#10: How to get the human and AI split right, with Greg Symons

(upbeat music) - Hi everyone, and welcome back to the Revenue Architects podcast. We are joined today by Greg Simmons, founder, president, and COO at Assembl. For those that aren't familiar, Assembl is a workforce orchestration company that connects the staffing and the recruiting of people, AI, and software as one singular workforce. Welcome Greg, how are you? - I'm doing great, thanks for having me. You found time to do a podcast on a Monday morning while still being president, founder, and COO of your own company. Kudos to that, what's your secret to time management? (laughing) - And what, three days before quarter end, too, just to put a. - There you go. - Yeah, if I had a secret to time management, I would probably talk more openly about that, but got a great team around me that helps kind of keep the trainings running on time. - Oh, that's great. - In our last episode of this podcast, we covered the AI stack and the people who are running it. So this week, we're sort of rolling up our sleeves and getting into an actual sales play that Greg and his team rely on to drive the run pipeline, running end to end by a revenue team that's part human and part AI agent. So we want, take away today for listeners to be some actionable steps, best practices, and kind of a rundown of what Greg's doing, and how they could potentially do something like this on your own. So Greg, when I hear go-to-market leaders being asked about what AI does to head count, they answer with predictions most of the time. So you've kind of already lived and breathed this experiment. Is that right? - Yeah, very much so. We've been at it for a little while now, starting to figure out really what works and how we're able to apply. I would say just more generally the technology to our revenue function to drive outcomes and the technology changes quite a bit. So fortunately, I have at least a few years under our belt and getting to where we're at today. - Yeah, and you're running roughly what, 120 employees alongside 200 digital workers, AI agents? - Yeah, so our company, we've been, the sample's been around for about three and a half years, coming up four years now, and we've been very fortunate to have some pretty significant growth really early on as the company was getting off off the ground. And yeah, a lot of our business was, and still is very human-centric, right? But also as a function of our ability to keep up with the growth coinciding with advancements in AI and ILLMs, we did have also an opportunity to really think more specifically and intentionally about how we can grow and expand the scale of the organization and what I would say are more non-traditional ways if we were talking about this five years ago, where historically, I think a lot of companies wore their headcount as a badge of honor, right? In terms of how many employees they have and employee growth. And that's not really talked about as much anymore. And for us, I wouldn't necessarily say that we wear our total headcount of both humans and AI or digital workers as a badge of honor. For us, it still is ultimately a derivative of how we're able to meet the demands of our customers. And at the end of the day, still deliver meaningful outcomes that it'll offer us to be an important partner to those companies that we serve. - Well, it's been exciting to follow your growth. And I think, you know, as I said, you're living in breathing, you know, your own business model, running at sales play across both, you know, the humans in the loop as well as the AI agents. And I think in my experience, that leads to some of the most successful, you know, plays not only from, you know, a data governance and a QA standpoint, but from an efficiency standpoint. I would love for you to take us step by step through, you know, a tried and true sales play that's running through both, you know, your human workforce and your AI workforce. - Yeah, definitely. And I can walk you through that. And I think it's also important that you commented on this and we're something we're really proud of of. You know, we treat ourselves as our own customer zero. And I think that's so important because of, again, as I mentioned, there's been so many changes in technology. There's so much noise in the market today. And it's very easy to get distracted by trying to plug in a tool or technology to get something done. And at the end of the day, our view on this is that what we've been able to be successful with, say a year ago, might not be the same cocktail of tooling, if you will, that is going to create the same type of success for us today. So for us, our journey on our initial entry point into more of a hybrid revenue play really started about two and a half years ago when we were very deeply integrated in the Salesforce ecosystem. So we use Salesforce internally where a Salesforce consulting partner were also the Salesforce ISC partner. So a lot of the customers that we work with are, as you would expect, in around the Salesforce ecosystem. And because of that partnership and alignment, about two and a half years ago, we started to experiment with some of the tooling that Salesforce was coming out with. At the time, it was Einstein co-pilot and then it later got rebranded as agent force. And years ago, at Dreamforce, the announcement of their SDR agent became GA. And even proceeding at becoming GA, we had been working in some beta programs within Salesforce to really start to wrap our heads around one, the technology and then two, really the purpose of how we should think about applying this to our business. And it matters at a really good time in our company's growth because of the growth that we had experienced, we had just hired about 100 people in 100 days. We've talked a lot about a lot about that growth. Our workforce changed materially, right? 100 people for us, we've got quadrupled the size of the business in about a quarter. And up, it's a lot of stress on the business, especially operationally. The other side of the business was we continued to generate significant top of final demand that was too much for a single person or really multiple people to navigate. And so it really worked out very favorably for us where we were able to now have an opportunity to grow and meet the demand, not with just humans. And so we started to experiment with the SDR agent framework and we learned a lot from that. And ultimately we went live with our first SDR agent, we call him Teddy. That was one of the learnings that we had, for example, is the first big understand that we had was that we named our SDR agent Teddy because our ability to create success with Teddy was a function of applying the technology in the same way that we think about hiring a human. And this is not the same mechanism of deploying AI into a revenue function or any function for that matter, is very different than a software deployment, right? You're not capturing business requirements and building them and testing them, deploying them, and then you're done. You really have to think a little bit more about the outcomes that you're trying to achieve, the work that needs to be delivered as well as the motivation to get there. And once you have done that, you don't just stop, right? It should be in the same sense of the vitality we hired a sales intern and never talk to them again. You'd be like, well, you're probably not a very good manager, right? And that's because you have to invest time, you have to continue to make sure the messaging is right. And so our very first revenue play was really to have Teddy start to engage with our inbound leads and do it from a place grounded in knowledge and understanding of our company and our differentiators and the personas that he's targeting. And ultimately, the goal for Teddy was to schedule meetings with a human BDR so that we could then take that hand off and really change the work of our human BDR to be less hitting the phones and sending out emails and more having set meetings every single day throughout the week to have more meaningful conversations. And so that really became kind of the first part of our hired workforce and our initial top of phone revenue play. - Great, and so yeah, that's a really helpful context behind the motivation for creating Teddy. What's the trigger that activates Teddy's ability to either qualifying inbound lead or schedule that meeting with the human BDR? What's step one for giving Teddy an account that he can do something with? - Yeah, so the very first part is whenever we're ingesting top of funnel information, there's first the understanding of the validity of the information, if you will, right? There's plenty of times and plenty of channels that create noise or spam or solicitation, et cetera. So the first is to categorize that this is an actual lead that we are able to engage with and is not just some generic solicitation type email. So there's a segmentation process that goes and moves that data into a queue and will automatically position it in a way for it to be action against. We then also are leveraging other tools within Salesforce that we've used to quantify the grade and the grade on the score of the lead. And as those two variables start to meet certain thresholds, it then creates the trigger for that lead to go into Teddy's queue. And once that lead is in his queue, we then take it a step further and have Teddy start to do some additional research and understanding of the company as it relates to our ICP, as well as the persona, to make sure that we can match the messaging with the person, again, grounded in our data to have a more specific first touchpoint so that it's not just some generic email catering that kids. that goes out to the CEO of a 20 million dollar company or the CTO of a half a billion dollar company. Those are very different messages. Got it. So step one segmentation and step two, you know, obviously, like, let me actually bring this back a little bit higher level. You're making sure Teddy is educated with your ICP and the correct personas and, you know, actionable buying groups at the company that is qualified so that once Teddy finds the company, you know, it knows exactly who to try to pull from that company for our reach. Is that right? Yeah, that's exactly right. And if you were to take, take the technology function out of it, it's no different than what we would do if we were on board in a human SDR, right? The human still needs to be trained and who assemble is, what we do, why we do it, how we do it, who we sell our products and our services to the differentiators that we have in the market, and also how to position those products and services to a buyer depending on their role in the organization. Teddy's no different. And that was a big unlock for us. And I know that, you know, some of what we talk about today is about reproducing these plays and helping others achieve success. For us, the big unlock, as I mentioned earlier, it was it was treating this less as a software problem and more of a human or a people opts type problem, if you will. And that was a big unlock in terms of how we went to market and thinking about creating these hybrid teams to go ultimately still deliver the same outcomes, but what if it were just purely human? What else are you using to inform your like grading and scoring? So obviously you have first party data that lives in CRM and Salesforce. Do you use third party, you know, intense signals, data coming from other sources as well to kind of qualify the lead or at least, you know, maybe help with the outreach, personalizing an email campaign or, you know, a cold call campaign. We do. Yeah. So Teddy V1 was really just using our own first party data that we had and we accumulated based on the data and information that we would get. A little over a year ago, we started working with Zoom Info and Zoom Info started to create a much water aperture in terms of the data and information that we could use to create more specific and intentional outreach. They also expanded the role that Teddy plays within our organization and also introduced some other digital workers that we can certainly talk about. But Zoom Info has been a really good additive for us in the sense that he gives us, gives us now the third party data and the insight to understand not necessarily the information that we're getting from the lead itself, but then reconcile that against other signals that we might be finding from other individuals within that company, per se. So what are other people with the new organization talking about, how can we then start to make some inferences as to the type of messaging that should be shared with the person who came to us looking for some type of information? Got it. Where would you say, so, you know, step one, segmentation identification, step two, research and persona matching, is step three, the handoff from Teddy to a BDR to create email campaign or or a cold call campaign or just Teddy sort of help with that process. Is that is that collaborative between human and AI or is that a clean handoff? It it's very collaborative and I wouldn't necessarily that it's a defined next step either. Teddy on his own has been trained to follow up and establish rapport with these prospects over a defined period of time with a specific number of outreaches, a specific number of responses that he can respond to over a defined period of time. And what that does is it allows for Teddy to execute his job in the same way that we would again train a human to do so. But the biggest difference here is that when we talk about Teddy, Teddy has his own email account with us. He has his own Salesforce user record. Teddy also lives in Slack as well. So when we think about the the 360 degree view, the emails and the back and forth in the commentary is all still very visible to the human BDR who eventually then takes over that lead. And that accomplishes kind of two forks, right? One is the lead goes dormant or we get some type of response from that prospect saying, Hey, I was just looking for information. I'm doing some research. We're not actively looking to buy in right now. But we've created a lot of data now, first party data so that we can continue to nurture that lead after Teddy has finished his job. And that nurturing could be done in the function of an email campaign. It can be a function of a follow-on cadence with our human BDR or another follow-on cadence with another with other SDR tooling that we have. The happy path here, of course, is that we meet the buyer or the prospect where they are. And ultimately are able to engage with them at the right message at the right time. And then Teddy is able to take our human BDR's calendar and schedule on behalf of the human. But as that handoff occurs, all of the conversations, all of the information are still fully accessible and visible both inside of Salesforce and in Slack so that when we do have that first handoff meeting, we're not starting from scratch. Our human BDR's name is Jordan is able to reference conversations that we have between the prospect and Teddy and just have a more meaningful first human interaction with that prospect. That's great. And how many, so let's think about, you know, companies big and small having to apply something similar as you've done for a sales play. How do you scale the use of Teddy across, you know, your 100 plus employees, maybe, you know, less than that are BDR's? But, you know, just Teddy have like weekly roundups with people. I imagine you're not doing everything on a one-to-one basis where, you know, they're dedicating one day, one call to one prospect at one company. But does Teddy do like, you know, weekly standups with summarized, you know, reports of, you know, multiple accounts? How does that work? Yeah, that's a really good question. So, um, Teddy actually reports to Jordan. And this is really important. And it's pretty fun. And this was another learning that we, that we came to is whether it's Teddy or any of our other digital workers, they have a manager that they report into similar to an org chart. So, on a weekly basis, Jordan is reviewing Teddy's work in the same way that a sales leader would review the, the output of the KPIs for, for the account executives in the organization. So, on a weekly basis, we have their observability dashboards that let us know how many emails have been sent, where at each process in the cadence, these are how many meetings have been scheduled, of those meetings, how many have been converted into opportunities. And then the other thing that we do in terms of reviewing Teddy's work is we take an export of all of the emails that have been sent. And we run that through a separate AI tool to, to validate that the message is in what the message should actually be. And it's a, a trust but verify mechanism is yes, we've trained Teddy, right? But these LLMs are not, they're not going to produce the exact same message the exact same time. If you're looking for that, we don't need AI for that. Just go use a marketing automation tool and you got your email templates and off you go. So, this is really interesting because it allows for us to take an audited approach and still make minor tweaks or refinements to how Teddy has been trained to make sure that the messaging that goes out on a week over week basis continues to improve. And that can be done at the individual email level, but it also be reconciled against the persona level as well. And as it relates to the scale, this is one thing that's really great, especially when you consider the role of an SDR or BDR in any sales organization, these are largely early employees of their career. A lot of them may be coming fresh out of college or may have only been in the workforce for one or two years. And ultimately, at least for our company, a lot of times the BDR serves as a foundation for them to understand and learn the business, become from all the phones and then graduate into a full account executive role. But within that role, your BDR function used to be rebuilt. Well, Teddy's not going to get promoted. He's not. He's very happy in his job right now as an SDR Haitian. And so, when we think about scale, it gives us the ability. First of all, Teddy has the capacity to touch about six times more prospects than a human would as a function of not having to take time off or have working hours or moving up in his career. So we're able to create scale on the top of funnel that is consistent. And if we make changes to his training, we don't have to then retrain the next person. Teddy just continues to get better over time. And you go from having maybe the quality of an intern on day one to a one year BDR after like month three and then come month six of of training Teddy at that point over about 26 weeks. You have a very, very well trained BDR who deeply understands the business and has been trained consistently to continue to provide better messaging. So the scale problem really, really gets solved in that regard. And allows us to focus our time and frankly, how we invest in our revenue function in more strategic ways. Hello, Teddy. Teddy seems to be growing fast enough to maybe expect an invite to president's club or something like that. Just be careful. Do you see, you know, six months from now, a year from now, I know that that's a crazy timeline to ask someone to predict Aitren since things are moving so fast. But do you see the teddies of the world still reporting to the Jordans of the world? Yeah, I do. And I think ultimately where this goes is, and you're already starting to see this of, you know, these, there's been articles. This is not my own unique thought of, you know, the $250,000 BDA. right? And this is not in associated BDR in a traditional sense. It's somebody who is really more of a go-to-market engineer. And ultimately I think where where I see things going is you're going to have more capable technologies and tools that is going to change how companies operate. And for us the role of the BDR is still a very valuable role for us as an organization because it's still deeply ingrained the humanistic traits that are needed to build relationships, understand and communicate value, and respond to and a number of scenarios that might exist within the sales process. But I do think that in the world of where at least where I see us going is is also taking a bit of a pendulum shift if you will. And what I mean by that is Teddy is is a completely you know was designed initially is a fully AI baked out resource and tool. But we've also seen the market that there's a cost to that right there's a a real cost of consumption of the tokens and the spend that goes into building out AI capabilities within workforce. And so for us we've actually started to see the pendulum shift a little bit back towards center. And we are now able to use all the first party data coupled with the third party data that we have and apply just you know what I would say is old school compute against it right it's it's essentially free for us to go calculate data consistently every single time. And so I think for us in the next six to 12 months then becomes a function of how do we leverage compute with AI and rethink how AI tooling is ultimately leveraged to drive a better outcome whether that outcome is delivered by a digital worker or a human. It changes the makeup of how we present the information ultimately to the actor is going to be executing on it. So I had vision that Teddy is going to still be very successful in his role. And we're already starting to build out additional digital workers within the revenue function to change the way that our human employees are able to come to work each day and invest their time on really working on what is is going to be the highest priority item or the highest value items within any given day the week. Love that. And you touched on some of the like the foundation that these plays are are being driven by and we've talked a lot on this podcast about garbage in garbage out you know Teddy is only as smart as the signals and the data that you're feeding him with so like you could you speak to the importance of having a clean you know underlying Salesforce foundation to this to this play. Yeah I'd say that's probably not probably I think that's much more important than the technology itself and that's not a function of of AI that's a function of anybody trying to do their job right a human or Teddy are going to have just the same amount of egg on their face if they engage with the prospect and that prospect is being rounded in the wrong data right nobody wants to be in that scenario. So an ongoing exercise that we have is to make sure that we have the right structured data and structured workflows internally to present the information in a consistent and accurate way and then also introduce tooling as as you mentioned to to hydrate and validate that data and again zoom info is a great tool for us to constantly be doing the data enrichment the data validation cleanliness of data the duplication or de duplication of data and so for us that's that's so important because it's not just Teddy's job that's impacted it's the information the conversations that we have with that prospect for the very first time that contains meaningful information that then gets pulled into and through our opportunity funnel address conversations about the outcomes that the customer was coming to us from the very beginning we can validate and show that customer how we can help them deliver those outcomes during the sales process once the software has been implemented the customers live we can then reiterate and validate that we have been able to deliver those outcomes that creates a really meaningful customer success pathway for us that creates customer stories and all of that as a function of how the data is structured and moved through each stage in the process and if there's one thing that lalms are very good at it's understanding trends and signals with large amounts of data and finding finding things that humans just otherwise would be able to do or it would take them a long time to do it. So the the underlying data in the flow of data is another area that we apply just general lalms not necessarily digital workers but analysis and insight into helping us refine our messaging and our positioning to understand what are the common outcomes that our customers come to us looking to achieve where the common challenges within these different buckets what are the types of buyers what are the size of companies etc and that just gives us so much more information that is first party hydrated with third party that we carry that thread all the way throughout and ultimately it creates much more meaningful conversations and what I believe has been shown through our company's growth much better experience for our customers because we're able to really just get to the truth of what it is that we're trying to solve for them and consistently check and validate throughout that we're still on the same page and we're not just trying to sell our product we're truly trying to help sell and reconcile against the conversations that these customers are trying to solve. So you mentioned growth and I'm curious to dig in around what you know key metrics KPIs you're looking for when you run this play over and over again what is the like you know what's where do you look to determine you know a benchmark of success here do I've seen a huge difference in meetings booked your sales cycle length your pipeline numbers can you get into the growth a little bit and the impacts of this play and what they've what they've had on the company yeah that's a good question and to be completely honest when when Teddy came to life we didn't really know what to expect just like honestly how could we right this is the best way so how we've looked at growth has has changed a little bit based on our overall maturity of our our hybrid revenue function so the initial growth metrics that we looked at were what we knew we could control with Teddy which is his outbound emails how many emails is Teddy sending on a period over period basis that's that's something that we can see and we can control and record on and that should ultimately then produce a specific number of meetings just based on some some expectations that we'd have of human interaction right if our human is sending this number of emails how many how many actual meetings would we expect to get and so our initial baseline for that was to achieve the same output as a human would and we were able to see that actually pretty quickly so then it became a function of scaling that up how do we create now and refine that messaging as a function of reviewing and auditing the messaging via the emails that I talked about with Jordan to start to see if we can increase the number of outbound emails to meeting conversions the next area that we get into then is of these meetings that are being booked what is the percentage of these that get actually converted and of those meetings that get converted then you go into the the ultimately then time to fill quality of leads etc so for us the the back half of that is still much more of a human function and for us it gets back to the jobs that we have Teddy do Teddy perform which is the the quality of outbound and the biggest measure of success for Teddy is the quality now of meetings booked we've already established that Teddy is more than capable of scheduling meetings and now the real goal in terms of his coaching is how do we get better conversion on the meetings that he is scheduling and that's a function of the ICP and the persona and the signals right because there's plenty of ties within a sales cycle that people are just interested and frankly kind of curious talking to a digital worker and just wanting to start to have a conversation and kick the tires if you will and so we started the train Teddy in a way that allows for us to prioritize and move people into the right tracks that they need to be in and we've already started to see a pretty good uptake in the in the conversion quality of these opportunities that Teddy's influence which has been really exciting I love how your first you know your first metric here was not just like driving revenue or dry or even driving meeting conversion it was can the AI agent replicate and improve on the efficiency of the human releasing a sales campaign so I kind of like how you did a you know waterfall or a step-by-step you know take on on you know how it's actually improved the business and that's led to a greater meeting conversion and meeting quality so yeah thanks for walking through that in you know so like you're talking about you know replicating the efficiency of a human if you did run this play with a hundred percent human support and Teddy was not in the picture at all what would you what do you think the outcome would be I would probably answer that by saying I would expect that the outcome would still be somewhat similar but our cost of acquisition would be significantly higher right and that's purely a function of Teddy's capacity and output and it's also the time horizon that we did this right so if we were to answer that question say over the period of a year right imagine that we're now hiring five additional humans we are training and ramping and onboarding those humans which let's say it's a three three month onboarding for them to be up to the same level of Teddy or maybe even the six month onboarding for them to really be speaking at the same level that Teddy speaks out and then let's assume that within that that 12 month horizon two of those SCRs get promoted into an account executive function right it's a really interesting question because it's it's not confined to just this nice box of hey if we had five people do this and Teddy what would the outcome be yeah the outcome initially of Teddy's output could be the same but the impact to the business would be significant not in a good way it would really hurt our ability to scale operationally, and you would never really be able to create the momentum that you need to create, because you're constantly in the sale of hiring, training, promoting, hiring, training, promoting, and that's not just the BDR function. That's the management function that goes into it. You have more variables of touch points where things can go wrong. There's a lot of impacts there, so would the output be the same? I can be pretty close, but the impact would be way different. Got it. Got it. So it's more of a resource isn't banned with question. So same thing for 100% Asian support. What if there were no humans in the loop, and it was Teddy running this play all alone? What would happen? Probably we would not see the same output. And I think that's not a function of Teddy. I think that's a function of the industry that ran in the buyers that we engage with. I think there are some markets that are really right to have much more of an autonomous AI revenue model, ours is not one of them. Companies come to us because they are looking at how they can effectively better operate their staff and their recruiting organization. These are oftentimes very big financial decisions that companies are making with a lot of nuance and a lot of, I don't want to say emotions, not the right word, but people are very passionate about their brand and the services they provide. And they need a technology that can help them represent their differentiation. And those require a lot of trust building and knowledge and understanding about the pain points that exist within the customers that we serve. And there has to be a level of familiarity and trust and understanding to really help our customers ultimately achieve the value that they're looking for. So our business is not set up in a way that we'd be successful if we completely automated the entire top of funnel cycle without humans. And that's our business. I'm not saying there's not others out there that are right for it, but ours is certainly not one of them. Now I think that's been my experience when working with clients as well. I still see hesitation and I think that's well founded in completely replacing a BDR team with BDR agents. I still think that the hybrid model is the best place to go and you can hedge some of that uncertainty and maintain a good level of QA and data governance. I think that's just such a big, big part of the equation. So when your team is building out this sales play or others doesn't have to be Teddy, but what's something you tried and failed at? Well, what didn't work? And how did you solve for it? Yeah, I mean, like I said, the very, when we first brought Teddy to life, we started thinking about it as a purely as a software play, right? And ultimately, we didn't consider that Teddy was part of a much bigger process, right? And it really impacted his overall throughput and his outputs because it was really just looking in a box of, here's the data, when it's true, go do your job. That's a software mindset, right? So what we really started to do thereafter was we started to look at more holistically as a part of the revenue function to help us think and get to that point that I mentioned earlier about enriching and segmenting and having more specific training and understanding. We didn't necessarily change Teddy. We changed the way that we were managing and nurturing Teddy. The other example that I'll give is not Teddy specific. It's another digital worker that we have called Stanley. And Stanley helps provide overarching support and understanding internally for us to better manage the efficacy of our sales team. And this was more an example of that pendulum shift that I mentioned earlier where we previously built Stanley inside of Claude. He was a 100% built inside of Claude. And he was trained on looking at the data that existed in Salesforce, presenting the insights and the information, the analysis from emails and Slack messages and call recordings and providing coaching and guidance on how we're tracking towards the quarter recommended coaching tips, etc. The job was correct. The output was correct. The problem though was that Stanley would take anywhere from four hours to two days to do his job because it was purely built on LLN tooling. And that four hours to two days wasn't a function of it just taking that much time. There was actual human oversight that needed to help Stanley get unconfused on things. And the output would be mostly consistent but sometimes different. The numbers that it would produce would be mostly consistent but sometimes different. And that was not good. We can't have mostly right sales data, right? So that is somewhere that we tried and failed. And it got us back to this mindset of the pendulum shift of saying, all right, Stanley's just looking at this same data set. Let's just apply compute to the data and structure it in a consistent way each time. And produce that in a JSON output that allows for other digital workers to consume that in a consistent way each time. And then let's produce the output in a structured UI so that we don't have any variability on the interface and how it's represented to our external users. So the job Stanley's job didn't change but the four hours to two days went down to five minutes. And that was a huge, huge change for us. It's a big one. And I think that's a trend in the market. We're starting to see a lot more is that just because AI and LOMs are out there, it doesn't mean that you have to use them. You still have to really, really understand the ultimately the outcome that you're driving towards. And then what is the right tool that you need to use to go and help deliver those outcomes and it might not always be AI. And the human made, you know, the human in the loop helped Stanley be more efficient. So it's full circle from the standpoint of AI helping us be more efficient. But only if we are the one to educate it and feed it the right data to do so. That's right. Well, it was also the distraction too. And our, you know, back to the Teddy example. And we see this not only in our own company, but in a lot of companies as we engage with companies to help them actually architect their hybrid workforce. And a lot of the themes that we hear, you know, one, people don't really know where to start. And then two, maybe they started to roll out cloud or open AI to their business. And now all of a sudden, all their employees are spending their time building agents instead of doing their jobs, which is a real thing. And so for us, it's about making sure that the AI complements the role of the human, as you mentioned in a way that doesn't distract them from getting enamored, frankly, by the addictive qualities that exist when you tell a digital worker to go do something. And they're very powerful, but they can also be a big distraction. Got it. Got it. So as we tie everything together here, what would you say, you know, what would you say Jordan's real benefit is at the end of the day from working with Teddy, you know, how is the AI able to, you know, benefit the existing, you know, humans on your team too? I would you summarize that in one or two bullets. And that's a great question. Jordan is trained in the same way that the Teddy is trained, but Jordan now has so many more touch points to help round out his perspective of conversations that are happening within our company within the market and a more first player point of view. We share success stories, we share challenges, we collaborate a lot within our revenue process. And so I've slack to help unlock better insight. But if you think about Jordan now in his role, he's able to further craft his skills and his point of doing perspective based on his ability to coach Teddy on what to get outcome. So he's hearing and seeing what all these additional touch points are and seeing and consuming a lot of additional information within the market from people that he's now engaging with. And he's able to have much more impactful conversations to compliment again, taking then that process. Teddy stops at the point of the meeting. Jordan takes it further. So his ability to have a wider aperture of understanding what a good conversation looks like on the front end and the value that creates on the back end makes them much more effective in the conversations that he has day in and day out. Got it. Got it. So Jordan's being set up for success with all the right information, you know, how to personalize, the right person, the right company, the qualification, takes it from there and runs with the play. Yeah, he's able to see not only what good looks like, but maybe what bad looks like. Yeah. Right. That's an important part of any training process in any role. Right. We're always training to the good outcome, but you learn a lot more from the bad outcomes than you do the good ones. And so this gives him a lot more perspective on on touch points that require a little bit of handholding to help him out. Got it. Well, as we tie things together here, we want to make sure in every episode we want listeners to walk away with some actionable insights that they can use themselves to get building. So in your mind, what does a team need to put in place before their first agent similar to Teddy or Stanley is even worth building? What would you recommend? Yeah, I would say you have to first and foremost understand the the outcome or outcomes you're looking to achieve. You have to deeply understand what those outcomes are and the impact that it's going to have. Once you understand the outcome, you need to then start to dissect that into the work that needs to be done to go and deliver that outcome. That's really then where you start to get to a decision point of is it AI? Is it a digital worker? Is it compute? Is it insight that is provided? first understanding the outcome that you're trying to solve. Maybe you want to increase your top of file, maybe you want to increase your conversion rate, maybe you want to increase your meeting schedule, maybe you, whatever the outcome is that you're looking to achieve, you have to put a line in the sand and say, "Hey, this is the outcome." And if we can go and then execute on this, we can look back in 30 days and know that this was a successful endeavor. And then you can start to build on that. And I think that that's the biggest thing is, is on the building on it portion, is don't think about this like a software project. You have to invest the time into getting the technology to understand and learn and think and behave in the same ways that you would want your human workforce to think and understand and behave. You look accountable and coach it and nurture it, but you gotta have that outcome first. That makes sense. All right, final question, Greg, across everything we've covered, what do go to market leaders most often get wrong? Or misinterpret about putting agents into a revenue team? - I think one of the things that I hear most frequently and have seen and lives myself is the expectation that you're gonna go and deploy this worker and it's gonna be a silver bullet, you're gonna be drinking peanut colladas on a beach in six months, right? This is a incredibly powerful technology, but it requires the right mindset motivation to ultimately make sure that you are building a efficient revenue engine that has a lot of different levers that you're calling to go derive the outcomes that you're trying to achieve, but it is not a silver bullet, it is not going to solve all your problems and it does require the specific and intentional effort to go and be successful. - Got it. And the humans are very valuable in the loop as far as Teddy is concerned. So Greg, thanks again so much for your time to summarize it up. Greg and his team are running a hybrid sales play that not only helps to segment and validate opportunities, but really make the handoff from agent to human with first party, third party signals combined, set them up for success with a scaled, personalized, outreach campaign and helps measure and refine and perform some AB testing as well to improve meeting quality and meeting conversion. So it was really great to hear from you, Greg, and I'll let you get back to running your company. - All right, thanks, John, I appreciate the time. - All right, everybody. Well, I'm John Lloyd. This was Revenue Architects. Keep an eye out for the next episode in two weeks. See you then.