VOTF Episode 8: Artificial Emotional Intelligence?! How AI Could Help You Win Friends and Influence People.
About this episode
What does it mean when your spouse is covering their mouth? What about when your colleague is clenching their jaw? What if you had a cheat sheet for each of your daily interactions – with your partner, your colleagues, your prospects – and that cheat sheet made each conversation a little bit easier and more productive? That’s a real possibility with advancements in AI-based personality profiling, which could help us not only identify our own personality type but also how to engage those with similar or different styles. The applications of this technology are limitless, according to Behavior Sales’ Ben Bressington, and must be used judiciously. Tune in to decide whether or not you’d like AI to help you win friends and influence people.
Sam Zegas: Welcome to Deepgram’s ‘Voice of the Future Podcast’, AKA, ‘Our Favorite Nerds.’
At Deepgram, we’re obsessed with voice, and this podcast is our exploration of the exciting emerging world of voice technology.
I’m your host today, Sam Zegas, VP of Operations at Deepgram and our guest is Ben Bressington, founder and CEO at Behavioral Sales. Ben, welcome to the show.
Ben Bressington: Hey, thanks, Sam. I’m excited to be here.
Sam Zegas: Awesome. Alright, Ben. This is ‘Our Favorite Nerds’ after all. So why don’t you tell us a bit about yourself and specifically what kind of nerd you are?
Ben Bressington: So I’m the nerd that never got into the comic books.
You could call it the band geek-style nerd.
But I recently like, I only learned to skateboard when I was, like, 37 I think I’m like, hey, I need to learn a new skill. So, like, I’m the type of nerd that was, like, too nerdy to pick up a skateboard right? So I’ve even got into mountain biking and cycling, and I’m an avid cyclist. So, like, 150 miles a week is kinda my baseline goal. So that’s the type of nerd I am.
Sam Zegas: I love that. If I tried to learn how to skateboard today. I would break every bone in my body right down into, like, my ears. So–
Ben Bressington: Hey, I can promise you, like… you gotta forget what people think you look like. Like, I’ve had the land paddleboard sticks. So I’m like, I have the pads on, the helmet on, and, like…
My wife at the time would go, I’m not going outside when you look like that. I’m like, hey, like, I’m not falling off. Like, I this is not gonna be good for me if I fall off.
Sam Zegas: Hey, more power to you.
So you’re the founder and CEO at Behavior Sales. Why don’t you tell us a little bit about what you do and what the vision is for the company?
Ben Bressington: Yeah. So at Behavioral Sales we’re a behavioral intelligence company because one of our biggest… our vision is to be a conversation coach.
Sam Zegas: Mhmm.
Ben Bressington: Because we find that a lot of people struggle with how to have conversations in today’s market. Like, in today’s world, we’ve become so technology fixated but yet we’re so disconnected and isolated than ever before.
Sam Zegas: Mhmm.
Ben Bressington: And this is a huge problem because we’ve become ignorant to the emotional impact of that text message you just sent or that remark you make, which gets interpreted incorrectly, and you may miss it in an email. You may miss it in Slack. You may miss it on a Zoom call.
And when we say, miss it, it may result in you churning a client. It may result in you losing a deal. It may result in you having that deal you thought was so awesome. Ghost you. And you’re like, what happened?
Like, there’s with all the technology and the analytics we have in today’s world, there was no analytics giving us feedback on what’s happening in our conversations right now. What’s happening with our body language? What’s happening with this person, like, am I truly creating a connection?
Or am I overwhelming them and shutting them down? There’s a big difference. And our goal at Behavior Sales is to create this platform that gives you real-time insights into what’s going on with the people you’re talking to and showing you insights on how to better connect with people faster. So that’s what we do at Behavior Sales.
Sam Zegas: That’s really interesting. That’s something that even humans might struggle to do. And yet somehow, I can see that what you’re researching is like how do you develop an algorithm that can understand the, I assume, communication styles or social styles of the participants in the conversation so that you can give tailored advice to each of them.
Is that how it works?
Ben Bressington: Hundred percent. So, like, for example, one of the first things we started doing was predicting personality type.
And we took all of the personality research out there because there’s multiple from ocean to disc to the bird model, which is the simplest model that we used. And we integrated them into our entire offering because we found that a lot of corporate culture issues are created because of personality conflicts. And we can identify that in our language.
So one of the first things we did was, we converted an email you received from somebody or a tweet. We can deal with a tweet or a LinkedIn profile, and we can do a full personality assessment on that person, and they never answered a question.
So our AI can tell you, “Hey, you’re about to have a conversation with Sam. How do you connect with Sam most effectively?” because I have to speak Sam’s language, not Ben’s language, which is bad Australian. Right?
So the thing is this is what’s important and this is where we started and what’s been evolving as we move forward to be able to provide you insights into really creating a connection with people and having a quality conversation and understanding the emotional impact of that. I’m sending the body language impact of that if that makes sense.
Sam Zegas: Oh it does. There are a couple of bookmarks I wanna come back to there. One of them, I think you said bird models. So I’m gonna ask you about that in a minute.
And another thing I wanted to talk about is the body language component of this, so we’ll come back to that too.
I’m wondering first though, would you give us a demo? Could you maybe end analyze something about my online presence that would tell you what personality type I am?
Ben Bressington: Yeah. Hundred percent. So I’m gonna share my screen real quick and I’ll literally just walk you through it.
So you should be able to see my screen.
Sam Zegas: Yep. And for those following along on the audio-only version, this is now the behavior sales website. Yeah. Go ahead.
Ben Bressington: And I’ll do my best to narrate visually for everybody. So we’re just on the dashboard right here. And what I’m gonna do is on literally gonna run a personality assessment on Sam. So, like, I’m about to meet Sam for the first time. I don’t know him from a bar of soap. Right? Which is one of the most challenging things for an SDR, business development, customer support person to do.
Right? They have this fear and anxiety of actually a cold approach.
Sam Zegas: Mhmm.
Ben Bressington: So we’re trying to give people confidence in their ability to know who I’m about to talk to.
Sam Zegas: Mhmm.
Ben Bressington: So therefore, create rapport faster. So one of the cool things we can do is we can literally copy and paste Sam’s email address into our system, and it does a lookup to find a match on LinkedIn. And then within 15 seconds, we can create a report on Sam.
So that’s what this report looks like. I’m sharing it on the screen right now, and it’s giving us a summary, and it’s telling us Sam is an eagle. And I’ll dive into the four bird personalities very soon, but it’s a way we’ve simplified it for people. And as an eagle, it means he’s a doer, very direct He’d be classified as a D C on the disc profile, and on the ocean profile that’s a C A N.
He talks fast to the point, no-nonsense, which could also be some of Sam’s biggest conflicts.
Because if he meets with one of the other bird personality types you call a peacock, right? That’s a bit of a social butterfly. They like to build rapport. They like to talk about the weather. They like to find out what you did on the weekend. Sam’s sitting there going, hurry up, and get to the point I’ve got crap to do.
Right? So the thing is a lot of salespeople make this massive mistake of always presenting in their style and not the style of the person they’re communicating with, which is a big problem. Does that make sense?
Sam Zegas: Yeah. It does. This is really interesting technology you’ve developed. So without my answering any questionnaires or anything like that, what we’re looking at on the screen here is a behavioral profile of an effective way to communicate with me and positives and weaknesses about the way that I communicate.
And I feel very exposed in a good way by this because I think that this is spot on and it probably could give me some valuable insight about how I should approach conversations with people who have a different kind of a style.
Ben Bressington: A hundred percent, and this is the key to it. It’s like understanding their style and until I communicate in their style, I can’t connect to them. And one of the biggest problems we experience right now is it’s all, like, sales they treat as a one size fits all model. Like, they have one sales script but you should have four variations of that sales script based on the personality style.
Because, for example, you need to know how to talk fast to connect to that eagle. But to you also need to know how to talk slowly to connect to the owl who’s all knowledge-driven and wants the details and the blueprints and they want the engineering flow on how this stuff works.
You have to talk a lot slower and you have to give space for thinking, like even how you close them, they’re buying desires, they’re buying fears, they’re buying worries, are all different. And what I’m doing on the screen right now is I’m literally comparing my personality style, which is a peacock to an eagle. And it’s showing us side-by-side comparisons of the buying styles and how they’re different.
Like Sam commands respect, where for example, I’m looking for prestige and reputation and showiness. I want a little bit of flashy stuff to it.
Right? Where and the sales approach is completely different. Sam can make fast decisions, concise, very business-like where I want warm and friendly and I care about how things all look and make me feel.
Alright? So there’s a big difference. And If you don’t understand that, like, if you’re an eagle selling to a peacock, you make the mistake of talking your language, and therefore you won’t close them. And you may not even connect with them, particularly if they’re the influencer or the champion you’re trying to empower for your product.
And this is just out of the gate.
So imagine this, like, the first email you sent can now be a better email. Like, we’ve seen a 33% plus connection rate to our cold emails when we start to personalize them versus what we’re sending just that generic one-size-fits-all. These are cold emails.
Sam Zegas: Yeah.
Ben Bressington: So that’s massive.
Sam Zegas: Yeah. This is a, thank you for screen sharing this. For anyone who’s interested in learning more or finding out what their personality type is, you can visit getbehaviorsales.com. Right? I’m sorry, behavoirsales.com to take this test.
So I wanna pull out some of the themes about where the speech industry and speech technology is today and how you’re at the cutting edge of that. But first, let me just understand one thing about the product today.
So you’re building this more and more into a conversation coach. And right now, are you doing mostly post-call analysis? Are you doing some real-time coaching at this point? What’s the latest?
Ben Bressington: So we set everything up to be pre-call and then post-call analysis.
We did initially create a Zoom integration to our own Zoom app. And we found that in the initial beta of that application, the sales reps were more focused on trying to game the indicators that would then overlay on this screen.
Because what we did create, we literally created an overlay on your Zoom which can literally show you different body language. So for example, if there’s dis- if there’s disbelief or maybe seeing him needs reassurance in this moment. And if I don’t reassure him, that’s gonna cause objections or problems later down the conversation.
Right? Because things fester in our body and our words are always telling the truth. It’s just if we process it. Right? So we did that initially, but we found people were trying to game the dials and game the system rather than focusing on the connection, which is the goal of the call in the first place. So we can do real-time, but we found that the effectiveness wasn’t as great because we need you to focus on connecting on a conversation.
So that’s why we kinda just kept it as pre-call and post-call analysis. And bear in mind, through our partnerships like with Deepgram, the transcript of an audio or video call can now happen in seconds.
So the post-call analysis we can provide you can be within 5 minutes of a call. We’re telling you if you even had influence over the person you’re talking with or where you lost influence.
So now you go, okay, how do I follow-up if I lost influence with Sam? How do I regain authenticity?
Things like that, which is a totally different conversation you’re gonna previously have.
Sam Zegas: Yeah. And I can tell clearly the thing that you’ve gone to market with at this point and are developing your expertise in is things related to sales conversations, but this sounds like a tool.
Like, I would like a conversation coach for life, for my relationships, for my, you know, for friends, for things like that. And some version of this technology probably really applies to all human communication. Is that in the long-term vision for you?
Ben Bressington: Yeah. A hundred percent. Like, we would love to be universal for everybody. Like, a lot of people initially said, why don’t we start in the HR world? And after we were doing some initial research, we found that the budget for HR tools isn’t really there. Where’s the budget? It’s for sales. And where can we move the dial the fastest? It’s in sales.
However, from a company culture perspective, we can analyze your email inbox or we can analyze Slack, for example, and show you which companies I think, which employees are thinking about churning.
Sam Zegas: Mhmm.
Ben Bressington: We can predict churn of employees, and we can predict churn of customers based on language change.
And one of the biggest variables for that is, for example, the disgust ratio. If you trigger the emotion of disgust, you literally are starting to force somebody to lean out from you, run away from you.
And if you trigger it badly enough, you’re literally gonna force people to leave. And this is a huge problem particularly for remote teams. Like, are you really happy? Are you enjoying your work? Are you being fulfilled?
And as you said, like, this is something that you could apply to your family, your kids, your parents, like, hey, is my dad says he’s okay right now. But, like, is the last few text messages he sent me? Is he really okay? Like, there’s a myriad of applications for this? It’s just time and energy and the ability to expand. Right?
Sam Zegas: Yeah. Yeah. Very interesting. And we’ll get a little bit deeper to the partnership, the being behavior sales in Deepgram a little bit later. But clearly, Deepgram is working specifically on the audio component as everybody who’s listening to this will know that body language is also a major part of communication and something that people are more or less good at reading. And how do you incorporate body language into the model that assesses communication?
Ben Bressington: Yeah. So we actually created our own machine learning process based on the techniques used by the CIA. And it’s the process in which the CIA uses to interrogate black site interrogators. So I had the privilege of spending over 12 months being trained by black site interrogators who train, like, these crazy individuals on reading body language. And at the time, I’m like, oh great. I’m gonna just teach my sales team all these tricks.
And, like, we like, there was, like, over 120 different body language indicators that I had to learn, right, and try to master. And I realized that that’s not gonna happen for my sales team. Like, because they’re like, how does it impact my commission today? And how do I make that happen? But what I could do is I could train our AI to literally analyze a call or analyze a video and look for these body language indicators. Right? Which is where we started and where we’re evolving from. And so we layer that with linguistic analysis, and we literally perform on every cold transcript that we get provided from Deepgram over a hundred different types of linguistic analysis to determine personality, determine like, there’s way more than just emotion and sentiment. Right? There’s over 17 different emotions for happiness alone. So it’s like where are you placed on that scale? And it’s the changes in these you need to be aware of. A lot of people create like these sentiment tools, which are great. But, like, they misuse that data and they misuse it in making decisions and they miss spotting the change.
Like body language is all about change detection. It’s all about at what point did this happen? and it’s creating that contextual relevance, which is really powerful. But it takes a long time to master.
Sam Zegas: Yeah. Sentiment analysis in speech technology is a very difficult challenge that a lot of different companies are working on right now, including Deepgram and even just coming up with the base classification of what you look for? what are the buckets that you’re trying to be able to identify? Are very difficult in sentiment analysis.
But it’s a really exciting area of research, and I think that in the next 5 years or so, we’re gonna see a lot of advancement there.
Based on this experience that you had working with folks from the intelligence community, what are some body language signs that you should watch out for just in any conversation?
Ben Bressington: Yeah. So there’s some very simple ones. One of the easiest ones to look for is what they call a lip compression. And our lip compression means, like, I’m biting down on my lip, and I’m actually… most you actually see this when a baby says no to breast milk or when they’re being fed. Right? Like, they close their lips and you see this in politicians or people on TV all the time. And it literally means withheld opinion–
Sam Zegas: Mhmm.
Ben Bressington: –which is a really interesting thing. Another simple one you can watch for and look for this in a conversation is blink rate.
So if I am blinking slowly, right, very slowly, that means I’m focused, and I’m captivated by what’s going on.
Now, if I’m blinking fast, you’ve lost my interest.
And this is something to be aware of, like, so Sam and I are on Zoom right now. His blink rate is very slow. He’s captivated and he loves this body language stuff. Right? Because that’s what Sam’s passionate about. Right?
However, it’s when his blink rate changes. I need to work out what’s going on? Is Sam talking and the context has changed or am I talking and the context has changed? And that’s what I have to come back to. So there’s two indicators right there.
I’ll give a final third one for people if they wanna challenge themselves at Starbucks this afternoon or on their lunch run. So disbelief is really interesting.
And the thing with body language is your body is always communicating, and it’s more like a video. Right? So you can actually spot disbelief.
And it’s what we call a single shoulder raise. It doesn’t really matter any shoulder, but that shoulder will pop in conjunction with what you’re talking about.
So the thing is you then have to ask the question of, is it do I do they disbelieve what I’m saying? Because, like, they’re literally spending with a shoulder pop to the words coming out of my mouth? Or is it something they’re saying?
Like, you go, oh how was your day Sam? And Sam goes, it was alright. Like, his shoulder just pops. left or right shoulder really doesn’t matter. And it means he disbelieves or disagrees with the words that literally just came out of his mouth.
Sam Zegas: Mhmm. Interesting. But like a quick shrug. Yeah. Interesting. That must have been a really cool experience of learning about body language like that. And now you mentioned the linguistic layer which obviously is coming from audio plus this body language layer that you’re talking about.
Is it those two components together that create behavioral intelligence because I know that that’s one of the keywords and part of the motivator for what you named the company.
Ben Bressington: Yeah. Hundred percent because it’s more than just that spoken word. And, like, so for example, one of the things I mentioned already is we can measure through various sentiment analysis.
Authenticity. Were you presenting authentic authentically? Right?
If you weren’t, your deception score increases. So we can literally connect all these emotional things that people typically use sentiment and start to connect them to bigger things– deception.
For example, if the words you’re using are too buzzword-ish you could be trying to deceive. And you’re losing connection and authenticity with your audience, which is really important to understand if I’m a manager.
Like, one of the easiest ones we track is influence. Literally, do you have influence over the person you’re talking to?
Sam Zegas: Mhmm.