Innovative Revenue Leader - Agents in B2B Compilation
===
Speaker 2: [00:00:00] Welcome to the Innovative Revenue Leader Podcast. I'm your host, Seth Mars. Join me as we deliver practical insights to help B2B CROs Find new and innovative ways to grow in this fast changing environment. The Innovative Revenue Leader is sponsored by Sandler, a triad company, empowering sales professionals and leaders to master the craft of selling at all levels.
sahil-aggarwal_1_03-05-2026_131023: Yeah, I think we need to differentiate between an agent and AI powered automation,
seth-marrs_9_03-05-2026_161024: Okay. Interesting.
sahil-aggarwal_1_03-05-2026_131023: What I mean by that is, I hate to be the one to say it, but building an LLM workflow in Zapier is not an agent. yeah. I think we've conflated the two words, [00:01:00] which. Whenever you're building a node in Zapier that makes an LLM call, suddenly we call it an agent, and now we have hundreds of agents running for us and doing these smaller tasks.
That's not what I call an agent. I call an agent, I call Waymo an agent. if you listen to Waymo CEO, they will say, we are building the world's safest driver like that. That's an agent. I think ChatGPT is an agent. Claude, is an agent, like an agent is that can do extremely complex tasks and can run through multiple variations or loops, uh, and get the job done.
Uh, so I think in that vein I would call one agent. But if, but if you're calling Zapier automations as agents, then I think we will call V or Waymo a super agent.
seth-marrs_9_03-05-2026_161024: Got it. But I mean, really that, the, the bar for an agent is, I'm, I'm doing something that a person would normally do that I can [00:02:00] delegate to a, to a to V an AI agent that they'll do it. It isn't, I'm doing, it's so, it's like I'm doing a task that a seller would normally do. Like you talked about, like I'm building a board deck.
You would typically off offload that to like a rev ops leader who would then go build all that stuff out
sahil-aggarwal_1_03-05-2026_131023: Yeah.
seth-marrs_9_03-05-2026_161024: And then build the deck for them. But what you're saying is I could just have Von do that. So instead of a person doing it, a person's gonna be reviewing it rather than writing it.
sahil-aggarwal_1_03-05-2026_131023: Yep. And maybe another way to say that is, um, an agent should be able to spin up these subagents to do the task. Um, so these sub an agent might spin up a hundred a, uh, subagents or a hundred thousand subagents. Uh, eventually what you and I care about is the task should get done. I don't care what happens in the background.
And, uh, that is what a super agent, like a Waymo, uh, or a chat g pt or a Claude or a one should do, which is, uh, spin up its [00:03:00] own subagent to get the job done.
seth-marrs_9_03-05-2026_161024: So like one of the things that, that always seems to break all this stuff is so Vs. Goes through all of your data and kind of builds all this stuff out. How. Does it handle the crappy data? So like that's always the limiter that you'll hear over and over and over is. Yeah, great. AI's great. I can do a lot of cool stuff, but my problem is, is my data isn't good enough.
So I shouldn't even focus on an agent because I haven't done the hard work to be able to figure out how to get my data right. So an agent could actually function properly. Like how do you think about that in terms of like when you deploy this in a company, deploy bond in a company?
sahil-aggarwal_1_03-05-2026_131023: I tell people it does not matter anymore. Yeah. So, so where is crappy data in your system? So let's start with the first node, uh, which is the actual data, which is buyer communication, your emails, your slack messages, your, your [00:04:00] call recordings, maybe even seller note, like that's, that's your actual data. Then crappy data exists when we ask humans to convert that into structured information in Salesforce.
So your your first level, your first node is not crappy data. Like it is just data. Your second node is crappy, uh, in Salesforce or in your CRM, and then you are trying to do some reporting or insights on the scrappy data. But what I tell people is that any true advanced AI system will not connect to the second node.
It'll connect to the first node. So you bypass the whole process. You basically go from first node all the way to insights, and you don't have a crappy data, uh, issue anymore. Um, so that's where, that's where tell people it does not matter. What matters is are you capturing your core data or not. Now, how do you capture your core data?
Emails are automatically logged, [00:05:00] so you don't have to worry about that. Um, slack messages are automatically logged. You don't have to worry about that. What is not automatically logged are the calls. So there are still a lot of companies that are not recording their calls or not transcribing their calls, uh, to them.
I, I say like, you have already lost last couple of years of data.
seth-marrs_9_03-05-2026_161024: Yep.
sahil-aggarwal_1_03-05-2026_131023: to start now, and if you can't, then the only thing that you do is force your salespeople or your CSMs to take notes of every conversation. Uh, so at least we have that. Uh, but I think as long as you have that, the whole messy data concept is does not apply anymore.
seth-marrs_9_03-05-2026_161024: That makes sense. So yeah, as long as you let, you'll let the AI do the metadata, which is the update in fields that you ask a salesperson to do. And by doing that, you get consistent updates versus every seller kind of either doesn't do it at all or does it in their own odd way. So then everything cleans up over time on its own.
sahil-aggarwal_1_03-05-2026_131023: Yeah, but I would say yes, you should have AI update your Salesforce, but [00:06:00] what will happen is that nobody will look at the data in Salesforce. Because no sales manager or sales leader or a CRO wants to go into every single opportunity and read MedPAC. They just want to understand, tell me deals that are late stage without a decision maker or without an economic buyer.
seth-marrs_9_03-05-2026_161024: Yeah. so what I tell them is that if you ask ai, AI will again go into your root data and it'll get you the answer. Uh, now, yes, you need this data in Salesforce for reporting. No doubt about that. You need stages, amount, close date, uh, but. AI will take care of all of this.
In doing that, you're hearing agents, agents, agents. And when, when we were talking, when we were talking about, about that, like we had the, you had this analogy where you're like, yeah, agents are like the new dashboard where, You know, back in the day, everyone wanted a cool dashboard to make all their, their information visible and now
sahil-aggarwal_1_03-05-2026_131023: Yeah.
seth-marrs_9_03-05-2026_161024: Their own version of an agent
sahil-aggarwal_1_03-05-2026_131023: Yeah.
seth-marrs_9_03-05-2026_161024: And, and the result is likely going to [00:07:00] be the same, where I'm gonna have a bunch of crappy dashboards that nobody uses. Over time. If you keep just building agents and agents and agents, you're just gonna have a whole bunch of agents that nobody uses.
sahil-aggarwal_1_03-05-2026_131023: Yeah.
seth-marrs_9_03-05-2026_161024: To me contextually around, like, I think you used that as kind of like, I don't wanna do this.
How are you avoiding that with what you're doing with V?
sahil-aggarwal_1_03-05-2026_131023: Because V is a super agent V will spin up its own agents. The analogy that I use is, how many times have we both spoken to rev ops people where they will tell us that? I hate when their CRO will bing them in Slack, uh, or teams and say, Hey, can you gimme this piece of information? And the ops guy says, I have created 20 dashboards for you, and I have sent them to you 8,000 times and you're still asking me, uh, for the same dashboard again.
Why does that happen? Because for the analogy is for Rev, uh, for the CRO or for a revenue leader, uh, their rev ops person is that super [00:08:00] agent.
seth-marrs_9_03-05-2026_161024: Yeah,
sahil-aggarwal_1_03-05-2026_131023: don't have to, they don't have to remember all the, the nitty gries or the URLs or the business processes. They just go to this rev ops person and they can ask, make a request, and the rev ops person decides, okay, where I, where I'm gonna go for this piece of information.
And I think what has happened is that because we have overused the word agents and we are calling AI powered automation agents, we have built hundreds of agents. And vendors are selling agents. And don't get me wrong, I was selling, uh, update your Salesforce agent and a risk agent. And
seth-marrs_9_03-05-2026_161024: Yeah.
sahil-aggarwal_1_03-05-2026_131023: Saw that with rattle, where we will, where we will build 30 of these agents and each of these agents are doing their own thing.
Uh, but then something breaks and the person who has, who built the agent has moved on. And now agents are the new dashboard.
seth-marrs_9_03-05-2026_161024: Yeah.
sahil-aggarwal_1_03-05-2026_131023: like you suddenly have 10,000 dashboards and you don't know how you got to this [00:09:00] point. That's why the concept of super agent is your super agent is just your person who understands what is happening in the background.
So I can just do my job without worrying about the technical complexity.
joseph-miller_1_03-25-2026_160407: I like to, I like to give like a, a framing that I think will help people like sort of reconcile what's going on, uh, with, with that space. Um, I'll try to be brief about it, but if, if you sort of imagine like a, a, like a big circle that, that represents all value that can be had. Generated in the world, and then inside of that circle is like another smaller circle that was like the value that humans could deliver. Themselves. Um, and now, and then now there's like this new circle that is like the value that agents can deliver and they overlap with the human circle, like kinda like a Venn diagram and, and all the things that are getting the tension is that shaded space in between the things that agents are starting to do that overlap with humans that are sort [00:10:00] of taking the jobs away and et cetera.
And like, oh, I'm worried about, You know, I, I'm not gonna have a job and all this sort of thing. But if you just step back for a second, there's a really interesting dynamic that's happening. One, there's a whole bunch of stuff that only agents can do now. there's a whole bunch of value that they can generate that we couldn't generate before because they're able to work, You know, around the clock because they're able to scan large, uh, like large spaces of data and et cetera. And, and then there's like the part that is infringing upon our space. And yeah, that's an, that, that's, there's contention, there's real issue there. But then there is this also interesting space that is just human. And I think that what's going on is that you'll realize the space that is could be offloaded to the agents, will get offloaded to the agents.
You just can't compete on the labor dynamics there. But the things that are remaining for humans actually become like more and more important. So like the premier that of, of a human touch, if you will. And then there's the question of like. As an agent starts to, to capture this value, um, this additional value that was never had before.
There's like the overall [00:11:00] circle is growing very quickly. So there's just more value that can be had and captured in the world. And, um, I think that those are like the sort of dynamics that are at play that help people sort of understand what's going on.
seth-marrs_18_03-25-2026_160407: I think this is a really good way to phrase it because if you, if you think about like great situations where you're doing cool things that you want to do more of, you can almost never get all of that done. All the conversations that I wish I had more time to do XI wish I had more time to do y. This kind of enables that to actually, yeah, well we are not doing that anymore, so spend more time on x, spend more time on, on y.
It's kind of almost, it, it, it can almost be perceived as an age of abundance in that you're doing more of the truly great work and less of the work that you're kind of muddling along. 'cause you have to do it.
joseph-miller_1_03-25-2026_160407: Yep, that's exactly right. I mean, there's a lot of like, there's a dark side to all of that, but we don't have to get into that here. But
seth-marrs_18_03-25-2026_160407: No.
joseph-miller_1_03-25-2026_160407: A, there is, You know, there's a flip side of like. Human, like there's more, there's this age of [00:12:00] abundance, but humans as a percentage are actually controlling a smaller and smaller portion of it. Um, and then of course there's only so much human touch value that can be derived, period.
So You obviously need less humans to be able to capture all of that value. So there's all these kinds of, You know, complicated dynamics at play, but, um, certainly interesting times to be alive.
joseph-miller_1_03-25-2026_160407: You, you do need to find people that have spent time to think about, like, how do we make this thing productive in the real world environment?
How does it actually deliver value to your customers or to your own business? But then what you realize also is that, like, that is gonna take some time. But the bigger thing is, is that even if that works, let's assume that you can, You know, disrupt yourself or get these agents to be, um, value generated in your company. Um, what you'll realize very quickly is that all your other business processes need to be reimagined as well. Like it doesn't make like, I kind of like, the easiest way I would say is like, imagine that I can give you a thousand employees right now. If, if
seth-marrs_18_03-25-2026_160407: Yeah,
joseph-miller_1_03-25-2026_160407: thousand employees, you would work very, very [00:13:00] differently than you do right
seth-marrs_18_03-25-2026_160407: so true.
joseph-miller_1_03-25-2026_160407: And the problem is, is that because you don't have a thousand employees, you're not thinking about how you should be working if you did. it will take you some time to adapt all those other processes that actually have nothing to do with the agents. Work that they're doing, but more about your ability to manage the work, the amount of work that is going to be done now, um, that, that just changes the whole business.
Uh, the business. All your business processes have to be rethought that way. And also the company cultures have to be rethought. All of that sort of thing. All those, those sorts of things need to be, you need to be getting on them now because, uh, um, You know, if you wait until thinking, oh, I'll just plug this in later.
It's not like that. It's not that kind of technology. It's really disruptive to the way you do work, um, not just the work that you do.
joseph-miller_1_03-25-2026_160407: we made bets on obviously the, that this was gonna be the right move. Um, but also we built it in a unique way. Um, at the time, nobody was talking about knowledge graphs and grounding LLMs, like that wasn't a thing at all. Um, we were [00:14:00] building, we were, but because of my time at Bridgewater, I got to work with a very famous AI researcher named Dave Ferucci that really impressed upon me. importance of knowledge, representation and, and, and, You know, graph definitions and ontologies and this kind of thing. And, um, And so I sort of looked at LLMs and the transformer model and just, I, it's kind of like, You know, you're interpreting the mathematics, you're kind of reading the tea leaves of is this going to. Be able to do proper reasoning. Is it going to be able to do what we call like multi hop reasoning or complex reasoning? And I, I, our bet was like, no, this is actually not going to work that way. However, it could work if we can ground it in the domain in more explicit domain knowledge. And so, You know, two years ago. Uh, the first thing we built was actually, uh, actually we were always, we always had an ontology all along, but the first thing we built was the ontology that was built for an LLM to consume and be grounded in. And now people are even, I mean, just now people are starting to come around to the realization that, [00:15:00] oh yeah, if you ground these things in knowledge graphs, they do a lot, lot better. Um, but even still, not everybody is doing that, You
seth-marrs_18_03-25-2026_160407: Yeah.
joseph-miller_1_03-25-2026_160407: Salesforce didn't start that way. Um, You know, I, I think Gong is now doing a, um, claiming that they have a graph and
et cetera.
seth-marrs_18_03-25-2026_160407: Well, they're calling it, so they're calling it a revenue graph, so it's
joseph-miller_1_03-25-2026_160407: exactly. So now, now everybody is, uh, is real savvy on graphs. But, um, You know, two years ago, literally nobody was talking about that, which is, which is crazy because we didn't invent that concept.
That concept is as old as, You know, as old as dust. Like is the early seventies of people thinking about like, how do you do knowledge engineering and, and, and graph constructs and ontology design and such. Um, we were just. It to the forefront and bringing it to the, um, You know, to modern technology.
laura_1_02-13-2026_170244: Very good and very good context. Um, and while we were preparing for this session, like last week, I was thinking what is the most innovative thing that I'm seeing in, in B2B right now? [00:16:00] And it's not the tool. What I think is innovative is, more of a redefinition of the seller role, from a seller operating tools to orchestrating AI intelligence. Um, I have the fortune, to be working with a lot of like forward thinking companies, which realized something really important. Their sellers weren't struggling because they lacked data. They were struggling because they were overwhelmed by a lot of disconnected signals. And so instead of asking, how do I give my reps more ai, they started asking, how do we surround.
The seller with the intelligence that doesn't overwhelm them. And, um, one of our [00:17:00] customer at, uh, at Spark, spark is our annual conference. Um.
seth-marrs_5_02-13-2026_120245: Okay.
laura_1_02-13-2026_170244: Last year, they presented their own approach and they decided to, uh, sur um, to surround the seller with, uh, domain specific agents. One, for example, was focused on, uh, margins and deal economics.
Uh, another one was, uh, focusing on, uh, pipeline health and, uh, forecasting. Or another one was, uh, translating product updates, um, into, uh. Still radio messaging. All these agents were operating with the, from the same, um, trusted data foundation. And, um,
seth-marrs_5_02-13-2026_120245: Wow.
laura_1_02-13-2026_170244: When these happens, uh, the seller becomes the orchestrator. Um, And so they spend less time in gathering information and uh, more time in, um, uh, thinking critically and thinking about how do I move the deal forward? And [00:18:00] so. Thinking about your question, the biggest shift for me has been moving from thinking about the seller as an operator of agents to a orchestrator of um, um, of agents.
seth-marrs_5_02-13-2026_120245: Uh, that, that it's so timely. And when you say operator of agents, that's kind of like what a company would do when they're just. Asking a seller to do the same thing
laura_1_02-13-2026_170244: Yes.
seth-marrs_5_02-13-2026_120245: As they've done before with te with technology they've given to a seller. The orchestration side is these tools like pe people need to start looking.
I'm seeing the, the same thing. And it's, and it's, it's kind of that, that it's good you bring that up because I think that's the limiter in a lot of companies is they don't see it. They see it as a tool working for the seller instead of the seller orchestrating around them. And, and you almost have to look at your sales world now, not as a.
Bunch of just sellers, but sellers and agents making up your entire execution engine.
seth-marrs_5_02-13-2026_120245: what's the biggest challenge that [00:19:00] you run into in enterprise organizations when trying to deploy an agent to a, to a, to a Go-to-market team or a sales team?
laura_1_02-13-2026_170244: Actually, this is one of the question that I'm hearing more and more and, uh, the biggest challenge of deploying the agents isn't the agent or the technical implementation itself, is the data quality, as I said, and the clarity of the use. Because, um, as we said, the agents, um, learn from data. They act on context. And if your data, for example, is fragmented or the definitions are inconsistent, so inconsistent, or the systems are disconnected, the agent doesn't become intelligent, it's actually very confused it becomes unreliable, and that's when you lose the, uh, the trust of the field.
But the other challenge that you see is a lack of focus.
Um, there are some companies that are trying to build agents [00:20:00] for everyone, and um, if you build an agent for everyone, you build it for no one.
So The real question, um, when these companies ask, um, You know, can we deploy an agent? Um, I want them to move, um, to the other question, which is. What performance problem are we solving for here, and do we have the operational maturity? To support it. Um, and, uh, and, And so you might ask me actually, um, so do, do I need to get to a certain operational maturity before I'm able to deploy an agent? And, uh, the answer is, the answer is no. For example, there are the agency we're talking before,
like the rule-based Agents, um, um, this is straightforward.
Um,
I, for example, you can ask an agent to archive all the content that has not been seen in the last 90 days, for example. Really easy.
But if you want agents to [00:21:00] recommend the next best actions, um, in complex deals or, um, the, the, you want them to really guide the seller, um, on the, on the next steps that requires clean data. processes and methodologies and shared definition.
n_1_03-20-2026_083826: It's, it's, everybody you ask is gonna be very different. And, uh, agent washing is a thing, right? Like everybody, every AI in the product is now being called an agent. Is that really the case? The way we look at it? is agents are not just chat bots, right? They're not just like, Hey, I, ask something.
I get a response. Agents have to have a sense of autonomy. Agents have to be able to complete tasks on behalf of the user, they should be able to take a series of actions in the path of the task. or even a workflow, like an end-to-end workflow that it can orchestrate on behalf of the seller. Uh, so the way we look at [00:22:00] agents, as I said, there could be task-based
seth-marrs_13_03-20-2026_113827: Yep.
n_1_03-20-2026_083826: Are very simple agents.
I can complete one task at a time. It'll summarize, it'll create an email. Those are. Individual tasks that the agent is automating for you. But it could be the entire workflow,
seth-marrs_13_03-20-2026_113827: Yeah.
n_1_03-20-2026_083826: I can basically take an entire sales play and the AI is conducting that sales play. It could be your outbound motion, it could be your inbound motion. So multiple steps in that workflow, uh, could be fully orchestrated by ai. But I think for us, again, what's the step we have taken is. You have to be very careful in how much autonomy you give to the
seth-marrs_13_03-20-2026_113827: Yeah.
n_1_03-20-2026_083826: And that really depends on the use case you have. So we have given you all the controls, all the flexibility for the autonomy versus human in the loop. But depending on the use case, we see our. As customers kind of determining when to turn up that level of
autonomy Versus, no, this is a strategic customer. I never want this workload to be autonomous. It can still do a lot of legwork, but let the seller review what is gonna communicate to the customer. [00:23:00] So there's a lot of those things, but I think to me, the agent is really completing work on behalf of the seller.
That's how I look at an agent versus,
seth-marrs_13_03-20-2026_113827: Yeah, just asking a question, getting an answer. It's actually doing the work with you and, and it's, I mean, we talked about this. You guys have been very laser focused on the agent concept. I would say just like that, that part more than, than anybody. So as, as you have, uh, like brought, brought that and looked at bringing that to life, can you walk through what you think that you are doing and what you've learned since you've started so early on this that is different and unique to other, you talked a little bit about it, which I, I don't wanna underestimate some of the things you talked about with controls, because that is.
Kind of where it's at, right? Like you can build an agent that can just go do anything you want. No enterprise organization is gonna let you deploy that. So you touched on it, but I just want to double click, or I just wanna emphasize what you said. That control lever, [00:24:00] the, the control panel that you're putting in place behind the scenes is what's gonna make this, or at least my perception is what's gonna make this enterprise ready?
Because if you don't have the ability to do that, it's gonna be, it's gonna be challenging. But in addition to that, like what are some of the things that, that you, that, that you are doing today that you think are, are that, that are a step ahead, that are unique, that are adding value that, that, that you've gotten from this advantage that you've had by starting so early?
n_1_03-20-2026_083826: Yeah, as you said, I think our, our focus right from the beginning, uh, as we got on this journey, especially in the last year where we've gone really deep, is that autonomous agents, uh, that we can bring to your, uh, and. Our focus is to make sure that the agents are delivering the right actions to the sellers at the right time.
Like that's kind of a, at a high level, always simple framing, and it's orchestrating all these workflows to bring those actions. What do I mean by that? These could be different type of sales motions,
seth-marrs_13_03-20-2026_113827: Yep.
n_1_03-20-2026_083826: You can run your entire outbound, the industry calls this ai, SDR, but for us, that's one [00:25:00] use
seth-marrs_13_03-20-2026_113827: Yep.
n_1_03-20-2026_083826: Of, of our autonomous agent and ai, SDR.
So you're targeting it to your specific ICP can go find prospects in that ICP can research them, generate that personalized, relevant outbound messaging. It's all done by ai and at the end of it, again, you have the choice because we had, I think the Advantage Outreach has had is we have thought about automation so deeply since the beginning of outreach,
seth-marrs_13_03-20-2026_113827: Yeah.
n_1_03-20-2026_083826: Was our core. So we already had these knobs where the human can intercept through our cadences and sequences. And so you can say this email is personalized. It goes into the sellers inbox, or this email can be fully automated. Like we had that flexibility already in the platform, except AI is coming in helping you in every step of the way now. Right? So. that's just one use case. I think I, the, the barrier, it's another use case is inbound. Like you have an inbound lead coming in. Again, I automatically research that inbound lead and, um, I have to say inbound lead is probably the ha um, biggest use case where it's getting fully, it's fully autonomous because there's [00:26:00] intent. more, there's more comfort because You know what the user is looking for so you can target them more effectively and make it fully autonomous. Speed to lead matters. So in fact, even in our internal implementation. We were writing the messages for the sellers, and then it was still sitting in a seller's queue.
And then we checked those messages and they were amazing. And again, our CRO said, why is it even sitting in the seller's queue? Just, just send it.
seth-marrs_13_03-20-2026_113827: Yep.
n_1_03-20-2026_083826: Uh, and the results were incredible.
n_1_03-20-2026_083826: I mean, you as a user really don't even care about the data, honestly.
seth-marrs_13_03-20-2026_113827: Just want it there.
n_1_03-20-2026_083826: yeah. So all I need is the action, right? Like, is the action and the action could be done autonomously, or the action can be brought to me. So that is our focus. Like, like we harvest all the data and the signals. We orchestrate it via our agents and we deliver the action to the seller. So this is it. It's simple. It's like our focus is an AI that acts. It's not about ai that answers ai, that gives insights. I think that's still a lot of work for the seller to go sift through that we simply get you to what you need to do.
seth-marrs_13_03-20-2026_113827: That makes total sense. So let's expand on that a little bit. When you think [00:27:00] about like all these agents are gonna be working alongside sellers and enhancing in some places, optimizing, providing capacity in others. What do you think the core job of the seller is going to be in the future when these things are fully deployed and working with the salesperson to help them sell?
Like what's that core job that still is going to be what they are doing to make a difference with their customers?
n_1_03-20-2026_083826: Again, when I take that example of coding and and revenue, it's very different. These are very different use
seth-marrs_13_03-20-2026_113827: Yeah.
n_1_03-20-2026_083826: Pattern recognition. You can do the same thing and get the same results. is touching people, touching humans,
seth-marrs_13_03-20-2026_113827: Yep.
n_1_03-20-2026_083826: You probably are not going to get the same result, even if I do the exact same
seth-marrs_13_03-20-2026_113827: So true.
n_1_03-20-2026_083826: Person, right? So this is where the seller's creativity, the seller's judgment, the seller's ability to build relationships. still really, really essential. Like AI can do a lot of that legwork, but what you do with it. Is really where the seller is [00:28:00] still going to be the pretty critical part. Otherwise, you're not gonna see the success you need to see by just letting AI do uh, everything that you need.
And the other thing is AI look around the edges, right? Like, You know, information about your deal, about your buying committee, right? So you, they see information. I'll take example from my own sellers, like my own best sellers at Outreach for what they do, right? They're coming to me, hey, or coming to Nadi and saying, oh. are the current people in the deal, but I don't think I'm at the right level. Uh, that's not the kind of information he has and who should be the right person? How do I break into this account and how do I build more influence? What is the path to get around this blocker, which was not evident from the data I have rightly.
These are the things they are getting ahead of, they're thinking through it. And those are the sellers, right? That that kind of skillset is where the sellers are. Absolutely critical. And, um, You know, their value is going to be in that and also gives them more time. I think the [00:29:00] beauty of AI is because it takes that, the legwork off their plate, they can spend more time doing that, which is what's going to get you to the outcome you're
looking for.
seth-marrs_13_03-20-2026_113827: It's gonna be, and I mean those, if you, well, those things you said are what makes great sellers, great sellers. And I think to a certain extent, the capacity that they have to do that is limited by all of the problems they have.
n_1_03-20-2026_083826: Yeah.
All,
all that grunt work and they really are not able to do the strategic work, right? Like this is the
strategic work to
seth-marrs_13_03-20-2026_113827: Yeah. And it is so interesting, like how that works because everyone thinks of, I'm going to do AI and it just does my job for me. But that's not the way the world's gonna work if you're talking to another person because they're not, you don't know what they're gonna say. Like, so whatever they say, you have to pivot.
Yeah. That makes, that makes total sense.
n_1_03-20-2026_083826: The one other thing as I'm thinking about this is also right I, where is AI learning from?
Speaker: And that wraps up another episode. Thank you for joining. For show notes and other episodes, visit us@innovativerevenueleader.ai. The Innovative Revenue Leader is [00:30:00] sponsored by Sandler, a Trilia company. Sandler provides top corporate sales and business development training while empowering sales professionals and leaders to master the graph of selling at all levels.