How to Use AI Roleplay to Improve Sales Conversations and Close More Deals with Paul Burani

Speaker 10:51

Let's figure out how to use the AI to beat the AI and keep us relevant in the future.

Speaker 20:56

Most of you are thinking that, oh, I'm going to train my people to use AI, but this is AI actually training your people to be better humans.

Speaker 11:03

What happens is if everybody has access to the same AI, powered by the same underlying data, we're all basically a commodity unless we think about ourselves as competing on our ability to leverage that technology. And that takes original thought. It takes creativity. And in the world of business, it also takes people working with people. We have a role play app that does this, it simulates conversations based on detailed personas that gives you objectives. You execute, you connect your experience to coaching and feedback, and ultimately the business outcomes. Can I onboard faster? Can I close a deal faster? Close the deal bigger, whatever it may be. And then from there, we're helping executives read the data at scale to understand the skill gaps that are persistent across their organization.

Speaker 21:46

I'm going to go to roleplay.atlasprimer.com and check it out myself, actually. Hey everybody, this is Chris Daigle. Welcome to Using AI at Work. This is going to be another fantastic episode. Before we get started, I know that in the uh ads that we have for our services, some of the links may have been off, particularly the 50K link. We fixed it. So if you tried it in the past and been like, but I want to talk to you, Chris, and it didn't work out, please try again. At this point, we've done 120-something episodes and we've covered a lot of topics. And what I've found is that most business owners that, you know, they're, they're on this thing, they're uh because they have a belief that AI training means teaching employees to use AI. But in today's episode, we're going to kind of flip that script and show actually how AI can be used to train employees to be better at human conversations, which uh is kind of the point of distinction for a lot of businesses. The IRL, the in-real life angle, is um becoming and the authenticity of a human conversation is becoming more and more desired by people who are getting inundated with long emails being generated by AI and agents and those sorts of things. And today our guest is co-founder and CRO of a company called Atlas Primer. And what they do is they actually sell an AI-powered role play and training platform. So again, remember, most of you are thinking that, oh, I'm gonna train my people to use AI, but this is AI actually training your people to be better humans. So, Paul, thank you for being on the show. And um uh we're excited to dig into this because this is not a topic that we've covered before. So before we jump into kind of the conversation, is there anything that you think the listeners need to know to kind of better understand the direction we're gonna be taking the show today?

Speaker 13:36

Good question. And thank you. Um appreciate being here. Um, this I've been looking forward to this. So, anything they need to know? I'm gonna say no. Let's let's go in a blank slate. Um, the only thing I'll say is open your mind, right? Because uh we're we're in this mood uh within within the the the workforce, within the sort of professional community, we're in this mood of looking at AI as sort of a race of incrementality. And uh we've seen evidence, right, that we can shave a few points of margin or we can, you know, we can find a few incremental gains, but really this conversation is going to be a lot more about a paradigm shift, like you said. Let's let's figure out how to use the AI to beat the AI and keep us relevant in the future.

Speaker 24:27

Well, you know, uh quick question, I guess. Before Atlas Primer existed, what were you doing? And what was kind of the thing that the event or the or the thing that you saw that convinced you that AI could fundamentally change how people practice hate these high-stakes conversations that you train that that Atlas Primer trains their users on?

Speaker 14:46

It's an interesting question. I would say it's probably been a series of observations over my career, um, rather than any one sort of Big Bang. Um But you know, just to answer your first question, I spent most of my early career, uh, well, first 15 years or so in media and advertising. So I was a research analyst, I was a media agency founder, um, had the experiences of starting a business, growing it, exiting, and then ultimately um, you know, went went to Google where I got to practice that trade, but with monster accounts, spending nine figures on our platform, big global automakers. And so it taught me a lot about sales, partnerships, BD, uh, how to take care of your customer and all that goes into it long before we were actually applying AI to our daily work. And uh more recently, I pivoted into ed tech, and that's where I saw a whole other, a whole other host of applications. Uh, because really, instead of selling clicks and ad impressions, I was selling economic mobility to government and and nonprofit and philanthropic partners who wanted to be part of creating creating change in American society. Uh so we would do big upskilling initiatives. We would try to basically take find people that had the intelligence and had the grit, um, but maybe didn't fit the profile of how opportunity works in this country and give them access to the kinds of resources for skill development to put them on the right path. So that certainly shaped my thinking about how technology can support people. Um, but whether you're reading pages in a book or on a screen, uh, you know, the the modality of learning is important, but uh there are many paths to the finish line. I think what's important above all is this mindset that we can use technology as an enabler. It does not have to be this race to the bottom where we are um really just using it as a crutch. Yes, there are important things about how technology can help keep us focused on high-value work, but at the end of the day, um we need to, we need to work with the the raw materials up here. Um, and so that's really how I now feel a lot of relevance to the work I'm doing with Atlas Primer.

Speaker 27:10

Yeah, um makes sense. So uh I guess we should probably explain what Atlas Primer does. Um, you know, we're we're on the plane, we're flying cross-country. I lean over and say, hey, what do you do? How do you explain it?

Speaker 17:22

Yeah, so we're helping organizations augment and invest in the the human skills that are gonna make them competitive at the next level. So you look at the status quo today. Uh, I'm gonna assume all your listeners uh are enjoying the same benefit, the same privilege of having the world's most powerful technology in their pocket. Uh a couple years ago, that that became a reality and we haven't looked back. AI has obviously been around a lot longer than ChatGPT's version one, right? But um, what happens is if everybody has access to the same AI powered by the same underlying data, and it's the same $20 monthly credit card payment that gets you access to, you know, a fairly robust version of it, then how are we competing? We're all basically a commodity unless we think about ourselves as competing on our ability to leverage that technology. And that takes original thought, it takes creativity. And in the world of business, it also takes people working with people. So I firmly believe we are not catapulting to some future where the bots do everything and we sit back with our universal basic income and just don't do anything. No, we're still gonna talk to people. Spent 25 years in sales and marketing. So in my world, that meant I go to another human being, build trust and rapport, get them to like me. No one buys from somebody they don't like, um, and and show exercise judgment, show wisdom and be a trusted advisor to that person. That's how you sell. You find out their problem, you figure out how to fit your product to the solution, rinse and repeat. And so I think that's the mindset we want to be thinking about. With Atlas Primer, we have a role play app that does this. It simulates conversations based on detailed personas. It gives you objectives, you execute, you connect your uh experience to coaching and feedback, and ultimately the business outcomes. So, can I onboard faster? Can I close a deal faster, close the deal bigger, whatever it may be? And then from there, we're helping executives read the data at scale to understand the skill gaps that are persistent across their organization, particularly in sales. So we work with sales. It's it's um, I would say it's a very obvious uh, you know, playing field for us. Yeah. But we work with leadership development. So imagine a new manager doesn't know the first thing about giving feedback, right? Bunch of 22-year-old Gen Z employees who don't know the first thing about taking feedback. It becomes very helpful. We've worked with people in cross-cultural situations where you have to show executive presence in a unfamiliar format, even recruiting, right? How many people don't actually know how to properly interview? Yeah. So there's a lot of applications. Anywhere that the spoken word is is meaningful, where we have a place.

Speaker 210:16

So then let's kind of steer it towards that sales environment. Um, one of the things I'm I I wouldn't consider myself a salesperson, but I am somebody who's got a strong business development background, which does require the ability to build rapport. Because, like you said, people buy from people they know, like, and trust. Um, there's, you know, most of the salespeople that I've encountered have a high degree of confidence in their ability, whether or not the numbers back that up, they have that. So how are sales professionals reacting when a company says, hey, so we're gonna have AI actually maybe not teach you how to sell better. I'm sure there's a more diplomatic approach that doesn't harm the ego of the sales professional. But how, how are the the I get that look, if I own the company, whatever mechanism we need to use to increase sales velocity, volume, uh value, all that kind of thing, I'm into it. But that idea and that intention meeting the the, I don't know, we'll talk about the we'll think about the cocky salesperson who's, you know, the best at their craft and that sort of thing. How do the sales teams react to the introduction of AI actually wanting to coach them and make them better?

Speaker 111:29

Well, um, there's a couple ways, right? So historically, sales teams are supported by enablement, or maybe it's the learning and development team that's part of HR. Uh, maybe it's an external trainer that comes in every so often. And generally speaking, I think we've reached a point in in corporate America at least where you don't want to be known as the person with the fixed mindset who says, Well, that training's not for me. So if we're bringing people into a training, let's say on a quarterly basis, hey, we got this ringer, he's gonna come and talk to us for four hours, you're gonna learn a ton. People aren't gonna really they know they're gonna learn something. Now, where it's changing is that we're seeing uh a lot of drop-off. There's a lot of decay in those skills when you just do that one session and then go back to your work. We need learning the flow of work. So five minutes here, 10 minutes there. What we're trying to do is show people that that is a smart investment of their time. Any salesperson is gonna say, hey, any uh time that I spend that's not in front of clients is compromising my ability to hit the number and drive business for you, the boss, right? But um, but these are these are investments. And if somebody believes that they are imperfect, which we'd like to think almost everybody can agree on some level, yeah, then it's a question of well, how good is the signal that we're giving you? So if you do one role play and you look at the readout at the end of that uh exercise, it knows you. It feels it the same way if you were asked AI to analyze your writing style, right? It kicks up on things and it makes very incisive observations. And that's part of a little bit of a you know, a trust game that we play. Okay, well, if it was useful to you, then work on these three things the next time and watch your score improve. And if you do it long enough, we've got evidence that you're gonna improve your overall performance. You wanna be earning more at the end of the quarter, right? Yeah, so that's kind of how we look at it. You're always gonna have some friction. Um, but I think people start to follow the pack when they realize actually this stuff works.

Speaker 214:21

It's working. Yeah.

Speaker 114:22

Yeah.

Speaker 214:23

So I I get it. And I've been in those environments, I've hired the ringer to come in on a quarterly basis. Sure. And, you know, uh recency uh and frequency are gonna be kind of like key uh contributors to somebody's growth and their skill. Once a quarter, maybe I take some notes, maybe I remember a few things, but most of it's gonna get lost. But with in this environment where I've got an AI enabled sales coach, I guess, even sales trainer, let's say, um, what's the frequency of exposure in like an ideal situation for Atlas Primer? What's the frequency of exposure for that sales professional and the tool? Like how often are they working together?

Speaker 115:05

Um, that's a great question. So on the high end, I can't tell you because I wish I could say that there are people that are putting in, you know, 15 hours a week. Maybe one day we'll see that and I can tell you. But on the low end, we've seen that as little as three sessions a week can get you a 30% bump on whatever the outcome may be. Yeah. So that that works for SDRs that are doing 200 cold calls a day. Uh, it works for on the other extreme, right? If you're in a call center and you're just, you know, your phone's ringing off the hook with disgruntled customers. Yeah. And everything in between, enterprise sales, partnerships, you name it. Um, so those sessions typically you're talking maybe 10 minutes um all in. When we structure our pilots, that's what we when we build the initial user cohort, that is always our guideline. We say find people that are prepared to do this 10 minutes times three times a week. And that's usually enough within about four weeks. We start to see the signal.

Speaker 216:04

It's pretty fast. So, what what does the experience look like? Because I've I've had some friends that have attempted to build kind of like um almost like a co-pilot that rides along on the call that it's on the side of the screen if you're on Zoom or whatever, and it's capturing both sides of the conversation. It's kind of like, hey, say this. Like, like almost if the sales manager was on the, you know, in the other ear saying, tell them this, you know, use this technique or whatever. Um, what's the experience like in the the training environment? Is it me looking at like a video avatar saying, hi, Chris, you can sell better? Or is it is it much more you it we we talked about role-playing and that sort of thing. I can load up my customer profile into that environment, maybe even give it some past recordings of calls that we've had so that the models can understand what a quote unquote typical sales call would look like. And then based on that, it can start to give my salesperson repetitions without them having to be on the phone and and burning up a uh expensive lead.

Speaker 117:00

I think that's pretty, pretty accurate. Um so let's start with a few things we are not. We don't do the video avatars. Um, what we've found, I mean, if you're gonna put a lot of your energy and resource into user experience, what we're trying to do is is make the actual uh the the sort of ingestion of intelligence the leading edge of the product experience rather than acute avatar, which never looks quite perfect, right? Yeah. It's sort of a distraction. Um so it's an audio first platform. Um the second thing, you talked about the coach that's there, like the little, you know, uh, there might be a little pop-up in your window saying, okay, client just said that. Now's your chance to say this. If you've ever used one of those, it seems cool in a practice environment, but if you've ever done it in a real live call, it is very easy to screw that up. Because if your eyes are darting around on a Zoom call, well, your customer's gonna see that. Yeah, and in a live environment, you don't get that. So now you're used to something that you don't get. Um, so what are we? What we have is the the flow goes something like this. You log in, and if you were responsible for the program, right? If you were the program manager tasked with getting this team skilled up, uh, you would have your objectives. It might be that they need help closing. It might be that we just need a better onboarding process. And that's usually going to come from your leadership. So you're gonna come armed with some objectives and some some you know backing from leadership. You're gonna come with a uh whole body of data and and knowledge and intelligence. Um, and so you everyone has it a repository of documents, your pitch decks, your spec sheets, your objection handling.

Speaker 218:48

Sure.

Speaker 118:49

As you create the scenario, there are there are a few ways to do it. One is you type a few sentences into the scenario studio, and it will make inferences on what you're trying to do. And it'll basically build out a scenario with a persona, uh a little bit of background on this person, some of their personality profile. So you can adapt to that. Um, it'll set some objectives based on the role you are. So it's gonna ask you about you too, what you sell, uh, what you're trying to get out of the meeting, to put together a package for what are we trying to do here? So you could uh do it through the scenario studio and just basically either uh click accept or revise what it gives you. The second thing would be a more of a detailed questionnaire, uh a little more time consuming. Uh in both of these cases, you can drag and drop all of that information into a secure environment. The other way to do it is through integrations. So, for example, we work with Gong. Gong obviously is best in the business for call recordings. And when that integration is enabled, the scenario creator flicks a button, and now all of that intelligence is present right there in the scenario creation process. So it makes it not only more realistic for the learner, it makes it a lot quicker for the creator. And we all know those teams are understaffed, under budgeted, trying to do more with less. So you're now building this scenario and you check it, you stress test it. We have a couple mechanisms in place to make sure that there aren't like goals that are colliding with one another. You'll set the amount of time uh for somebody that how much you're gonna give them, and then that's it. You're if you're the program manager, you should probably practice it yourself before you distribute it to your team. Uh, but once the team gets it, they get to execute the scenario. And there are a couple of options. You can enable visual feedback. So it's gonna check your eye contact, make sure you're not like reading a cheat sheet and looking off screen. Um with body language, right? So if the body language is a little bit too much, it's too distracting, you'll get that feedback. Um but otherwise, you know, it's pretty straightforward. Click here to practice, have at it. Once you're done, you're gonna get this readout that tells you how you scored against your objectives. It's gonna give you your highlights and low lights, which is basically here's where you're strong, here's where you need development. It's gonna put it in the language of a coach to say, you know, I recommend you practice this, do it a couple of times, do it in different formats, et cetera. Um, there's some analysis on tonality and that kind of stuff, confidence, uh, persuasiveness. And uh, you know, you then at the end you can share your results with your manager. He or she will see it anyway. Uh, but we also try to gamify it a little bit.

Speaker 221:33

Interesting. So it's not just like teaching me the mechanics of like, here's the the sales talking points that we want to do. It's actually looking at the the practitioner and giving them feedback on, you know, sit up straight, or yeah, you know, you're it was obvious you were reading off of a script as compared to. So that's pretty cool then, because like you said, I mean, if it if regardless of what it is, if a tool helps some helps my closers close better, I'm all for it. But if it's something that becomes the crutch and it's like, oops, you know, the tool is down, and I should expect a decrease in performance, that's not ideal. So this is something that actually uh the more they use it, the more they in a natural setting uh they're they're more capable at delivering the result.

Speaker 122:26

It's it's just the simple theory of of repetition, right? Nobody ever disputes that practice makes perfect. Why? Because as you take many swings of the bat, you start to get more comfortable and you develop that muscle memory. And so, you know, the little variances from take one to take 10, yeah, uh, all of that gets crystallized in here. And that's what's important because this is what's going into that real meeting, right? So, too often we we we force our people to go into the meeting cold. How often are we really saying you better practice, right? We're not making that space for them. And so that's part of the conversation we have with our customers is that this is a new opportunity as we're saving so much time by automating all this work. Now is the time to go into our habits.

Speaker 223:15

Okay. So for the listener, this is one of the things that I've kind of been like wrestling with like, how do we answer this? Great. We come into your company and all of a sudden there's more bandwidth available. But like, what do you fill the void with? And if it's you know more free time for the team and higher morale as a result, that's fantastic. But in this case, in the sales environment, you're filling that void with like because now they're not entering. Into the CRM because it's happening automatically because of AI is supporting it and blah, blah, blah. But they're actually filling that void with sharpening their skills. So I like that a lot. Okay. So here's a question I've got. Let's say that's great that I can say, hey, listen to the last 50 calls and get like an idea of what a sales call goes like for our organization. But what if it's, hey, we've got a specific client. Is it able to like, am I able to feed in enrichment and give it links to their LinkedIn profile and the AI will actually start to build a synthetic avatar of Mr. Jones, who's going to, who is actually going to be sitting on the other side of that Zoom screen in 15 minutes? And therefore, I can't anticipate this. A buddy of mine wrote a book called AI Driven Leader, Jeff Woods, and he talks about how they did that in preparation for board meetings. They built like a synthetic panel of the board based on true behaviors. Am I able to do that with a specific, if I know that who's going to be on the other call, am I able to build something that specific and train on like what it would most likely be like talking to that person in advance of the live call?

Speaker 124:45

Absolutely. Yeah. So during scenario creation, that's another thing you can do is import the LinkedIn profile. Um, that is right now, that is the only way that we connect to social profiles. Um, but it it's obviously the one that is gonna have the most utility, right? Uh yeah. The LinkedIn is gonna give you their professional history. It's gonna show a little bit about like what they value and their personal mission if they fill out that about section. Yeah. Um, and so you know, a practical application for that. If you're if you put in that person's LinkedIn, right, if they were uh PhD in data science and you start coming at them with fluffy storytelling and right brain kind of stuff, you're gonna lose them, right? So this is a perfect way for the AI to say no, you need to steer it a little more in this direction.

Speaker 225:33

So I like that a lot because look, uh I've spent a lot of money to generate leads. And the last thing you like and you know, in this business and previous businesses I've been involved in, you're just gonna write off some of that expense for the first X number of calls for that new salesperson. You're just like, that's just part of doing the cost of doing business, let's say. Absolutely. But this, like the cost for me to, you know, generate that synthetic activity, that synthetic interaction, as compared to $250 a call, a thousand dollars a call, whatever your lead costs are to get somebody on the phone, that's attractive. How is the math working out for these folks? I know that you you indicated that with a little bit of training, there's measurable impact pretty quickly. But as far as like training up the new sales team or the new salespeople, uh, do you guys have data or or numbers on like the the cost associated with you know those lost leads because somebody's got to get practice somewhere as compared to not doing it in Atlas Primer?

Speaker 126:31

We've had those conversations in silos, Chris. So sometimes you know, we will get, we will know from the marketing team the target CPA is this, and we've achieved it, which means it's a proxy for how valuable the lead is, right? So presumably somebody on the back end has done the math and said, we we earned this much on every new deal, and then they back into the close rate and you know the the whole all of the funnel analysis. Yeah. Um we haven't aggregated that data because frankly, like you would need an absolutely immense number of just a sample, right? In order to find like a real reliable narrative. Uh, but to be honest, you know, the the other way to look at it is so so we're not yet at the point where we can say, uh I I long for the day when we get there, when we can say, hey, based on your performance in this call, your likelihood of closing this deal went from a 42 to a 46%. So congrats, you're making progress, right? And then obviously you can back into some dollar value of that. I'm sure we'll get there one day. We're all going in that direction, right? But in the meantime, we're also looking at, well, what is the legacy uh approach? Every company is spending something on training, and how are they measuring it? They're basically saying, did people do it? Did they do it all the way through? And did they like it? Like that's as far as it typically goes with with LD in in these corporations. Connecting it to business outcomes, a lot of a lot of people are working on this, but it's it's um, there's a lot of sort of behavioral noise in between. And if you have consistent practice, that helps to turn that noise into a bit of a signal. Uh, so that's how it becomes easier for us to connect, for example, to CRM data, and we can export everything, put it into Tableau if that's easier. Uh, you know, any any sort of Power BI, what have you, uh, to be able to say, okay, so we put these people through this program, and then compared to the control, we're seeing that their deal sizes are five percent better. So the little bit that we spent on this is clearly getting a return, right? And then and then you have a conversation about well, what would it look like to scale it up to the whole team?

Speaker 228:47

So here's kind of a curveball a little bit, I guess. Has there been any uh benefits? I mean, because look, it's it's obvious. I think any of the listeners could say, great, if my people can get some more reps, it doesn't cost me, you know, the expense of the marketing and the time to get the people on the call and all that, and hopefully we don't blow it. That's easy, right? They're gonna be better prepared when they get on the phone call. But has there been any other benefits like I don't know, retention or um like people hitting president circle faster because of this thing or morale going up or like has there been anything like that that you guys have noticed that you didn't quite expect? But actually, wow, that's an as they call it, an unanticipated enhancement.

Speaker 129:30

Yeah, a lot of those are anecdotal. So we hesitate to you know speak, certainly we we we don't, you know, share our appliance data, we leave that up to them. Um, but we um we see anecdotally, for example, that you you can see the engagement of people by looking at a metric like training consistency. And when you see that they're recommended to do it three times a week, because that's how we structure the pilot, and then we get out of pilot mode and we're into like full-blown day-to-day usage and they're using it five times a week on average. Okay, well, that's that's a cool signal. That means they figured out a way to work it into what they're doing. Um, we do see we don't by default have visibility into these metrics. So we leave on our relationships to understand, hey, what's the CRM telling you about all this? Um, is it saying that you are closing more deals? Um, or is your deal cycle getting shorter? Um, or after after the signature, are your customer success teams more effective or quicker in getting the upsell and the cross sell? Uh how about a renewal, right? After 12 months, like are they more likely to renew? And anecdotally, we hear yes. Uh so for now that that suffices. Um, the other the other thing is 100% of our pilots have converted. And so what we found, you know, that that speaks loudly. Um we we look to embed deeply with these organizations when we kick off because we know that our support will help the odds of success. Yeah. We also just learn a ton. And that's that's part of the benefit is you know, if we can build a better product by shortening the cycles of just absorbing that from our customers, then all the better. Everyone wins.

Speaker 231:20

Okay, got another question. Um, because different, you know, different sales managers and and companies they have their own kind of like styles, like in Atlas Primer, are let's say spend selling or Sandler method or anything like that, are they already pre-programmed to follow that? Or can I say, hey, this is the, you know, we're using this particular sales, I don't know, theme. Sure. And it can accommodate that.

Speaker 131:43

Yeah. You can get it out of the box. Like even if you're typing, hey, uh I am a junior rep that needs to start closing bigger deals, and I want to prepare for a uh you know, fictitious discovery meeting. And our organization uses MedPIC, and that's the only way. So the first thing it'll say is, well, who's the meeting with? And it's going to use medpic language. So, you know, are going to be the champion. Is this going to be the economic buyer? Is this somebody that's, you know, maybe an influencer or an enabler in the process? And that may manifest a little differently if it's Challenger method or if it's Sandler. Right. Um, so yes, you can bring that into the process the same way you would drag and drop collateral or you know, anything else, any other documentation. The other thing is, and this is kind of interesting, um, we work a lot with coaches, uh, coaches that have come to understand this is a complement to their business, not a, you know, it's it's it's not putting them out of a job. It's actually augmenting the coaching business because it does two things. It gives them an easy way to assign homework. So, hey, I'm here on Monday, we're doing our regular weekly session. Uh, by the time we meet next Monday, I want you to do 10 sessions with this bot. And uh, you know, we're gonna work on some of the skills that we talked about. Or the other thing is uh we've even had a few that after their engagement is over, uh this becomes a leave behind. So, hey, hope your team enjoyed it, right? You were only budgeted to have me for let's call it a month. Great. Uh, but for whatever price, call it 500 bucks a month, you can have a version of me that is there 24-7. You can all use me at the same time, right? Because I'm just an AI. So um we work with those methodologies a lot because as you know, whenever a new CRO comes in, they're usually bringing some methodology with them, and they're gonna have some designs on you know, baking that into the whole operating system of their organization.

Speaker 233:48

Yeah, it makes sense. Okay, so we're gonna shift gears a little bit. So just for the listener, like, I get it, I hope you get it. Like, this thing is intended to uh tune up your sales team without you having to, for the expense uh of you bringing in the outside trainer, not saying you don't do that, but bringing in the outside trainer, or hey, we yeah, well, we got these, we got these leads, we've already paid for them. Let's let the people train on our leads. Like it sucks. Um, but what I want to do is I want to move into uh a section of the show that I call the handoff, right? And it's got nothing really to do with your your business or the industry that you're in, but it does have to do with how you as uh you know an executive in this organization are using AI. So what I'm gonna ask you is like what's one task that you personally have stopped doing because AI just now does it better on the first pass than you would have.

Speaker 134:46

Yeah. Um any salesperson knows this uh conundrum. When a deal had momentum and it starts to stall, and this happens to me on Fridays for some reason. I must be in a I start thinking, what about that meeting from three weeks ago? Yeah. Why haven't I heard anything yet? There was so much energy and momentum. And you look back in the CRM and you say, okay, I followed up with them twice, and there's still nothing. And the hardest thing in the world for any salesperson is why what is my reason for reaching back out? I know my reason is to bug them and remind them that they need to do this because I need to hit my number, right? Yeah, yeah. But but I need a reason that is acceptable to them. And so salespeople will start like looking through, you know, press releases. Had the company announce anything I can congratulate them, or maybe there's some like industry story I can spin. And what we've done is we've connected um, so we use open claw to basically bring a bunch of different data into our Slack instance, which becomes the front end for a very simple chatbot experience where I'll go and and you know, HubSpot's our CRM and I'll go and I'll say, Hey, look, this deal with this individual, uh, it's stalled. I already tried this and that. Uh I'm I'm you know getting a little frustrated that I can't seem to move the needle. Write me an email that will properly and respectfully interrupt the pattern. And I specifically use that language with it this morning where I said, interrupt the pattern because the guy's not responding. And it'll come up. I mean, I would say 95 times out of a hundred, I use the first draft because it's that nice and it's what was the example this morning? Well, you know, it's an example with a uh a company that uh it's a Canadian company that we're we're working with to try and and actually uh you know get a pretty large piece of business. So it's kind of a channel sale, it's a little bit outside of our typical comfort zone in terms of how we go to market. Um, and the company is in the news a lot. And so I was thinking, hey, just pull something from the news that is germane to our dialogue. Yeah. Yeah. And uh I saw that she read it. So maybe the reply's coming. Uh nice, but you you know, you you you look for you look for progress however it comes.

Speaker 237:12

Yeah. So for the listener, I think that's a fantastic idea. If you're, you know, that that outreach thing, it's not like, hey, still haven't uh, you know, uh paid the invoice yet or whatever, to come up with that angle where it's like, hey, hey, buddy, I'm just reaching out. Oh, and by the way, and obviously they know what, you know, oh, there's that thing. And maybe they got busy, maybe the kid got sick, maybe the whatever they got a busy travel month like I do. And it was like, oof, yeah, and that's a good one. Um, so you you mentioned real quick, you mentioned one of the things that I know that um executives uh would use AI more, but they're a little unsure of the risk or like what happens if I put this into the models or those sorts of things. And so I like to also touch on um kind of uh soothing them a little bit when it comes to uh AI professionals and their perspective on how they're answering this question. But let's say they're using, let's say they're using Atlas Primer. Um, all that information that's being captured, how do they know that um the the the client information or whatever isn't uh is being used safely? Like what are you guys doing to manage that? And how do you suggest other executives, like what questions should they be asking when it comes to using, I guess, any tool? Like, how do I know that my data's safe or that I could put financials in here or whatever?

Speaker 138:33

So there's that's a great question. And I think everybody needs their own checklist. Uh if you go beneath the very surface, there's always gonna be nuance to your business, your industry, right? If you're in healthcare, it's gotta be HIPAA, right? Like there's gonna be some of that. But if we were looking for something scalable for this conversation, I would say first thing you ask is, are you using my data to train third-party AI models? And it's gotta be a hard no. Um, I mean, I think even for us as consumers, we should be unchecking that in whatever, right? Encryption, right? So what you want to hear is that the the type of encryption, whatever it is, they'll let you know. Hopefully, they know what it is. Um, but what's important is that the encryption is uh applied both in transit and at rest. So uh there's different vulnerabilities in either of those scenarios. In the case of our product, since we have, since we are actually recording from the microphone and from the camera, um, we're very quick to remind people that our recordings are discarded immediately after you. So the second you click end scenario and it starts generating the feedback, that recording is gone. Uh the same is true with transcripts, although we do give somebody the option to override that because it might be useful. So this is an important principle, right? There's saving all the data you generate, there's saving none of it. And the middle ground is save what you need and throw out the rest. So we don't make assumptions about what is needed. Okay. We will leave that up to the user. Um, those are some of the important things. I mean, of course, you know, if you're doing like a if you have to go through like a proper security review, you're gonna be looking for things like the SOC2 certification and you're gonna want somebody accountable. So let's get the CTO on the phone or or whomever it might be, just to be there and do a little bit of like AMA style, like give me your your hardest questions and I'll answer them. Um, I think that's that is the bare minimum, I would say, nowadays.

Speaker 240:39

Yeah. So my big takeaway from that was that nugget of like as a listener, if you don't have some uh, I don't know, decision matrix about what do we keep versus I mean, some of you may want to keep all context and create the company brain context layer that has everything. That's great. But not all data is the same. So some of the data that you're capturing on those calls, like if you don't have that uh green light, yellow light, red light environment for what do we keep, uh, I would suggest that that be something you explore introducing into your AI use policy. And if you don't have an AI use policy, shame on you. Let's get that together.

Speaker 141:15

Oh, we shouldn't be using AI.

Speaker 241:18

And I'll tell you, a lot of companies don't. So if you're listening to this, you don't have one, listen, don't don't like cancel your account or anything. You're a part of the majority, sadly, but uh it's an easy fix. Um so, Paul, like can people get a demo of this? Or how do they conceptually I think people get it, but until you are sitting in front of the robot, you don't necessarily know what the experience is going to be like. So is that an option?

Speaker 141:43

Yeah, absolutely. Um, so first thing is you could uh just go to roleplay.atlasprimer.com and there's a little button right there in the nav to create a scenario and you can get started. Uh you could reach out to me personally over LinkedIn, and uh I can even give you a promo code so that you know if you want to see some of the advanced features uh you know behind the paywall, that's always an option. And you know, we like a challenge. So if somebody says, hey, look, I've got uh this is this is my company, this is my product, this is my team, and here's what we're not doing well, we'll come back and we'll say, okay, here try this scenario, and and it'll be something sort of tailor-made. Um, so you know, we we accept that challenge.

Speaker 242:24

Very cool. So I'm gonna make sure that we have uh the links to all of that information in the show notes. So please check that out. But uh, if you're at all curious, because I am, because this sounds like just uh such a like low friction environment for improving the performance of your salespeople. It's not a full day where they're not on the phone. It's 10 minutes. They can, I guess they can train at home if they want to, right? I mean, okay, so like there's let's that's pretty low friction for improving your sales team. So uh Paul, thank you for that. I I don't see uh this is the first time we've had a conversation with any um of our guests about this type of tool. As soon as you say it, it's like, duh. Why didn't I think of that? It's like pretty obvious that that AI is very well suited to support this. So um I'm gonna go to roleplay.atlasprimer.com and check it out myself, actually. So I guess any any closing remarks or anything, any uh final bit of advice for anybody who's out there that's that's because we all want our sales teams to improve. None of us are like, oh, they couldn't do any better.

Speaker 143:25

You know, right, right. I think you know, it's this is probably not the kind of advice that that people are hearing for the first time. But you know, 25 years ago, email started becoming a big thing, and and what did we tell ourselves? We said, Oh, we're finally gonna leave work at five o'clock, it's eight grade, it's gonna save us so much time. And all we did was cram more in. And we're at risk of doing that again, right? So if I'm if I'm pushing off, you know, whatever percent of my work day full of busy work to an AI, then we all need to be asking ourselves, okay, so what are we gonna do with that time? And it's very tempting to just put twice as much on every rep. And that's just gonna burn out your people. That is not a human forward strategy in the age of AI. So what we're saying is the the way to really keep your human capital, your biggest investment, right? It's the biggest thing on your payroll or on your on your budget is your payroll. Like in order to continue to get a return on that investment, if we're gonna really look at it as like hard numbers, keep your people energized, keep them motivated. And that means, you know, give them new ways to just improve themselves and and move forward. Uh, so you know, that's the the ones who we see thriving on our platform are the ones who think that way.

Speaker 244:49

I dig it. That that that is good advice for sure. Hey, well, Paul, well, thank you for taking the time out of a Friday afternoon um to share with our audience, even though you guys are listening to it at least on a Monday. Um, and I again, very cool what you're doing there. I totally get it. Um, and I'm excited to go and role play with it. So, everybody, thank you so much for joining us for another enlightening discussion about how AI is impacting uh all of our uh all of our work environments. So um, and just as a favor, look, if you like this episode or you got something out of any of our episodes, and you know some other people out there who are trying to figure out like how do we use AI more, how do we use it safely, all those sorts of things, um, we're doing a good job here on the podcast uh exploring how others are doing it. So please, we'd love it if you'd share it. Um if you really like the show, leave us a review or whatever the case might be, good or bad. I love the feedback, although we haven't gotten any bad feedback yet. So that's a good sign. Um but with that look, I look forward to seeing everybody on the next episode of Using AI at Work. And thanks again for being a listener. Thanks for tuning in to Using AI at Work. Don't forget to subscribe for more conversations about how to use AI at work. And a special thank you to our sponsor, Chief AI Officer, for empowering businesses with AI education and training. Visit their website for free AI readiness assessment and AI strategy guide to help you get started using AI at work. That's www.chiefaiofficer.com. Follow us on Twitter at the handle usingAI at work. And visit www.usingai at work.com for free resources to help you harness AI in your role.