Automation That Feels Human

Coby: Hey, Dallas, you want to grab a drink?

Dallas: Coby, let’s go grab a drink.

Coby: All right, let’s do it.

Nicole: Hi. I’m producer Nicole Denson, and today we’re talking with Dallas Olsen, director of revenue operations, a big leap. Dallas has been solving operational problems long before DevOps had a name. Building automated systems in industries like mortgage fitness and now digital marketing to help teams work smarter, faster, and with less friction. Known for connecting dots that others miss Dallas blends logic and creativity to build infrastructures that actually work in the real world. Back to the show.

Coby: So we’re going to Swig right?

Dallas: Okay, good. Let’s do Swig. Nothing beats soda. No you know I’m from Utah.

Coby: Do you have a favorite Swig?

Dallas: Swig drink? Got a couple. But I love that beach babe.

Coby: Ooh, I’ll one of those.

Dallas: Yeah, I think this time I’m going to get the founder. It’s the Diet Coke with, like, peach and cream.

Coby: Oh, no, no, no, I’m going to do that because Diet Coke is the base. Yeah, that’s the way to go.

Dallas: I think I might do that too. The beach babe is kind of a summer one, you know?

Coby: Yeah, I get it. So what’s new at Big Leap for you?

Dallas: Oh, man. There’s a lot of new stuff going on. A Big Leap. I would say probably the most exciting thing that’s going on right now is just how much innovation is taking place. Seems like ever since I has come on the scene, it’s been nothing but trying to keep up with the tide and make sure that we have all the skills and knowledge we need.

Dallas: So it’s been fun to see how many different ways we can incorporate AI into our workflows and create new efficiencies. So that’s always fun to me.

Coby: Lots more words. Tons more words on paper. Now that AI is taking notes and making all that happen. You notice that? There is like three times as many documents in the last six months. If you weren’t, if you weren’t doing what you’re doing now, what would you be doing?

Dallas: That’s a good question. You know, surprisingly and well, maybe not surprisingly, I actually love what I do. And I think the reason behind it is because I just love. I love logic and numbers and puzzles. So I think if I wasn’t doing what I’m doing right now, I think something maybe like architecture would be fun to look into just cause I love, like, math and how it combines the logic and the data and the physics and all that stuff with just creativity.

Dallas: And try making something beautiful.

Coby: Right? You know, you could always just tell people you’re an architect like Art Findlay.

Dallas: I kind of am! Like a database architect.

Coby: Your data. Yeah. Experience architect. Yeah. Cool. Very cool. So, I got a question. This is what I want to know. Because you’re on the front lines, right? What are you seeing that nobody else is seeing right now?

Dallas: Oh, man. I would say the most repetitive and recurrent theme that I see that most people probably don’t know, that there’s a very easy, easy solution to is going to be just front end lead management. There’s so much opportunity, to really cut out so much manual effort and so much busywork that teams could be using to do other, more valuable work.

Coby: Yeah. Okay. Why aren’t they?

Dallas: You know, I think that there’s just a barrier to entry. I think that, you know, people specialize in the business that they work in. They’re not they’re not tech experts. They’re not. They don’t have the time to go on YouTube and start looking at the ins and outs of how automation softwares work and workflows and all of that stuff.

Dallas: And so I think that people are aware that the option exists. I think it’s just a matter of understanding how to do it and set it up in a way that’s accurate and in a way that’s, reliable. I think that’s another part of it, is that people maybe are hesitant that, you know, because AI is involved or, automation is involved, and maybe it’s not as accurate as a human did it?

Coby: Is it is it? I think it’s kind of funny, actually, that, where AI and automations can have the biggest impact. First, I’ve seen this in a number of places that I’ve worked and some clients that I’ve worked with where the, the place that AI can have the biggest impact is, is where people still have the most people engagement, like in sales. Right. Like you think. Okay, well, yeah. Spreadsheets and all that kind of stuff. Whatever. But no, it’s it’s in the sales process, which is traditionally that’s where the, you know, the, the handshake in, in the smiling and the glad hand and all that stuff goes on. Right? Absolutely. But they’re, they’re typically I mean, I want to, you know, I don’t want to paint with too broad a brush, but they’re typically a demographic that doesn’t really like tech as much or at least adopting their they’re a little later adopter when it comes to technology.

Dallas: Yeah, I can see that. It does seem like that there. I will say, it seems like once people can see how the process actually works and how it’s done, there is some trust building that’s done through that. Just because before people understand how that all works, it really is kind of just like what’s happening behind the scenes and how do we trust it, you know? And so I think that’s another fun, exciting thing that I get to do at Big Leap is that we’re not just creating systems and handing them over to people. We’re creating systems that people can take ownership over and they can own themselves later on. Right. So we’re creating transparency and clarity into people’s business operations and making them more efficient at the same time.

Coby: How do you automate everything without making it feel like somebody is in an automated thing on the other end? Like, I mean, everybody gets the errand or has an experience where they got the errand text or it’s a double text. I mean, those are those are mistakes, clearly. And so eliminate mistakes is one of the ways I just answered my question. But like, what if you’ve got something that’s smooth and like smoothly humming along, but it feels like, oh, I’m in a funnel. I don’t want to be in a funnel. Yeah, no, no control for that.

Dallas: That’s totally valid. And I think that that’s a concern for most people is as soon as AI and automation start being introduced, people worry that the messaging and the delivery is going to come across as from a robot, right? It’s not personable. It seems like you’re sort of just a number in somebody’s CRM. And so I will say kind of circling back to what you said earlier about, you know, you got your sales teams that are busy meeting with clients, shaking hands, smiling, building relationships and connections with people. Right. Yeah. That to me is where the sales team has the most value is in the human connection. Right? What we’re trying to do is free up the sales team so they have more time to lean into those valuable opportunities and automate where we’re able to, you know, and so there are some cases where automation maybe isn’t the best solution if we’re trying to have a more human approach. There are still ways of doing it. And AI has come a long way. But I would say to avoid having people feel like that, they’re just in a funnel and they’re having robots reach out to them. We’re specific and strategic about how we go about it.

Coby: So give me an example because this like, yeah, I think this is big. Give me an example of where you would use AI for this part, and then maybe or an automation for this part, but then a totally manual process process for this other part, like can you give me detail.

Dallas: Yeah. Why don’t we. I think probably what would resonate the best is just the front end process. I think most companies deal with a form or a lead intake process. Most companies are dealing with responding to those, right, trying to reach out to their contacts that have submitted interest as quickly as possible, right, to get that good conversion. You’ve got your sales teams who are manually taking notes after each phone call, putting them into the system. All of these things are things that bog down our sales teams from reaching out to high priority leads and contacts because they’re busy doing manual work, right? And so one of my favorite areas of application for this is right there in the front end. So instead of having somebody manually receive a form submission, just start at the beginning. Somebody fills out their form. The form comes in to the receptionist or whomever is taking the leads. They’re going to be manually reviewing that lead to determine if it’s a valid lead, if it’s spam, right. They’re probably going to do some sort of some sort of scoring on that lead. Yeah. Just see how valid it is. How urgent is it. Do we need to address this right now? Does it have a high high opportunity to convert? And then they have to assign those leads and send them out to their sales teams. So if you take just that process, that’s probably going to take a receptionist. Let’s be generous 7 to 10 minutes. Right. Yeah. If they’re doing it right. No mistakes.

Coby: And if they’ve got a bunch. They get tired, they want to do other things. So now it’s 7 to 10 days.

Dallas: And right there you’re already running into issues with your speed to lead, right? Because by the time that lead gets handed off to the sales team, a competitor could have reached out to that contact. You know, they’re probably actively looking and filling out other form submissions so we can automate that process so that we have a super speedy, response to our leads. We are automatically scoring those leads based on the criteria that we predetermine based on the buyer’s behaviors.

Coby: That’s a sexy truck? Oh my lord. Yeah. No. Keep going. That matte black. I’ve never seen one like that.

Dallas: Yeah. Okay. So all of these front end processes, your lead intake, lead scoring, lead assignment, reaching out to your contact, that full process, let’s say if we’re being efficient from lead receipt to the sales team, reaching out after it’s been assigned is going to take 30 minutes. Maybe if we’re quick we can have all of that finished in a matter of a minute. Wow. We can have that data in the form submission, add it into the CRM. We can have an auto responder go out. We can have it assigned, we can score it and we can do all of that instantaneously.

Coby: Well then what she going to do with her day? Hey, how you doing, brother? So I get the whole, efficiency, time saving argument? Yeah. You know, and you heard me go off a couple of times, probably on behavioral hierarchies and all that kind of stuff, and how time saving and avoidance of work and ease and all that kind of stuff is valuable. But it’s, removing friction from the sales process is among the behavioral hierarchies, the most ubiquitous, but also the lowest driver. So is there a place where an AI enabled outreach or nurture automation delivers a better experience to the prospect or the customer? And how would you do that? So let’s get out of the: “hey it saves you time” and let’s get into “hey, it’s a better experience for your prospects.” Can that happen?

Dallas: Sure. Let me think about this one. Is there? So there are so many different applications and uses. So let me think.

Coby: Oh, that’s really good.

Dallas: And maybe this isn’t exactly what you’re what you’re thinking of, but I will say, so one of the things that we have automated that seems to create a much better experience for a client that isn’t necessarily client facing is our call notes. So we have-

Coby: Okay.

Dallas: One of the most times taking processes that are what’s the word I’m looking for? One of the processes that takes the longest and most tedious is going to be for the sales team after a call to want to sit down and, you know, add all of the notes that they need on the client, you know, who are they, what are their pain points or their goals?

Coby: It never gets done anyway.

Dallas: It never gets done, right? And that right there alone causes such a negative client experience because you’ve built this relationship, gotten to know the client, and they have been vulnerable and giving you all this information, and then it gets handed off to fulfillment. And fulfillment is over here going, well, who are they? And so now we’re duplicating questions, duplicating effort, frustrating our client. Now where AI gets becomes involved and makes a much better seamless client experience- and I’m going to kind of talk about the process behind this- is we’ve actually set up automated call note taking, which informs our CRM and updates everything that we need by the time we get to that handoff. So what that looks like is that we have a call recorder join our calls, we get a call transcript from that. And then one of my favorite things is that we can use automation to essentially process that in whatever way that we want. So we take the phone call the call transcript and say what information is important for our fulfillment teams to have or our sales teams to have after this call. Let’s identify those points and let’s extract all of that information that we want from the phone call. Okay. So we’re getting all the data that’s available from the call transcript. We can put together all of the notes, the client’s pain points, the client brief handoff to fulfillment. But then on top of that, we can also layer in other research that didn’t come from the client. We can use AI to go and do competitive research that’s saving us time, right, that we don’t need to do. We can have AI go through and create strategy recommendations. We can have it do analysis on data on all kinds of stuff.

Coby: You can set all this up automatically?

Dallas: Yup. So in the instance of a call, if the call notes are only triggered, there is a new call. New call transcript available from the sales team. Cool. Now let’s take that. Let’s put it through our AI. Let’s put it through GPT or Claude. And let’s extract the data points that we need to so that by the time that client closes, we can put together a comprehensive brief of everything that our teams need to know. So at the end of the day, it really is kind of to your point, it’s about removing the unhealthy friction from the process that creates a better client experience.

Coby: Okay, okay, I like that. Where does it get used that it shouldn’t be used?

Dallas: Ooh, where does it get used that it shouldn’t be used? Well, I think to your point earlier, you do have to be a bit selective. AI can be trained and can do almost anything that you want it to do, with the caveat that you have to train it up. You have to get it to where it needs to be. Right? AI is very much a garbage in, garbage out system. So if you’re having AI write your emails and you’re not, you know, getting the AI, what type of tone you want to use or what context it needs to have, it’s going to sound very salesy.

Coby: I got a great question.

Dallas: Very robotic.

Coby: I want to know. Okay, so, I’ve been working with training models for three years, maybe three and a half years now as part of a, engagement as a service startup. Before what I’m what we’re doing now, and. The whole goal was to write meta prompts for AI that would truly make it sound different than it does. How do you prompt have you found success prompting AI so it truly sounds like different than because at this point it takes about a sentence and a half and I’m like, I know exactly where that came from, and I know exactly how that like. And it’s great for internal stuff, it’s great for it. But as a customer, you know, there is something about somebody putting fingers to a keyboard and a little bit of thought. So it’s not like you’re trying to fool anybody. But it does help if the AI that you’re using can output in a way that truly sounds like you. Yeah. So how have you found to make that happen?

Dallas: That’s a great question. So one good use case example. And the way that we make that happen would be your automated outreach. Right? So somebody fills out a form submission and you want your sales team to reach out to that lead that prospect as quick as possible. If that sales rep is not able to reach out within 15 minutes, we want to make sure that something goes out. Right, so we can do an automated email. The way that we can get those emails to be automated and not sound robotic is just a lot of training. Right? So let’s say I want these emails to come from me to go to my client. ChatGPT is going to do the best, and AI is going to do the best when it has context and when it has examples. So by that, I mean, if you’ve got 25 emails that you can upload into ChatGPT as reference material, it’s going to understand your tone a lot better, and it’s going to replicate how you speak to people, where you put your commas, what types of emojis you use, if you use emojis.

Coby: Right or don’t use them-

Dallas: Or don’t use them.

Coby: because you’re over 18 years old.

Dallas: So really it is a matter of giving it to some good context, giving it some good examples, and essentially a foundation and a framework to go off of. It kind of reminds me of like, did you ever do that project in school where you have to write out all the instructions to make a sandwich, and then somebody else has to do literally what those instructions say and how messed up the sandwich comes out.

Coby: Yeah, right. Yeah.

Dallas: Like you have to be so specific about, okay, you got to open your peanut butter, pull the film off the peanut butter. Grab your knife in your right hand, put it in the peanut- you know what I mean? AI is kind of the same way. You have to be very specific. And there is a method and a recipe for getting good outputs and.

Coby: Yeah, it’s not a set it and forget it. It’s as much work as just writing the email. In fact, a lot of ways I think it’s a lot more work than just writing the email. Yeah. But then you get to invest that work in a system that then you no longer have to answer.

Dallas: Yeah, yeah. Another really cool thing about I do I don’t know if we’ll have time to talk about, but is its application when it comes to data. So and we’re looking at, you know, potential markets or potential clients, we might reach out to people who are maybe close loss that we haven’t reached out to a while for a while because of maybe budget wasn’t right, maybe pricing wasn’t right. Right? Instead of coming up with a tailored campaign for every single niche client that we want to reach out to you, right? Because we’ll have people from different industries with different company sizes. That’s where we can also flex AI and say, hey, this is the general content that we want to put out to these people, but we need this to be specific and scalable, right? We don’t somebody doesn’t have time to go through and write 50 different styled emails to tailor to each individual segment or branch, right? So we can definitely use that to use the data to have more tailored, customized personal approaches, which is always going to be a lot more successful.

Coby: I like that. So my question then is there’s lots of folks that are like either threatened by this or. For various reasons. I mean, it’s still technology and it’s still I think a lot of folks still feel like maybe it just gets in the way of the personal experience or whatever, what somebody who’s still new to this, maybe they’ve been under a rock for a year. It’s still new to it. What’s the first step they can take to, leveraging this? In their personal process, assuming it’s like a sale or some sort of salesman.

Dallas: Sure. Well, the first thing I would say is that anything that you’re looking to automate and things that you can automate, let’s start there. Anything you can automate is going to be anything that you do repetitiously or on a schedule, right? There’s not a lot of like if you’re sending invoice reminders to somebody every day on the 30th to remind them to pay their invoices, that should be automated.So that should not be taking up capacity in somebody’s brain. So I would say step one is to take a step back and look and see what things are you doing that a computer realistically could do. You know, are we sending out follow up emails, reminders on a regular cadence? Is somebody, you know, repetitiously, going through and collecting lead source data, adding it into a spreadsheet and then adding it over here, anything that’s repetitious or done on a schedule, you can automate, right? So that’s the first step, is we want to look and identify a trigger. Yeah. Anything that’s got a trigger we can work with. From there, we’re just looking at what is our what’s our business flow. What needs to happen after this. Okay. And so we start looking at outlining our process. And then we start looking into the tools that can help achieve those like those things.So a couple of the tools that I love are HubSpot Hubspot’s got some incredible integrations and automation capabilities. If people don’t have HubSpot, I would say Zapier is probably the next best thing. Zapier is a if this, then that type of software, okay. Which means that if a new call recording is available, then feed it to ChatGPT, have ChatGPT extract the data we want, and then plug it into HubSpot, plug it into Salesforce, plug it into high level. Okay, right. Send that as an email to my sales team as a recap.

Coby: Ooh, last question. Yeah. HubSpot or High level. Ooh.

Dallas: Well, I’m going to say HubSpot just because HubSpot’s my baby. Yeah.

Coby: You love it, don’t you?

Dallas: I’m recently getting acquainted with high level, and high level does have some cool, some pretty cool and similar functionalities to HubSpot. But I’m always going to say HubSpot just because I think it’s a lot more user-friendly. It’s being more actively worked on and developed. To my knowledge.

Coby: It’s kind of becoming the platform to go to, isn’t it?

Dallas: They are constantly adding new features. New. I mean, they really do listen to their audience. So anything that people are asking for is probably under development, which I really appreciate.

Coby: I think that’s the key. Yeah. This was cool. Thank you. All right man, back to work. Take it easy.

Dallas: Thanks, Coby.

Coby: All right, man.

Dallas: All righty.

Coby: Back to work.

Dallas: Okay

Coby: Take it easy.

Nicole: Thanks for coming along for the ride! If you enjoy Sweet Takes be sure to subscribe and leave us a review. Join us next time. There’s more sweetness ahead.