Latest / Rev-n-u Unplugged / Beyond Automation, What AI Is Actually Doing In RevOps

Transcript

Beyond Automation, What AI Is Actually Doing In RevOps · Rev-n-u Unplugged

• james: Welcome back to Revenue Unplugged and to the second episode of our RevOps season. I'm sitting down with 10 RevOps leaders this season to try our hardest to demystify RevOps for CROs. Just a little. As my guest today points out, if you ask every RevOps leaders what RevOps is, you'll get a different answer from each of them. Joining me today is Anu Krishna Kamara, Chief Strategy Officer at Blanco. Anno's a RevOps person at heart and we're talking about what AI is genuinely doing in RevOps at the moment, beyond just automation. He explains how his team built a system that means people stop treating RevOps as the company's information desk. He talks about why he's never let AI run the forecast, because as he puts it, you can't choke AI's throat. And it says why those LinkedIn posts claiming AI has replaced SDR teams. Don't tell the whole story. If you enjoy these episodes, please give them a like and subscribe. It really does help us attract and get the guests like Anu. Right, let's get to it. Grab a coffee, pick up a pen, take some notes. This is Revenue Unplugged. Anu, welcome to rev-n-u Unplugged and thank you so much for agreeing to be a guest on this special season that we're putting together where I get the opportunity to talk to 11 incredible Rev Ops leaders in a hope that we might demystify Rev Ops just a little by the end of the season. As a way of getting started, I always ask my guests just to give a brief background as to who they are. So Anu, who are you? Anu Krishnakumar: I think that felt like a philosophical question, but I'll try. james: Yeah. Anu Krishnakumar: but I I mean is it said I'm a new, I I started in management consulting, consulting and moved into sales and kind of just like I would say probably most RevOps leaders would say, kind of fell into operations. and it just felt like a good mix of consulting and sales, right? Like it's like I I felt like I was consulting within the company for our sales team or the go to market team. felt like a I don't know, really good fit and and it actually kind of came through in my work clearly and kind of just went up the ladder and over time kind of absorbed a lot different functions. right now I am a sh chief strategy officer at Blanco, which is a a broader role. I have the RevOps team under me. I'm still a RevOps person at heart, but I also manage you know, customer support, professional services, IT and security and all that. but you know, as always, when when reached out, I was like, you know, it it's it's not just about, you know, talking about Revops, obviously it excites me, but it also gives you an opportunity to kind of think through things, you know, james: Yeah. Anu Krishnakumar: that you normally wouldn't pause and reflect on. So I was really glad to be here. james: Amazing. Well, the title of what we're going to be talking about today is Beyond Automation. What AI is actually doing in rev ops. Before we get into that, and I have some great questions for you on that. If you could describe what rev ops is, the function rev ops is, Anu Krishnakumar: Yeah. james: how would you describe it? Anu Krishnakumar: Yeah. It's it's actually interesting that you said in your intro about, you know, demystifying RevOps, which is absolutely true because you could ask you know, every RevOps leader, you know, and they will all have different definitions. the way I would like to think about RevOps, you know, especially because it can have so many it it can have the analytics, it can have the systems, it can have the process component, so many things in it. I think of that the role and the leader's role as someone who enables the revenue engine, right? And that is different in different companies, different types of businesses, right? You could be a you could be a very traditional go-to-market business where you have like this sales team, SDR team, marketing team, all that. You could be a product led growth company where it's, you know. primarily just, though your product drives the function and maybe you have a small sales team, you probably have a marketing team, but it's, but even there a RevOps team has a role because I see the RevOps role as, you're finding opportunities to streamline your revenue engine, right? Reduce revenue leak, you know, get better predictability in your revenue and when you you know increase the you know the productivity of your go to market channel, whatever the whether that's human or you know, machine or whatever the case may be. so that's that's how I see Rabops role. james: of it. So thank you for sharing that. going into topic one, where we want to look at where AI has genuinely changed a RevOp leaders day to day versus automation with a new label. So a lot of what's sold as AI and RevOps is really, in a lot of people's view, is just automation with a new different Anu Krishnakumar: Mm-hmm. james: badge on it. In your view, where has AI genuinely change what a revops leader does on a day to day basis and where is it just automation labeled differently. Anu Krishnakumar: Yeah. interesting. So I think it's people think of AI as our automation because you know, y that's what you hear a lot about, right? Like it just look at this. Now AI does this thing itself, right? Like james: Yeah. Anu Krishnakumar: and especially if you follow not follow, hopefully, but if you just see the LinkedIn post, you know, people make it look like no, we let go of our SDR team and it's AI does everything, right? It's like People think, great, we can automate everything. I I don't know if it's n it's not quite at that s spectrum, but I if if you ask me to define, you know, where AI genuinely or when genuine AI genuinely has an impact, I would think of that when it really impacts workflow, right? And by that I mean who does the task, right? How it gets done. maybe where it gets done, right? if that involves automation, great. but if it is just making a task that was done before just a little bit faster. And maybe, you know, y didn't really need AI for that, right? I mean it's great. I mean it's it's any any productivity improvement. I'm here for it. Right. james: Hmm. Anu Krishnakumar: but I think the real transformational power of the technology comes from transforming our day-to-day work, transforming our workflows. so th that's kind of how I how I define how I like to think of it. sorry, I think your question was w where it has actually transformed. so getting to that, I think in I think in in my experience, What I'm most excited about is, and I would probably say most RevOps leaders would agree with this, in the, f RevOps hasn't existed for the longest time, right? I would say probably 15 years, 20 years maximum. And through all this time, people have been trying to, you know, increase people's productivity, predictability of the business. But we always get caught up in the as an information desk in some ways, right? No matter the level of documentation you create, and people have been creating automation even before AI. So automation you create, we always get pulled away from work for the productive work for you know, you know, what would be like silly questions where is the forecast? What is the forecast? what stage is this op in? What should it be in? Did we send a quote? It's just like you're always answering these questions that are not really moving the business forward. It's just people find it so much easier to just ping someone or send someone an email and get a response. That's how humans, that's the easiest way for us to do it. as opposed to, you know, actually finding opportunities to improve the business, right? And I think with AI, something that we have been experimenting with and which I'm really excited for is now you can take all this documentation the teams have created, the supporting materials that we have created, or just even giving access to AI to the data and the systems that's available, Salesforce or otherwise. You can you can w or we now have a place where our sales teams are go to market teams, marketing, whoever can ask these questions without going to the RevOps team, right? So they could and that they can ask them in a natural language and they give a response that a RevOps leader or someone in RevOps role would have given them that answer. And and and they are satisfied and now every they get the answer almost immediately. And the RevOps team is not distracted from our, you know, the actual work that we That we want to be doing. And I think that potential never existed before. You you know, we have tried to solve this problem by creating documentation, creating dashboards, right? You know that, you know, tableau and local. They they solved it to some extent, but people go look at it and sometimes they don't even know what which is the right dashboard to go to. They still used to come to us. But now, AI can entirely eliminate that function. that RevWops was playing, right? And it wasn't the most productive or interesting work that we were playing anyway. So I I'm I'm glad to be rid of it. james: And so for somebody who, like myself, who doesn't have huge technical nous, are you using Claude to build that or is there a third party system that's part of your tech stack that you're using to be able to do this? Anu Krishnakumar: So yes, Cloud is our front end now, but it could be anything. And I think that's that's the biggest advantage with the s system we're building. so it and I I'm sure there are other companies that have built something very similar. All it is is essentially funneling you know all our software systems. and whether it could be In our case, it's Salesforce Kong right now. But and we have started to bring in our customer support data into it as well. And our idea is to expand our entire, you know, all the software tools we have in our business flows into that database. and we use MongoDB, which is, you know, it could be any data. MongoDB is better for this. and that is now the memory for Claude, right? And Claude then connects it to connects to the MongoDB. And now it has access to all the information that you know that any human can go find in Salesforce. and why that's exciting is first, you know, we are reducing our dependency on just Claude, right? I think if you start to put it just on it now, we could just take in any LLM and plug it to it. So, you know, if there's a big price increase on Claude, we don't have to be we don't have to be worried. But more importantly, it just having the chat interface gives makes people feel like what they've been used to, right? Like they've over the last couple of years, people are used to asking questions through their chat. so people could just open up and type any questions. For example, they could say, Hey, what's happening with this X account? And normally, even if it's a salesperson or RevOps person doing that, And let's an executive wants to understand what's happening with a big account, that information would have to come from at least three different places. We're talking Salesforce, Gong, and and the customer support, in our case, it's the Jira. And getting that, and that's not just three systems, it's probably five people you have to reach out to to get that information. But all of that is now in one place. And you are asking one question and maybe three minutes later you n you have everything you need to know about that account, right? so I think that that's and that's something we built internally, but it wasn't a huge effort. I I would say james: Hmm. Anu Krishnakumar: if you know, if you have a plan, you could probably get that stood up in like a couple of weeks. james: sounds good. So you've talked a lot there about productivity. And I guess the one thing I've learned for a lot of these conversations with rev-ops leaders like yourself is that you guys get, I guess, productivity is one of the key areas, but predictability is also a key area of Anu Krishnakumar: Yeah. james: where you're held responsible for. Where would you say AI is moving the needle most? for you and your business around getting the predictability, forecasting things right or productivity. Anu Krishnakumar: So I I would say it's I it's it's actually so interesting you mentioned that because I kind of talk about that with my team. Like, hey, our our primary goal is increasing the productivity and predictability, but it's a balance, right? It's it's james: Yeah. Anu Krishnakumar: you know, before AI, you try to increase productivity, you will have to open up, you'll have to reduce the controls and the system and the process so people can just run as fast as they can, but that automatically messes up the predictability, right? you're now you have no controls. You don't know what's happening. You don't know who does what. And again, similarly you try to increase the predictivity predictability. Now you're putting in more controls, you're putting in you're asking people to document more, capture at different points. Now you're slowing them down. I think AI finally might be the solution to you know get more out of both of them without compromising the other. and I think with the the productivity stuff, we talked about some of those things, you know, earlier, which is about like, you know, automation is one piece of it, transforming the workflows, you know, offloading some of the responsibilities to AI. And I think that's the place where it massively drives productivity, in my opinion. predictability, on the other hand, is you know, you are you're not compromising the controls are the information that's gathered at each stage, right? You are you're like, okay, you know, some action is being done, but when you're redesigning a workflow with AI, you are integrating the the the data capture or like the information capture at different different different points. So you're not losing it. And because it's a human who doesn't have to do it in many cases, or it can happen in the background. no one feels like they are being taken away from the work, especially a salesperson, the time they are spending with a customer. So you're not losing the predictability. So that's what I'm actually really excited about. And we have already seen a lot of benefit of that. and I think that existed in small ways before with the you know we used to have tools like Clary and for f on the forecasting side and you know gong for the call recordings and things like that. But It was very hard to bring that all together. I feel, I know there was this big concern about the SaaS apocalypse, know, a james: Hmm. Anu Krishnakumar: few months ago even. But I honestly feel like AI kind of just supercharged some of these tools. If I mean I g I'm sure there are other tools where it probably negatively impacted, but on the RevOps side, I feel like many of these things you always had to consider these trade-offs. And AI has finally, you know. reduce the number of trade offs you have to make. james: I really appreciate all your knowledge on that. You've talked a lot around processes and workflows there and probably that kind of takes us nicely into topic two. We're going to talk about an AI workflow that your team actually uses. And when we were planning for this, you spoke about Blanco's intelligence system. So you're building this intelligence system. As a starting point, what problem are you trying to solve? And importantly, why couldn't you solve it before? Anu Krishnakumar: I think that we're building this intelligent system mainly because we wanted to so again th the vision maybe I'll go back to like a year ago, right? we I had this thought if we were if AI can understand you know, learn and understand everything that's in the world, right? It's a smaller problem to solve for it to be able to understand learn and give responses about things that's just happening within the business, right? you know, it's just a smaller set of data to remember, if anything. and I think part of that is like the, you know, one, we didn't know how to quite get to that level of infrastructure. And two, AI was also evolving you know, as as time went on. But and I now it felt like it was the right time to start building something like this. And the reason why We wanted to do it because, like what I started with, we do spend an inordinate amount of time. And it's not just a RevOps team. My goal would be to solve this for, every function, whether it's IT or HR or, anyone else who has an help desk that, that gets a lot of questions, and usually it's the same question multiple times from different people. so it felt like a very natural, it was a very obvious pain point. And it felt like with the AI, it felt like there was a the solution was right there. hey, we went from searching from ten blue links to, directly getting a response. Imagine how much easier it would be if my team doesn't need to answer the question that they have answered a hundred times in just that week. Right. So that's that was the natural evolution of that. But what And and that solves one problem, but I feel like the james: you Anu Krishnakumar: the intelligence we call it the Blanco intelligence is probably silly, but whatever, right? So with james: It works. Anu Krishnakumar: exactly. but it can do a lot more than that. It's not just answering questions. Where we are evolving to is changing the day to day workflow for our sales team, right? We used to for the people who did the research before call, right? And not everyone did because it's a lot of work and, you know, people had it, you know, the in a spectrum, right? People who said, let me just quickly check the last email to, you know, there are people who are spending like an hour before the call doing research on everything that happened with that account so far. or if it's a new customer, you know, the deal's going on. It's a six month deal. You're following up on all of that. That now gets essentially eliminated or gets shrunk to know a 15 minute catch up at best. You know, something that could have been, you know, probably hours of research. Especially imagine a an enterprise sales team that works a deal for a year. The conversation with a customer from the first time to, you know, 365 days later, there's so much happening, so much changing. It's almost impossible for a human to go and, you know, see all those trends. But With AI, you could do that in, minutes. And before that wasn't possible because, you know, AI knew all everything that was trained on everything that's happening in the internet, but not what's in your Salesforce or whatever CRM you're using. But now the memory that we have created, you know, it can go and quickly analyze and, you know, not just give you the research, but you know, we are building skills that starts to align every salesperson to what we would expect a best or an optimum salesperson to behave, right? there was a lot of inconsistencies when you're, you know, people prepare calls different ways. Now you can create a structure, you can put all the information in and, you know, even prepare them with questions they need to ask the customer. And creating that level of consistency was impossible before, right? Because the your best alternative was You had an enablement training where you were giving them all this training and advice and all that. And you hope that they would remember that a month, forget about it, even a week james: Yeah. Anu Krishnakumar: after that training, right? Now this does it every single time just before they get on the call. So I think that kind of initially it started as like you know, an an internal search and a chat engine. And now I think it's more of a an in a just in time enablement and or just in time everything, prep and research, all of that, right. james: How quickly does it evolve into that? People talk about AI developing so quickly, like by day, let alone by week. You're starting quite a big transformation project there, I guess you can say in terms of building this product. Who's looking forward at actually how this develops and becomes more than what it was when it started? Anu Krishnakumar: Yeah. I mean, I think some of that is you know, you you wanna have the right people involved in these things, like people who are genuinely interested, excited about the possibility of the technology, and ideally they have a vision of what this could be. You know, I would like to think, you know, I play an important role in that. I'm genuinely excited about this technology. and I think a lot about like, okay. my team gets excited about something we were thinking of doing and I'm immediately thinking, okay, what else we could be doing with it? where else it should go. but it's not it you need people who are on the ground who are involved in the day to day who are equally excited about it. Right. So I think we have nominated different players, you know, whether it's in sales or marketing or sales engineering, wherever it might be, on, you know, people who just are constantly not just learning about what's available in in the world, but you know, they're excited about like they're they're they're doing something in their day the way they've done it for the last 10, 15, 20 years. And they are thinking, how could this be different? So you need people like that, absolutely. Right. And it can't just be like one person. It's not you know, you have a visionary and then they are because the we are talking about transforming workflows. You need people who are involved in those workflows right now, the way it exists now, to be thinking about what it could be. Right. And I I f I feel like we are fortunate to have some right individuals in that space, in in in in our company. And but yeah, it's a it's definitely a team effort. It's not just, you know, one person leading it. james: I guess the question that everybody's asking right now, which is, it's an almost impossible question, but I'm ask it to you how do you decide which jobs, again, replaced by AI, and which jobs will always need, or for the time being, will Anu Krishnakumar: Yeah. james: need the human element in the middle somewhere? Anu Krishnakumar: well that's yeah obviously yeah no matter what I say, I think it's gonna be it's gonna be pretty polarizing, I'm sure. But it's it's interesting. I I haven't I would say when when all this started, I don't call it like a couple of years ago when this thing initially came out, you know, I was definitely more apprehensive in terms of like, my god, like you know, AI is actually here, you know, none of the jobs are gonna exist, right? But now that we have had more time to process it, right? we don't know how this technology will continue to evolve, right? But as it but everything we have seen, what I see is it's not so much as it's eliminating any roles, it's certainly going to increase the productivity of a lot of the roles. and it should, right? I can't imagine any role that it wouldn't. And then that would reduce the need for too many people in those roles, right? But the expectation is if capitalism works the way it should, people people don't settle for you know less spend and same revenue, right? We all want more as much revenue as we can get. You know, you want the market and the whole pie to grow. So the expectation is now you don't have a a five hundred people company that, generates a hundred million in revenue. You have a five hundred people company that generates two hundred million in revenue, right? And so and then obviously, you know, the com company continues to grow and so does the s the revenue and the and the company. So I don't at least at this point in time, it's almost impossible for me to think that it would eliminate any role because you need ultimately We need ownership, right? You know, james: Mmm. Anu Krishnakumar: you w the companies are being run by humans. You want, you know, maybe a crude expression, you want the a throat to choke. You know, you cannot choke AI's throat, james: Yeah. Yeah. Anu Krishnakumar: right? Like you don't you don't want it to be like, yeah, well, we have eliminated these three roles. Now something went wrong and we don't know who to hold responsible. I think most of the workflows that we are thinking of and we are you know trying to evolve, we're thinking about where humans get placed in that process. You know, it it could be like you're initiating and you know call it an ideal situation towards the end and someone is validating, the human is validating. But what we are seeing in most longer workflows or any impactful workflows is there is a human constantly involved throughout the workflow, right? And yes, it massively improves productivity. you're not, think of a difference between, writing an email and reviewing it and having someone else review it to an email is written for you and you make some quick edits and then you're ready to send, right? That's obviously a very simple example, but you I cannot imagine a situation where we are going to just have AI completely write and again, simpler example, completely write an email and send it and fall manage responses without a human looking over it. Because again, if something goes wrong, who is who is responsible for it? I think the accountability factor never goes away, right? So I think that's why I can always imagine humans being involved in it, but I can certainly see the number of people needed to do that being dramatically reduced over time, right? james: I guess the noise from LinkedIn is Anu Krishnakumar: Mm-hmm. james: what kind of drives those conversations. Anu Krishnakumar: Right. james: And there's one point that I wanted to bring up. My LinkedIn feed is constantly filled that you don't need SDRs, BDRs, the replacement. And what's Anu Krishnakumar: Right. Right. james: fascinating, talking to all these leaders for this show, Rev Ops leaders, Anu Krishnakumar: Yeah. james: is it's not something that they're seeing as possible. And then also, I hosted a roundtable with some incredible CROs last week, 10 of them. Anu Krishnakumar: Yeah. james: Five of them are in growth mode. And what are they growing? SDR teams. They're Anu Krishnakumar: Yeah. Yeah. james: actively hiring. But if you look at my LinkedIn feed, it's you don't need STRs or BDRs anymore. So yeah, look, it's Anu Krishnakumar: Yeah. Right. That james: fascinating, like the real world against the social media world of what it's saying. Good? Yeah. Anu Krishnakumar: Is it okay if I quickly comment on it? That's a very valid point because I I have seen it and sometimes we have at least one time we spoke to one of them, right? Because we were fascinated. We were like, this used to be done by five STRs. Now it's done automatically. And we were like, okay, let's talk to this person. We want to know what it is. And I wouldn't say it's they were lying, but the impact of it, of what they are seeing, it's like it It's at least massaged information. the the reality is you can, you can absolutely automate anything and just let it run. You're now probably hiring a lot of engineering people to constantly look at it and see what's wrong. james: Yeah. Anu Krishnakumar: And now they don't know how to write an email. Now you need to bring in an SDR or a salesperson, right? And I have seen you can tell like when an email comes to you and you're like, okay, this is definitely written by AI. And now that reduces the chances of you responding to that email. Right. We are not in the like I said, we're not in the business to s reduce the cost as much as possible. We are in the business to make money, right? And I think with the AI, unfortunately, at this point, it reduces the chances of you making money, right? While it james: Yeah. Anu Krishnakumar: reduces the cost. So james: Yeah. And so before we move on to the third topic, around AI, if time saved isn't the right way or wasn't the way that we were going to measure AI's impact, outside of that, what should people be looking at to measure its impact? Anu Krishnakumar: Yeah, I think time saved, you know, should be a given. but, you know, the the problem is w what are you doing with that time? And that's where it gets tricky, you know. people playing golf and I don't know, watching TV, whatever, right? But I I think the better way to think about I mean the way we are thinking about the impact of AI and and the benefits of it would be how the workflows are being transformed, and time saved is being is a part of it. it's more about, okay, how would how was something being done before? How is something being done now? And the impact of that, right? And the impact of that could be speed, which is time saved, could be the effectiveness of it. Hey, before we used to have, you know, X number of errors on an average. And james: Woof, woof, woof, woof, woof, Anu Krishnakumar: you know, now it's james: woof. Anu Krishnakumar: but no so you know the the idea is like okay what is the impact of the workflow, right? time saved, effectiveness, reduction in errors, and or, you know, sometimes it is like, you know, probably something you do want to eliminate. It's not no one wanted to do it, but you could now have something do it. It's just like a, you know, hey, this is a nuisance work nobody ever wanted to deal with, right? Think about cleaning up your desktop or your file system. It's n people never had to get wanted to get it done, but now you could have AI do with it. You know, I'm again, that is not a call it a business example. james: Yeah. Anu Krishnakumar: But there are opportunities in business we are finding every day where we're like, my God, nobody wanted to deal with it. It's like it was piling up, but now you can solve it. So that's why I think time saved alone is not an in you know interesting enough metric. I think the workflow is transformed is how I would like to think about it. james: Topic three, we're looking at the gap between using AI for forecasting and letting it run workflows totally autonomously. So if there's a team that let an AI predict revenue and run largely on its own, in your view, if they were doing this, where's the gap between what AI forecasting promises and where the teams really are? Anu Krishnakumar: Yeah, I I think the there's a there's a spectrum, right? And all of that is completely, doable now with varying levels of effectiveness. at the the beginning you have AI can help with the the analytical side of forecasting, taking the information that's coming from you know, the the managers, the roll ups or the information that's available in your systems and analyzing it. And then somewhere in the middle is like it's actually coordinating the process of forecasting, right? Whether it is like engaging different leaders, getting their input and consolidating it and and providing it as a know, rolling it up, I guess. And you know, and you know, the more theoretical, you know, maybe in the far future would be where it entirely runs the forecasting process, right? I'll I'll give my commentary on how I would feel about that. But james: Yeah. Anu Krishnakumar: the where we are is on the analytical spectrum, right? I think we are starting to engage in more of the the call it the coordination aspect of it using AI, but before AI, we have still had what we used to call a projected forecast, right? Which would be a very mathematical forecast that just considers a lot of quantitative attributes, right? Stage, age, how big an op is, all of those things. and then we would produce a number. And, you know, there were always tools in the market like Clary, which would factor in other things. They had machine learning to do that. and but now with AI, you don't necessarily need a tool like that. You have this quantitative information. It can go and analyze a lot of those unstructured signals like conversations with customers and emails back and forth and give you the sentiment analysis and you can put a number to it. And then now you can produce a what I would call a calculated or mathematical forecast that goes beyond just those quantitative metrics and you can you can you know you can be a little bit more confident that it, you know, other things that are factored. but that doesn't remove the Sales manager roll up. And that goes back to that countability piece that I mentioned, right? This is going to exist, but we still want the managers to be responsible for their forecast, for the numbers they are putting out. It can't just be like, it's an AI forecast, and everyone's like, well, I didn't know the AI forecasted it. I had no, you know, you didn't ask me for it, right? and that's and so this place I'm comfortable with, and we could potentially get to a point where we can reduce the number of reminders and emails and things like that that has to go out to roll up a forecast. And I think AI can definitely help there. but, you know, letting it I know people are talking about like again in social media about the whole forecasting process can be fully automated. Can it be sure, but do I want it to be? I don't think so. I mean I feel like I would always want in human a sales leader being accountable for the forecast, right? You because who do you hold accountable when the when when we don't hit a forecast, right? or for that matter, we exceed a forecast, right? And I think and there is also the element of even when engaging with the forecast conversation, there is a lot of human dynamic involved, right? Like you know the personalities involved, you know, someone's conservative, someone's aggressive, and, you know How do they react when when it when a deal is at risk? I think and challenging those opinions. Can I do that? Possibly, but I would feel a whole lot comfortable when it is you know a sales leader calling out their team and saying that, like, hey, you know, explain to me how that's going to happen. Right. Now again, your manager holding you accountable is very different than an AI holding james: you Anu Krishnakumar: you accountable, right? so I think that's why I think I can see value of AI, but I can never I can never see AI fully running a forecast, right? james: So you believe, just to make sure I understand right, that the running of the forecast, the accountability of making sure that the reps are doing it is where the human element comes into it. Do you think the human element comes into creating the forecast at all? Or do you think that's okay to be run totally through AI? Anu Krishnakumar: w when you say creating you mean just the idea of consolidating the numbers and rolling it up or james: Yeah, yeah, so it's like putting it into really basic forms. Is it okay to tap a couple of numbers on your keyboard and there's your forecast and then your role is just to be accountable to make sure that number's hit? Or should there be a human element played in between the taps on the keyboard? Anu Krishnakumar: a hundred percent that has to be human human again, right? So the I am big on accountability, right? Especially, you know, the the the degree of importance matters here, right? and I think forecast is like comes goes back to the predictability piece. It's one of the crucial parts that RevOps play or sales theaters play for that matter. I am I'm not comfortable with, you know, offloading that to AI. Right. because if if it's a couple of okay, my team is, you know, let's say a sales leader believes that team is so tight on how they manage an opportunity that they're always kept in the right stage, you know, they always put in the right inputs. Even then, if an AI produces number, it removes the air, you know, salesperson, sales leader from that you know, from reduces accountability for the sales leader. Right. Now it's AI number, it's not their number. Right. even if they've tapped the number you know, some keys. and to hold them accountable, you want them to be closer to that number. You want them to they can certainly and AI is massively helping with this, you know, giving them the insights to produce that number, but it's still they are making the judgment. And I think that's critical. And I would ideally not if that goes away. Now, the chances of us holding a sales leader accountable for the forecast also goes away with it. Right. And I wouldn't want james: Totally agree. Anu Krishnakumar: that. james: Yeah. I think that's the bit. To be accountable for something, you have to be really believing, be part of it rather than be giving it. So I totally agree with you on that part, just for the natural human element of anything. Anu Krishnakumar: Yeah. Right. james: And I think the difference between winning and losing has accountability involved in it in quite a large amount. Last topic before I let you get back to your day job. Anu Krishnakumar: Mm-hmm. james: Talking about whether alignment is something AI tooling can help with and where it lies and talking about the human elements of things. So do you believe AI tooling generally can help align across sales, support, product? Or is that still in today's world? And obviously things are changing at the moment. Is that still fundamentally a human job or can AI do that? Anu Krishnakumar: Well, I guess that depends on where the misalignment comes from, right? In some ways. if the if the alignment issues arise from you know, purely, you know, lack of information or the right information or the shared knowledge, call it, context, or just people remembering thing things incorrectly. air can absolutely solve that piece, right? but I'll I don't think the alignment issues are just because of that. I'll get to that in a second. But and for the information element, it can absolutely solve. And that's that goes back to the you know what we are building internally, right? Like your your you know the data lake and the dashboards were, you know, the first attempt at solving it. Hey, if everyone has that access to you know, shared data and everyone is working from the same information, you will reduce the alignment issues. But now it applies to everything. It could be like a customer conversation, it could be anything. You can you can all have access to the same information. So that reduces the alignment issues coming from people just misremembering or not having that context. But I believe that the misalignment happens for more than just that reason. It usually happens because people have different incentives or different motivations, or even risk tolerances, right? Like there's just the you know, we are all different people, we all have different diff think differently about the same thing, even, right? And that you cannot solve with AI, right? It can It can make it clear that, hey, everyone's looking at the same information. This is clearly a perspective issue, right? But to get people to align on a perspective, align on an idea, you do need that human element, right? Coming to an understanding is uniquely human, just because, yeah, okay, we have eliminated misinformation. Okay. Is that really going to solve the problem? I could, there are I I see that you you have a dog, right, you know, james: Yep. Anu Krishnakumar: I love dogs and my wife would never let a a dog inside our house, right? And james: Yeah. Anu Krishnakumar: I could I could I could spend an entire day trying to explain to her why we should have a dog. I'm never gonna win that argument, right? james: Yeah. Anu Krishnakumar: But it's so it's not about accurate information alone, right? But I think AI can definitely simplify some of those problems. But I do believe it's gonna be a human there's gonna be a big human element there. james: And so when you're talking about that, you spoke earlier about all your data being in one place. Has that then changed how the conversation between those teams are taking place now and how they're interacting with each other? Just to give the audience an idea of how that has aligned everybody or changed how they were before. Anu Krishnakumar: Yeah, it reduces the amount of time People have to spend getting that information, right? And it's again, and there is less time being spent on is the information accurate, right? I mean, well, actually I wouldn't say it has completely gone away. Whenever an information is not aligned with what you are thinking, it's like, okay, are we sure this is right? That question still comes up, but you know, it's a lot less than it used to be, right? Because before, you know. two people pull information from different places, whatever it is, and then now there is a disconnect. And you know, you do f spend a fair amount of time trying to debate the accuracy of the information. That has gone away. and we have moved on to like, okay, now let's think about what the action is and agree on the next steps. Right. I think so it has definitely made it more effective or efficient. but, you know, I can't see it eliminating That friction completely. james: I really enjoyed this conversation. I really appreciate you taking all the time out for planning this session and for sharing your knowledge and insight. Before I let you go, if there's a rev ops or CRO, which hopefully there will be plenty of them listening to this episode, what's one thing they should take away from what you've mentioned today and try in the next quarter? or just try and implement and do. Anu Krishnakumar: well, I would strongly support everyone consider like the kind of you know, the infrastructure that I mentioned that we're building. And it's not unique to us. We are not the first ones to have done it. and I don't think that one's that one's fake news from from social media people should attempt it. And I think it'll it'll be transformational. But that is the underlying work, but I think what I would like everyone to start exploring, no matter where they are in that AI belief spectrum, whether they are on the pessimistic side or the optimistic side, is to it's here, it's here to stay, and we should think about how we can leverage it and how we can transform our workflows. Move away from automation, think, you know, stop thinking like you're gonna eliminate entire sections of the role, but just thinking about how do you enable people to leverage it and transform their workflows. james: And in your opinion, do you think whether it's a rev ops and I think rev ops will always be optimistic, but from a CRO perspective, do you think you get more out of or will get more out of AI if they're pessimistic or optimistic? Anu Krishnakumar: absolutely. I think it's being optimistic with anything, right? just I think you were mentioning that the accountability or when how people feel about something has an impact on the results, right? I think james: Yeah. Anu Krishnakumar: being optimistic definitely helps, right? but I think you could be pessimistic but still accept the reality, right? you could, hey, this is where it is, and you want to, think of it as a tool, right? and try to move away from the from, you know, however it's being pictured in in in the world. And in its current state, it is it is a tool that's available to us. And as leaders, it's our responsibility to figure out how best to use it to our advantage. Right. And if you're not, then someone else is going to have that advantage. Right, whether it's another company, business, country, whatever it is, and we're gonna lose the edge. So that's how I think about it. james: Love it. Well, Anu, thank you so much for your time, energy, and of course, most importantly, your knowledge and insight. Really enjoyed talking to you and spending time learning so much about rev-ops. Anu Krishnakumar: Thank you so much, James. It was great chatting.