The three things a RevOps leader can operationalize with AI right now. With Lolita Trachtengerts

Speaker A: And you think about it and you're like, is it actually moving the needle for me on what I do? And sometimes it just doesn't. It creates so much work for you. And for me, when I get so many things, so many insights that I need to act on, and no guidance whatsoever. And what. Wait, what does the analysis even say? Like, forget the pretty dashboards. Tell me what it means. It just creates so much pressure and unnecessary. Like, I have so many things to do anyways. I don't want that extra pressure from that tool. You should look into your frankest.

Speaker B: Hi, everybody. Welcome back for another episode of Revops Unboxed. And we have another awesome guest coming on the show today, and her name is Lolita. So I'm going to go ahead and have her do a little intro about herself. Like, tell us about you, what you're doing, um, kind of what you're doing in the Revoff space, and then we'll go ahead and just kick it off.

Speaker A: Yeah. Thanks, Tana. Thanks for having me on. I'm excited to be here. It's my first podcast. I love it. Yeah, I know. I love it, too. So I'm Lolita. I'm a chemical engineer by training. I'm in Revops by mistake. It's an accident.

Speaker B: I love that.

Speaker A: Um, I'm currently VP of, uh, Go to Market Ox and Growth at Sports Spotlight AI, where we're building. Um, deal Autopilot. Before this, I spent my entire career in industrial water treatment. Um, you know, like a decade or so without giving away my age. But, you know, started off as a chemical engineer, project manager, operations manager, leading team, so on and so forth. Um, had so much fun, really. Was, you know, was looking at the transition into tech or looking at, like, from, you know, it's just interesting. So different from everything that I did before. So, um, it's fun. It's a different industry, but I still think it's the same job. Like, I'm an operator at heart by this time, you know, by being, um, by getting to this point, I know that for sure. And so, you know, when you think about operations and efficiency and making everything work better, it does, no matter what industry you're in. So that's, um, me kind of figuring that out. So, you know, doing that on pipes and now doing that on pipeline, kind of the same thing.

Speaker B: Uh, that is incredible. We're very similar in trade. Because I'm going to ask, how did you mistakenly get over to this? Because I was in nursing for, like, 10 years before I made a pivot to tech. So I say that same exact line of like, it doesn't really matter where you go, what industry, if you know operations. You know operations.

Speaker A: Yep, yep.

Speaker B: I love. Okay, so how, how did you mistakenly get into revops?

Speaker A: Um, um, I got into Revops through, um, basically one of the co founders of the company. I worked with him in the past. We were both in the water space. So he knows me, I know him. And when you know, it kind of, um, made sense timeline wise. Where I was, where he was. It just, you, uh, know, kind of offered me the opportunity to join the company and I was like, yes, 100%. Um, I'm definitely interested. So that's me. Yeah.

Speaker B: And now stuck in this ever changing world of now you're doing AI instead of chemical engineering. I mean, how cool is that?

Speaker A: It is very cool. And honestly I've always wanted, uh, to, you know, kind of do software. Like, it wasn't something that I was good at. When I was studying chemical engineering. We had a course for C and I almost failed it. It's a true story. It was so, so bad. Anything that has to do with software and writing code is just a language that some people feel very close to and understand it so easily. And some people are just like, what the heck is happening here? Um, so I'm on the other side. Um, I'm, um. The people who are like, what is happening? What am I looking at? But I was always so interested to learn how it goes and just how different that is, how easy it is. Somebody would code something and something appears, you have a product all of a sudden. Whereas in the water industry that doesn't work like that. You have to build a facility, you have to wait piping, you have to do all the design work. It's so different. I was always super interested in software, always super interested about coding. And I'm sure we're going to be talking about it more, but now anyone can, which is amazing.

Speaker B: Yes, that is a great segue into. Because I know one of your passions is everything AI and being at Spotlight AI, I know y' all are doing a lot of innovating when it comes to your own product. What are you seeing really happening right now with like overestimating AI today? And what is, what is overestimated in AI? Because everybody thinks, you know, AI can do everything. And then what is something you think is underestimated in AI? Not really being talked about that much.

Speaker A: M. Um, so I think at this point, overestimated, I would say anything that has to do with like summaries and giving you insights. Um, you know, there's so many software out there that can summarize emails for you and Slack and whatever, you know, um, meetings, calls, like everyone's talking about summaries and getting insights in. And I think it's a little bit of an overhype at this point. Um, and underestimated is AI that actually takes action. Like I think you know, from day one, Chris Sparkly, that was what we were trying to do is make sure that there is an actual action that is happening, something is moving something for you and not just giving you the data.

Speaker B: So I would say that, yeah, it's really interesting because all the conversations we're having, stuff that's taking action, like things going from co pilots to Autopilot, like the uh, age gentile stuff that's happening with you can program something to actually do actions for you, set the workflow and go. Have y' all been doing any of that lately? Or like, what has been in your, um, part of it that you've been able to build? That's more of that action oriented play?

Speaker A: Oh yeah, that's a good question. So co pilot versus Autopilot, you know, people have been talking about it for a while and I always struggle to explain what I mean by co pilot versus autopilot, especially when it comes to spotlight. Right. You meet new people, they ask you what you do and you're trying to explain that and it's really, really tough. Um, but lately what's been working really good for me is comparing Copilot and Autopilot to chatgpt like three months ago and how we were working with chatgpt three months ago, um, to Claude nowadays. And I seem to have success in explaining that that way because up until this point it would be like a blank face of people like, ah, ah, Yep. Don't, um, know what you're saying, but okay. And now it's like, oh, wait a minute, I totally get it because, you know, when you were working with ChatGPT, or I should say I was working with ChatGPT, but I think most of us it was for drafting emails or getting a summary or creating some sort of PowerPoint, but you were the one who was doing the work. Like it's a correspondence and you would get feedback, but then you have to do a lot of the actual work and a lot of the action. And then Autopilot is like Claude, where you have cowork and you have cloud code and now all of a sudden you can actually build flows and workflows. For yourself of if you have this data, what actions should I take and what actions do I feel comfortable giving away to an AI? Right. So I think that state of mind and kind of that shift is currently happening. Like I can see people, um, reacting to that more and more. But for us it's Spotlight. Like that's what we've been trying to explain all along. That's the funny part. Like that was the whole thing. Like we kept saying, don't talk to us about insights. Let's talk about act. Like, what does the AI actually do for you? So I'm having a lot of fun with this now because it's finally, you know, kind of, you know, in the world of MCPs and connecting things to Claude, people are finally really on the same page. And I don't get that blank stare of like, I have no idea what you just said. It's finally like it's clicking.

Speaker B: Oh, absolutely. I'm. I'm on the Claude train by the way too. So have you built an agent for yourself yet to do any of your revops back office work?

Speaker A: Um, I don't have to for web apps specifically because we have Spotlight. So it's um. And you know, now we have Spotlight on MCP connected to Claude, so it's even better. Like it's a completely new world even for me. Um, but um, I built a bunch of agents for the go to market function. So because I have wear a bunch of different hats. Spotlight, we're such a small organization. Small, fun and lean. Um, on the go to market ox, which is, you know, that's the way I separate my daily workflow.

Speaker B: Right.

Speaker A: Monday Lolita is doing this and Tuesday L is doing this. So for the more of a top of the funnel functions, um, that we don't have, you know, it's not on Salesforce, it's not something that we run through Spotlight nowadays. That part. I've built a bunch of them, uh, which I really, really love. And especially on the marketing side on the growth end. So many agents, so many workflows. It's really, really neat. But then on the rev op side, it's been a really smooth ride for me because of Spotlight. That's what we do.

Speaker B: Right, Right. So out of all those agents, where do you think your agents are having the most impact? Like, where are you seeing the value in the ROI on those now with the go to market team?

Speaker A: Um, so they. Wait, what? Say again? And we can edit this later?

Speaker B: Oh, ah, absolutely. We'll go back through this with the ones that You've built for the go to market team right now, where are you seeing the most impact, like the most value coming out of the agents, whatever it's doing.

Speaker A: Oh, um, so right now it's essentially, you know, on the go to market is all about the funnel that I can bring in. And we're seeing just better, um, I would say personalization and better intense signals on that side of the world. So basically we don't do any spray and pray, you know, like we don't send a bunch thousands of emails. It's not something that we do as an organization. So, um, for us it's more of can we still limit ourselves to, you know, a very limited amount of emails and outreach sequences, but make sure that we tailor those to people who are actually showing some sort of intent. And um, getting that intent is really, really working well with different agents that you can build while monitoring different platforms. So I think that is the biggest ROI because we're actually seeing higher conversion rates.

Speaker B: Oh, that's amazing. So thinking about that and you've built the agents for the go to market team, how do you think orchestrating multiple agents because you've got your customer success, you've got your marketing, you've got your sales team, you've got all the different functions. How would you, I don't know, advise people that are thinking about using agents across multiple teams without creating like another Frankenstack.

Speaker A: I love that she used Frankenstack. Thank you for that. I'm so proud of that. Frankenstack. Yeah, it's like our, um. So just to tell you, we all have them.

Speaker B: We all have them somewhere in our org.

Speaker A: Everywhere, I would say everywhere in the org. Um, but yeah, um, Frankenstack. Just a funny story about that. So, um, I have like, I'm fangirling over this one person who is running growth at a very big company. And I actually created a skill on Claude that is a replica of that person so I can run ideas by him without knowing him. And it works so good. It's really, really cool. I love that. So Frankenstein came from that. And it's such a proud moment for me because it's like really cool. It's like a really good word. Um, so thank you for using that. But yeah, um, I think at the end of the day, um, you have to. If you have multip agents, especially in an enterprise organization, which is not, you know, for us, for Spotlight, it's easy because we're a small team. You know, everything is controllable in a very Very efficient way because we're so small. Right. Like everyone knows everything. Um, but then when you're in a bigger organization and you have that Frankenstack of like so many tools, so many data, so many things, you really need to have some sort of brain, some sort of knowledge that combines it all together. It's like having multiple agents in a way, you have multiple agents but you need to control them under one orchestrator, brain, knowledge graph, whatever you want to call it. Um, and I think um, for Spotlight specifically it's um, it's pretty easy to do because what we did is we've developed that knowledge graph over the past four, maybe five years. So all the data and all the industry knowledge, um, you know, playbooks at work, methodologies that work are all in there. And then you have something that guard rails what your agents are doing.

Speaker B: Yeah, I think that's the biggest part which will lead me to my next question. When you let the agents free and loose to go do their thing, how much control do you think needs to be? Human judgment versus the autonomous agents being able to do any kind of like complex stuff or things that you're assigning them to do. How much do you think the humans have to still be involved to double check at all?

Speaker A: Um, I think it depends on what you're doing. Right. It depends on the workflow, it depends on the, on the agents and what they are actually performing. But I think right now like majority of the organizations or at least the ones that we're talking to humans are doing everything. There's very little AI. Ah right now especially in the enterprise organizations, you know, where it's very hard for each person to make a decision. It's you know, buying committees and everything needs to be approved. And the AI that they have is still more of an insights AI and it requires human judgment to do anything honestly. Um, but I'm hoping, and I think that there is, right now we're going towards that shift where you actually employ agents to do more and they can judge what to do and they can act on behalf of you in the guardrails that you set. Like that is the, that is the, you know, the most important thing. Also you want to make sure that they have the data to work from. That is not only the web. Like if you build your agent on Claude and you just assume that it knows how your organization is running, that's going to be painful. Right? Because the hallucination rate and like sometimes, you know, really. Right. Like sometimes Claude just wakes up and you're like Wait a minute. This is not.

Speaker B: What did you just say?

Speaker A: That's not right. Right. So I just think, like, we need to be smart about it. But I think definitely, if your AI has been trained and it's properly executed and, you know, you have the brain behind it. Yeah, why don't you let it work? Run, you know, make actions. Why don't you have it go through your, um, opportunity and update the, the stage that you're at, or create a material for you relevant to that stage. Like, why not? You don't have to do it yourself.

Speaker B: Right. That, that makes a lot of sense. And what I'm seeing too, we've gone through a couple different tool implementations recently. And to your point, I think a lot of the AI that's being built in tools is insights, you know, because all the data's in there, whether it be your CRM or, you know, activity trackers, whatever you're using for your activities. And most of that is pulling insights. Like what did you say during the meeting? What did you say during the phone call? What did this happen? But what was very interesting is the one that we're going through now. And I'm not going to name names because I'm not giving them props right now, but there is a lot of that same stuff you were talking about, like setting up scorecards, playbooks, instructions, so that when you're using the AI, your team can say, okay, based on this, show me these things. And I think it is giving, like you said, more in those instructions and swim lanes and guardrails that you need to actually pull. It may be insights, but it's also data. So how, you know, how can we move the deals farther down the line? Because we've got the right data being pulled in. So I find that very, very interesting. So do you think, in your opinion, as we talk about specialized agents per function, do you think, as we, especially with you being a smaller org, which will be interesting to see what your take is, that you're going to have more of unified revenue brain, you know, to where the entire life cycle is kind of run by agents while we all manage it in the background.

Speaker A: Um, I think we, uh, I think we'll have so many agents. Like, I think there's going to be more and more, but I do think we'll need some sort of a unified brain. Yeah, it's like you have a unified brain and specialized hands in a way. Right? Like you have a bunch of agents who are executing and doing things, but you do have some sort of a brain that Controls it all and make sure that you're actually doing the right thing, getting the right data. Um, you know, it's providing what you need and not just running around doing whatever.

Speaker B: Yeah, that makes absolute sense. I would agree. Since we are so used to saying, you know, this is our source of truth, or this is what our goal is, or this is what we're trying to work towards. Almost getting your agents to have that same behavior, to say, okay, this is, this is part of the goal, and here's where we need to get them to go. What do you think are the three things that say, you know, because there's a lot of people that are just getting into AI, they're not as far in as some of us are. Um, and they're just now putting AI, uh, initiatives on the table. What do you think could be like, the three things that a rev ops leader could operationalize now for AI and then also, how can they do this and help change behavior versus they're just building reports and dashboards?

Speaker A: Um, three things I would say. First of all, I would kill anything that doesn't do any action. Like nowadays, like, if you have tools that are just nudging people to do their work, get rid of it. It's not helping you in any way. Um, I would automate a workflow. Like, let's start with one. Because the tendency and especially the noise around. I know all of us are on LinkedIn. Everyone's talking about how they built a hundred different agents in one weekend and now they don't need to do anything. Or like, the one that I love the most is the one that, what they're talking about. Like, oh, look, all of my calendar is like booked with meetings with Claude, and I didn't listen. Like, dude, seriously, um, no, don't do that, don't try and do that, but do one thing, automate one thing. Like automate a workflow that you know and work with and just start there and then it's going to give you the confidence to move forward with more. Um, and the third thing is pick the workflow that is actually broken today. Because there is. Then there is no harm done. Like, you either fix it and it's now it's better and now something is more efficient, something works well, you don't fix it and then, you know, no harm done. So you don't.

Speaker B: That's a very, very good point. Yeah, I mean, it's already broken, so if you can't fix it with something easier, it's. It's already where it is. Now, that's a really good point. Where do you think most AI initiatives fail nowadays?

Speaker A: Um, I think with the hype, like not defining the problem, it feels like. And again, you know, it's a. It's very, um, I guess each company, like, I can talk for all of them, but I seem to stumble on a bunch of companies, my friends that work in different companies as well, that just initiate AI projects because AI is cool, because of the hype, because of knowing, oh, we need AI, uh, like, let's throw AI into this, but not defining the problem, or not even starting with the real problem. And that's a failure to begin with. That's just a shame.

Speaker B: Yeah, that's a very, very good point. And since we were just talking about the system of Truth, what do you feel like with everything that we have going on with AI, do you think our CRMs are still going to be like the system of truth going forward?

Speaker A: Um, so it's interesting because just this week, right, or I think maybe last week, CRM Salesforce announced, um, the headline Headless 360, which I think is a really interesting move. And I think they're actually understanding that they're not the source of truth, and that is them trying to be it again. Um, so what I'm trying to say is, um, the CRMs are a little bit late. Like, you don't get the data that you need at the time that you need it, because a CRM right now, you have to really work to fill it or work to update it in order to get something out of it. Right? And, um, you really want the voice of the customer, um, whether it's, you know, the voice of the customer talking to you or the voice of the customer in general out there, you know, announcement that they make things that is available online. You want that available for you in order to act on an opportunity that you have. So. And the CRM doesn't have it, But I think Headless360 is exactly that. They're trying to say, put us, you know, connect us with MCP to clutter, whatever it is that you're using, and just use us with other tools. And then you have the voice of the customer and every truth in Claude, like, you know, would multiple MCPs. Um, so that's a very, uh. Yeah, it was a very long answer to your question.

Speaker B: No, uh, honestly, and I think about it to that point too, because we're always, We've been saying, I mean, as long as I've been in operations, I've always said our CRM is the source of truth because that's where our customer data is. And we've got platforms like Salesforce that have everything in it. You've got your marketing, you've got your finance, you've got your revenue, you've got your sales. Everything is there. And it's almost as if you know, they've got their marketplace and you attach all these third party apps. Now when you can attach that other stuff to Salesforce, it's almost like they position themselves as an app, as a third party, because now you can kind of connect the whole ecosystem together, which is something Salesforce loves to do. Um, because I said the same thing with HubSpot. They haven't come up with a new thing, but all of that stuff we can integrate into it and get the same functionality across their whole set of stuff. Like you're not paying extra to use cloud in marketing or sales or revenue. So you can still kind of keep it all inclusive and keep it all together with the 360. Which, uh, which is still going to be. It's going to be fascinating how I think over the next year, year and a half, because I do see the trend to your point. Like my whole LinkedIn network is all about cloud, which, you know, I'm using it too. So I see what the hype is and I'm super excited about it. So I think it'll be really interesting to see we have this chat a year from now. And now what are we doing?

Speaker A: Oh my God, seriously? Even, uh, even tomorrow, I think I just an article that says that OpenAI is back in the running and something is happening. And I was like, oh, wait, I have to read it later. I'll send it to you. But I don't know, maybe, you know, tomorrow, this conversation, maybe, who knows?

Speaker B: I feel like anything is possible at this point.

Speaker A: Everyone's gonna be like, who? Claude? Why are they talking about Claude?

Speaker B: Right, right. So as we talk about AI and going through the different skills and we just talked about fast forward, what do you think revops is gonna look like? And I had in here, I was gonna ask you two to three years, but what do you think RevOps is gonn next year with AI?

Speaker A: I, um, think it's such an exciting time for RevOps. Honestly, like, no better time. I think up until this point when I was thinking of Revops, especially not coming from a Revops background, but having Revops team work as part of a sales team that I was managing, it was always, to me, it was like, oh, people who do dashboards people who prepare the pretty slides that we're going to have, people who are going to clean up the everything in CRM because my account executives are, you know, don't want to do it. But now it's like we are at the very front of everything that is happening with AI. So I really think that this role is becoming more strategic than anything else on sales honestly. So I'm really excited about it. I think it's going to be really different. You're really going to be working with workflows, automation, operationalizing things that you weren't able to do because you had limitations with the software, you had limitations with your organization. But I think we're going to be given a lot more um, you know, a lot more freedom to run with different things that we want to do because we're going to be the testers, like we're going to be at the front of the AI.

Speaker B: I agree and it's interesting because I just saw, I think it was this week and I love when people put the visuals together because you know I'm a big uh, advocate for like rev ops departments, you know, and how to have the right roles in the right places to support the department. But it was interesting because I saw somebody do a new mock up of what a RevOps org would look like with the AI. You know, you've got an AI, I don't know if you want to call them an analyst, a developer, a go to market like in all these other separate roles that really are not defined in a RevOps org because they are usually living with um, IT or somebody else on the technology side because you know, sometimes we are the owner of our technology stack. But I'm seeing a lot of orgs change that thought process because your rev Ops leaders are really the ones driving the AI adoption usage business, um, cases, all of that. They still should own the tool or whatever it is. So it is very interesting to kind of see maybe just even how the reorgs like, of RevOps in general kind of change over time. Um, so, so I got some spicy hot topic ending questions for you. They're not too bad. So if you joined a company tomorrow, what would be the first thing that you ripped out of the tech stack again?

Speaker A: It has to be like all the insights tools and that's exactly what I've been doing, you know, um, because it's just, it doesn't cut anymore. It doesn't cut it anymore. You know, you, you're being overloaded with data but then you need to make a decision on what to do with that data. And sometimes it's so much data because, you know, LLMs can handle so many things, right? So they just throw it all at you and you're looking at it and you're just kind of, you know, uh, you don't even know where to begin. So it becomes that next week's problem and it becomes somebody else's problem and then nobody does anything. So any tool that doesn't provide action and just provides you insights and data, I would just get rid of it completely.

Speaker B: You're going to make me go assess, like my entire tech stack and all of that stuff now?

Speaker A: You should. I do, I do. Like, that's what I like. Uh, seriously, that's what we should be doing. Because some of the tools you've been using for a while and you're like, oh, I like this because the UI is really nice. Or, you know, you have a contact in there and you think about it and you're like, is it actually moving the needle for me on what I do? And sometimes it just doesn't. It creates so much work for you and for me. I don't know if it's just me, but I feel like when I get so many things, so many insights that I need to act on, and no guidance whatsoever on what. Wait, what does the analysis even say? Like, forget the pretty dashboards, tell me what it means. It just creates so much pressure, unnecessary. Like, I have so many things to do anyways. I don't want that extra pressure from that tool. So, um, yeah, you should look into your frankest.

Speaker B: I absolutely. Uh. Or just to compound on that, it's. Or you're getting the same insights to multiple tools. Like, multiple tools are giving you the same exact thing, but they've written it into different word. And then your team is like, well, which one do I look at? Because I'm seeing the same thing in three different places. And which one do I use? That. That's. Yeah, I feel like one of them.

Speaker A: And I do kind of, um, going back to, you know, having that brain that we talked about, that orchestrator that can look at everything because, you know, like, I don't know, I'm not. Again, I'm not a software engineer. I now can run Python code using Claude, but I don't understand what it says. I can just test it and it runs. But, um, other smart engineers on Spotlight and the way that we design the system to really understand the context is just amazing. You need to have that brain to

Speaker B: connect the dots 100% agree. So do you have any parting wisdom to share? Any, I don't know, mottos you live by? Parting wisdom could be about anything in general from the Lolita. Parting wisdom.

Speaker A: Love it. If we stick to Revop's wisdom, I would say, and it's controversial and people are going to probably not like it, but your pipeline is a lie. Like, I say it all the times and in Munich, I wish, I truly wish to meet an organization, especially an enterprise sales organization, that's going to come and show us their pipeline and actually believe in it themselves, especially the leaders of that organization. And for us to be able to say, yeah, you know what, we didn't need to do anything. This is the great pipeline and your CRM is in great shape. So, uh, yeah, your pipeline is a lie. I'm sorry.

Speaker B: That. That'll be. That'll go along with my other it's fixable T shirt. I think we're just going to start branding. Branding T shirts bad. Yeah, that might be coming along with us.

Speaker A: You should do like, a T shirt that has the episodes on the back, you know, when. When there's a tour and they have like the tour and all of the tour in the back of the T

Speaker B: shirt, you know, And I'm going to have them edit this out so nobody can steal my idea and take this whole part of it out. So thank you so, so much for joining me today. It was an absolute pleasure. I know we'll have plenty of other conversations outside of here, but just wanted to tell you thank you so much for taking the time today.

Speaker A: Thank you awesome so much for having me. This was fun.

Speaker B: Thank you.

Speaker A: Thank you.

Speaker B: That's another episode of Revops Unboxing with me, your host, Tana Jackson, Click subscribe like, and all of the good things on YouTube, Spotify and Apple. Thank you. Bye.