for agencies who want to outsource their design, dev or PPC work at wholesale prices. Check out their special offer (10 free hours!) for podcast listeners here.

Artificial Intelligence Marketing | How agencies can use A.I. to communicate more effectively

What You Will Learn in This Episode:

  • How we as agency owners can harness the power of artificial intelligence marketing and use the technological disruption in our favor
  • How to use A.I. to get data and use it in your client interactions
  • What tools agencies can use to leverage data and turn it into ROI
  • Where agency owners should look for guidance on what A.I. can do and how to do it
  • What inspired Neil to write his book, “Own the A.I. Revolution”
  • Neil’s perspective on the ethics of A.I. in marketing

The Golden Nuggets:

“We live in an age where people are giving away their data all the time. We need only a little bit of information to construct a model using A.I.” @neil_sahota Click To Tweet “Start small to get your feet wet with A.I. so you can see what the effect is. Often people see that the results are mind-blowing.” @neil_sahota Click To Tweet “Artificial intelligence marketing is not going to solve all of your problems. But it will give you another powerful tool in your toolbox to communicate with different audiences more effectively.” @neil_sahota Click To Tweet “Even if you are already doing a microsegment campaign, A.I. can help you subsegment that to the individual level.” @neil_sahota Click To Tweet “Social media companies are now using A.I. tools to create packages we can use to buy a customer base of specific demographics and psychographics.” @neil_sahota Click To Tweet

Ways to contact Neil Sahota:

Speaker 1:

Welcome to The Agency Management Institute Community, where you’ll learn how to grow and scale your business, attract and retain the best talent, make more money and keep more of what you make. The Build A Better Agency podcast, presented by White Label IQ, is packed with insights on how small to mid sized agencies survive and thrive in today’s market. Bringing his 25 plus years of experience, as both an agency owner and agency consultant, please welcome your host, Drew McLellan.

Drew McLellan:

Hey everybody. Drew McLellan here from Agency Management Institute. Welcome back to another episode of Build A Better Agency. We are going to mind expand today. That’s my goal, is for us to think about things in a way where that we have never even thought about before. And I think our guest is going to help us do that.

Before I tell you a little bit about him and we get into the conversation, a couple quick things. We are doing, every week, we give away something from one of our podcast guests. So, a lot of our podcast guests have written books, or have courses, or done other things that they very generously give us to give to you. So, all you need to do to get in that drawing, and by the way you only have to do this once and then you’re in the drawing forever, is head over to agencymanagementinstitute.com/podcastgiveaway. And you just fill out the little form, give us your name and your email address and we will take it from there. So, again, giving away a lot of good books, a lot of courses. We’re giving away my new book, Sell With Authority. So, lots of great prizes and all you have to do is sign up, all right?

Let me tell you a little bit about today’s conversation. I think one of the challenges, one of the places where agency owners feel a little ill equipped or, I don’t know about you, but I feel like I get it. But I don’t get it. So, I’m talking about AI. I get in general what it is. I get some of the things that it does, but I don’t really feel like I understand all the things that it does. And I certainly know that there are probably things that, ways that we could be using it inside my own agency, ways I could be advising all of you to use it inside your shop that is not on my radar screen.

So, that’s what I want to talk about today with our guest, is sort of what’s possible? And how is AI changing expectations? How can it change our deliverables? How can it make us better at what we do? So, my guest is a gentleman named Neil Sahota. He is a professor at UC Irvine in California, but he has worked at IBM. He’s currently working with the UN to do some really amazing projects using AI, which I want to ask him about. Even though it’s not really marketing related, I want to chat with him about that as well because they’re doing some really remarkable work. Really my goal for our conversation today is just to give us things to think about.

So, with that, let’s jump into the conversation because I have more questions than we’re going to have time. So, let’s just get right to it. Neil, welcome to the podcast. Thanks for joining us.

Neil Sahota:

Hey! Really excited to be here. Thanks for having me on Drew.

Drew McLellan:

So, give the audience just a sense of your background, because I know you do some interesting things at UC Irvine, and how you came to write the book. Then, I have a million questions to ask you, so we’ll get to that. But first, introduce yourself to the audience and help them understand how you came to have this knowledge and insight.

Neil Sahota:

Well, I’ll tell you I cut my teeth as a management consultant helping global Fortune 500 companies with their business and marketing strategy. And I was always the kind of guy that likes to solve the problems more generally. About 14 years ago now, business intelligence was taking off. They had a lot of these sea level folks telling me, “Hey, it’s amazing the insights machines are giving us.” I’m like, “Machines aren’t telling us anything.” They had the tools that collect and store, and slice and dice, and write quick and precise reports. But it’s not telling us anything, but could it?

And it kind of put me on this quest to figure out a way for machines to do that. So, I went developing a lot of intellectual property, a lot of patents around what we call machine learning and AI today. And one day I gained a call from IBM RND about my work. The next thing I know I’m working on something called Watson.

Drew McLellan:

Wow.

Neil Sahota:

I was part of the original Watson team. That’s how I got my foray into AI.

Drew McLellan:

Not a bad foray, to start with IBM, right?

Neil Sahota:

Not at all. It was a fantastic endeavor, and when we won the Jeopardy challenge the question was, “What are you guys going to do?” Right? Because we spent a ton of money and we’re like, “Well, we’ll get back to you.” Because we never thought that far. But I was one of the people advocating to create an ecosystem, you know? Open up the platform, let the people and the business, expertise, the business problems come in; help them figure out solutions.

Drew McLellan:

So, how did you end up becoming a professor, or was this all concurrent?

Neil Sahota:

It was total accident. I enjoyed teaching, but I was just teaching those continuing education classes.

Drew McLellan:

Sure, yep.

Neil Sahota:

Back in the day I was advocating about we’ve got to teach more classes in analytics, some of these things to the MBA students. Full disclosure, I’m an MBA myself. And, one day they kind of called my bluff, so to speak, and said, “Hey, we’re talking about putting this class together. Can you help us out?” And so, I did help them put together, it’s a two year approval process that I learned, and when it got approved they called me up, said, “Congratulations.” They’re like, “You’re the only guy qualified to teach it. You want to do it?” That’s how I wound up breaking into a professor.

Drew McLellan:

Wow. And, before the interview, you were telling me that you split your time between the business school and the law school, right?

Neil Sahota:

Yeah.

Drew McLellan:

Yeah.

Neil Sahota:

But not a lawyer, never went to law school, but law is a great example of how much some of the professions were changing.

Drew McLellan:

Yeah.

Neil Sahota:

That there was a recent case a couple years ago where it was a murder and the only “people in the room” were the victim, the killer and Alexa. So, the DA essentially called Alexa as a witness. They allowed it. So, then the defense attorney is like, “Whoa, how do you cross examine Alexa?”

Drew McLellan:

Right. That’s crazy. Yeah, our world is changing so fast. So, I know you also, the culmination of all of these experiences for you, so far anyway, is that you just have recently published a book, right?

Neil Sahota:

That’s right, I published a book called Only AI Revolution, and it’s meant for a non technical business audience. In large part, I saw lots of companies struggling with the same questions, right? I’d hear it every time I worked with them. I know I need to be doing something with AI, what is it? And when I figure that out, how do I actually get started?

Drew McLellan:

Yeah.

Neil Sahota:

And most of the books I saw were very technically oriented or they’re more about the viewer’s concerns, not something that’d be helpful for people that want to figure out what to do.

Drew McLellan:

Awesome. That is exactly, I think, our audience and that’s where I want to sort of start our conversation, is I think a lot of agency owners and leaders are very familiar with AI. They’re trying to learn as much as they can. They know that it could and should be able to help them in their business. They just don’t know where or how. And I think the assumption is that it’s too expensive for a small agency to tap into.

So, let’s start with, I know you have sort of a premise or a belief that the disruption that technology is bringing to all or our worlds is actually being driven more by marketers than by technologist. So, talk to me a little bit about how we are driving the disruption, and maybe how we can harness and use the disruption?

Neil Sahota:

It’s interesting that one of the earliest adopters of AI was actually digital marketing. Right? We always talk about it’s a volume business.

Drew McLellan:

Right.

Neil Sahota:

Right? The more reach we get, the more eyeballs, the more conversions we’ll have. And with AI technologist coming out, like personality insights, the [inaudible 00:08:37] graphics and your linguistics you had, probably six years ago, a lot of sharp marketers realize, “I’m busy trying to construct this 360 picture by looking at all this outside information. Suddenly I can get information about kind of the inside.” You know? The psychology of the customer. So, I don’t need to worry about collect models and all these so much; I can actually know the person. I don’t need to worry about volume anyway. I know exactly who to target.

So, they want to call it individualized targeting, or some people call it precision marketing. The fact that we can know each person at a psycho graphic level, know what words to use, what they’re going to value the most. Are they going to care about the features? Are they going to care about cost? All these things, it’s become a really powerful tool because rather than say, “Okay, I want to go out and market to 100,000 people and hope for a 10% conversion rate.” Now, it’s like, “I can go out and market to 20,000 people and get an 80% conversion rate.”

Drew McLellan:

So, how do I scratch the surface of that? So, if I said, “All right, I’ve got a client that they’re game to have that kind of conversation about the psycho graphics and the neuro linguistics and all of that.” How do I get that data and how do I use it?

Neil Sahota:

So, the good thing, and maybe some of the thing’s the bad thing, we live in an age where people give away their data all the time.

Drew McLellan:

Right.

Neil Sahota:

You need just a little bit of social media information about a person to be able to construct this, right? You could create a whole psycho graphic profile across 56 personality traits with less than 2,000 characters, right? So, if you can ask for Twitter handle, LinkedIn profile, Facebook posts, or you could even ask them a couple of simple survey questions when they write out your census. All this actually gives you the data you need to actually construct these profiles on them.

Drew McLellan:

So, I get the data and I put it into a what? Or I run it through a what? So, probably a month or so ago, maybe two months ago, we had a guest on and he was talking about Crystal and how that draws the data from LinkedIn and gives you kind of a frighteningly accurate look at a person’s personality. So, how do I do that in a more aggregate sense, rather than I don’t want to just know what Drew’s personality is on Crystal, but I want to, let’s say to use your example, I want to talk to 20,000 people and I want to figure out which 20,000 I should talk to, to get even a 50% conversion rate.

Neil Sahota:

So, it’s an interesting question there, Drew, because the guys who’ve built a lot of the tools expect you to have data. Now, you’re seeing more and more companies, even some of the social media companies start to put that data together for sale. So, if you’re saying, “I’m looking for these type of people, not just a demographic level but a psycho graphic level.” Right? They’re actually now using some of these tools to construct those things for you so that you can actually buy the people, so to speak, buy the customer base based on demographics and psycho graphics. So, they’re actually doing some of the hard work for you now.

Drew McLellan:

So, what are some of the tools? Because I know before we hit the record button I was saying I think a lot of agencies want very much to use AI, they just can’t figure out how. And their assumption is too expensive. So, what are some of the tools that you see that agencies can take this sort of data and then use it for their client’s, obviously, ROI advantage?

Neil Sahota:

I mean, there’s some big, big notable names. All this stuff I’m talking about are literally pennies to the dollar. So, they’re meant for even the smallest of the small businesses. But you have IBM Watson [inaudible 00:12:41] the insights that help you do the psycho graphics. There’s actually a company called [inaudible 00:12:46] AI that ironically was started by a therapist, a neuro linguist to help depressed and suicidal teenagers. But what they realized is they actually constructed was an AI communication coach. So, with their actually API, they can actually take what a person is saying and figure out what they value, what’s the best way to communicate, what they’re going to care most about it, what their commitment level is. And they don’t need a whole lot of data to actually do that.

They actually did a test with an auto dealership because no one seems to believe you could do something like this. So, they said, “Okay, we’ll predict who will buy a car today, who will probably buy a car within a month, and who is probably just window shopping.” So, they went through the test. They were given 30 customers, and at the conclusion of the pilot the guy that owns the dealership came out and said, “Before I tell you how you did, I want to invest in your company.” Because they went 30 for 30.

Drew McLellan:

Wow!

Neil Sahota:

Yeah.

Drew McLellan:

Wow! I expected, because you were telling the story, that they did well, but 30 for 30 is pretty impressive.

Neil Sahota:

That’s really impressive.

Drew McLellan:

Yeah, yeah. So, let’s talk about how are you seeing marketers, I know because one side of the fence in your world in terms of both I know the book, but also that you’re teaching on the business side at UC Irvine; how are businesses and marketers… Tell us some of the ways they’re using these tools and technologies. What are they able to do that maybe we weren’t able to do before?

Neil Sahota:

Well, other than the inside look, actually goes to way it actually customize the message of a marketing campaign where we always segment it up, and we come up with the different messaging, maybe different visuals for each one. They have the ability to say, “Well, I still got to keep the visuals, but let’s say I chunk up my segments and I have a segment of 10,000 people. These tools are actually not giving us stability, and it’s automated to actually tweak that message for each one of those 10,000 people.”

I’m not just saying like, “Hey Drew, how’s it going?” Or anything like that.

Drew McLellan:

Right, right.

Neil Sahota:

But even pick what the specific words that’s going to resonate strongly with each one of those people. So, even though you’re still doing kind of a micro segment campaign, the AI is helping you subsegment that to the individual level. I mean, imagine just plugging in a few keywords here and there that’s going to up a person from maybe 60% to 85%.

Drew McLellan:

And what kind of a tool is someone using to do that? So, basically you’re saying if I have an email database of 10,000 people, if I have in essence scored them, and so I know different things about each of them on an individual level, then I’m going to write a newsletter article. I’m going to send it out, but this tool, which I’m about to ask you about, is going to say, “Okay, for Neil, you want to use touchy feely words. But for Drew, you want to use numbers and money words.” Right? Is that what you’re saying?

Neil Sahota:

Yeah, but it actually takes it one step further where it’ll actually suggest what word to use.

Drew McLellan:

Huh.

Neil Sahota:

Let’s just say, “Hey, you need to use a more touchy feely word.” It’s like this word over here you’re using, let’s say you’re using supports. If you actually use the word helps, it’ll connect stronger with this person. Right? The final decision on language can still be left up to a person, right?

Drew McLellan:

Sure.

Neil Sahota:

You don’t have to automate all the way through.

Drew McLellan:

Right.

Neil Sahota:

But that’s the level of sophistication that some of these technologies are. And that’s actually what Code Breaker Technologies is doing. When you want to send a marketing campaign or even write a followup email to one of your customers, that’s the level we’re actually at today.

Drew McLellan:

So, what are some of the tools that we could, if somebody’s like, “I want to do that.” What are some of the tools that we should be exploring that would allow us to do that?

Neil Sahota:

I mean, in addition to Code Breaker, and [inaudible 00:16:57] AI, there’s actually several tools. But you want to look at things that talk about psycho graphics and neuro linguistics. And I know those are big, heavy words. I’m not going to bore people with the science, but what people who tend to get caught up with is sentiment analysis. Right? And they think every time they hear sentiment analysis, this is what they’re really getting. But you’re actually not. Sentiment analysis just tells you how people are generally feeling. Do they feel positive, neutral or negative? But they don’t really give you these level of insights or give you that kind of coaching on how to speak with your customers.

Drew McLellan:

Yeah. Do you expect… I read an article the other day that was saying that within the next 50 years, AI is going to be writing music and TV shows and movies. Will we get to that point, do you think? I’m just curious. I know this is a little off topic, but I’m curious, because when I hear these things I think, “Where do people draw the line and say enough is enough?” And I don’t think we’re anywhere near that yet, but I’m curious what you think about how prevalent will AI be in the creation of all things that we create by hand now, by human, in the future? What’s your take on that?

Neil Sahota:

I chuckle off to myself because some of the things you’re talking about have already happened. I don’t want people to get me wrong. There are things that people are much better than machines at. Right? There are. Some of the things machines can do, we do find surprising. So, I’ll tell you two quick stories. Right?

So, one, my good buddy, Ross Goodwin, actually created an AI called Benjamin that wrote a screenplay for a short film called Sun Spray. Right? The language is a bit hulky and bit, but people like, “How in the world could it write an original screenplay?” Well, we can teach AI anything that we can commoditize and it turns out we can commoditize movies.

Drew McLellan:

Sure, there’s formulas, right?

Neil Sahota:

Yeah, every movie ever made fits into one of 12 archetypes. So you teach the machine those 12 archetypes, little bit about characters and dialogue, and you have a machine that could write a screenplay, right? They lack some of the deep emotion or connection, but it has the ability to do that. And we have AI that are already writing news articles and blog posts. And one of the things I do in my role as the UN AI advisor is we run the AI for good initiative, and last year in Geneva we had a big summit, AI and art exhibit. We actually had a performance. So, LGA Rich is a famous teaching. He came on with an AI that’s taught to help write music, right? So, it writes original music.

MEPS1, who’s a vocal artist came on. He said, “No one’s disrupted the human voice in 50 years.” Came on with an AI he did that’s disrupted voice and created new sounds. Then, you had Mio, who’s one of the greatest mime dancers in the world, happens to have a PhD in computer science, created-

Drew McLellan:

As all mime dancers do, I think. Yeah. Yeah.

Neil Sahota:

I think they go hand in hand.

Drew McLellan:

Yeah, right.

Neil Sahota:

Created an AI dancer that never saw a human dance before. So, it does some very human like dances, and some funky dances. And all three of them came together for a mutual performance. So, they had human AI music, you had human AI vocals, you had human AI dancer all together for a performance. So, some of these things that we’re talking about are not so far off, but don’t worry, we have not figured out a way to commoditize imagination or creativity.

Drew McLellan:

Right.

Neil Sahota:

Huge limitation for AI, and I’m not sure we’ll ever figure that out.

Drew McLellan:

And maybe we don’t want to. Maybe there’s a fine line between that, but it does remind us how powerful this is. And I think most agencies would say that they are not tapping into the power of this at all. So, let’s say there’s an agency owner who’s listening, who’s like, “You know what? We do a little bit of demographics, psycho graphics, that kind of thing. And we’re probably usi