Future of Revenue Teams in an AI-Native World | Sreedhar Peddineni | S1E16

Utsav Bhatt: Welcome to Alt Consulting where we do AI adoption conversations. I am Utsav Bhat, founder of Startoff. We help companies drive AI adoption and innovation-led growth. Today's conversation sits at the intersection of AI, revenue growth, and organizational transformation.

My guest today is Sridhar. He is CEO and co-founder of GTM Buddy. Sridhar has spent much of his career creating entirely new software categories. He previously co-founded which helped define the customer success category ⁓ ultimately grew the business and achieved a billion dollar plus exit.

⁓ that, he co-founded Host Analytics, which is now called Planful, which helped pioneer the Sreedhar Peddineni: and conversation. Utsav Bhatt: Cloud EPM category. Today with GTM Buddy, he's focused on what he calls revenue activation, a category that aims to fundamentally change how revenue teams operate in an AI-driven world. What makes this conversation particularly interesting is that most organizations are currently experimenting with AI through copilots, chat interfaces, and productivity tools, but very few are rethinking how revenue itself gets generated, supported, and scaled using AI.

So that's what Sreedhar Peddineni: you ⁓ Utsav Bhatt: we are going to explore today. Sredar, welcome to the podcast. Sreedhar Peddineni: Thank you, so glad to be here. Utsav Bhatt: So before we get into AI, I'd like to first ⁓ sort of start with the underlying business problem.

⁓ organizations, as we know, have invested heavily in CRM systems, ⁓ enablement platforms, content repositories, ⁓ systems, training programs. ⁓ if you speak to sales leaders, many would say that their teams are overwhelmed with the information that is thrown at them. They are multiple systems, it's all fragmented, and execution remains quite ⁓ Sreedhar Peddineni: . Utsav Bhatt: So from your perspective, what is fundamentally broken in how revenue teams operate today?

Sreedhar Peddineni: Great question. So to set some historical context about how the revenue teams have operated for the longest time. It's about, okay, I'm running a revenue organization, which includes sales and customer success, pre-sales and all of the teams. And part of it is about how do you execute on the revenue motions, but there's a supporting function that emerged over the past 15 years or so.

in a significant way and that function is called enablement. It started with sales enablement and sales enablement basically stood for, okay, I'm hiring reps, how do I train them? So you're typically talking about learning management systems, onboarding, create onboarding courses, create periodic training courses and certify people and so on. So there are a number of L and D focused enablement professionals who focus on that.

Then after reps are trained, and certified. We hope that they remember all the things that they have learned and we give them access to a content repository. So there's a sales content sales focused content management systems that have emerged in the past 15 years and companies are spending well over a billion dollars in this content management and sales learning management systems. I'm just talking about software.

If you add the people cause you're talking about tens of billions of dollars being spent on annual basis on enabling the reps. Okay. With a focus on, okay, I'll train people. I'll give them access to content and I'll give them a pipe or if they're full cycle is that building the pipe and hopefully they'll convert better with armed with these two pieces of information.

This is a very traditional model of enablement. And ⁓ with the times were good, money was cheap, all was good, right? But ⁓ As the economy is transitioning, the growth at all cost mindset is no more real. And the boards are demanding that do more with less.

That's today's norm. And no more the investors do not look at company headcount as a measure of company success. On the contrary, they saying that why do you have so many people? Okay.

So people are expecting, okay, you should have ⁓ hit $10 million in ARR with 10 people or less. And they will cite the examples of the lava Wilson of the world and so on. Right. So the world has dramatically changed and this added a lot of pressure on functions that are classically considered as support functions.

And enablement is one of them. And the function was hit hard. There were a number of really respected, ⁓ revenue enablement professionals who were laid off. And this is something that's a big topic in the enablement community where they're good to have and not a must have.

Right. Now, part of the problem when we reflect on it, there are external macroeconomic drivers, the rise of AI, that's one part of it. The second part of it is about, ⁓ how do you convert something that is perceived as a good to have to actually driving revenue? The connection between the good to have initiatives like training and content hygiene and stuff like that, how is it actually connecting to the revenue?

And we see that as the... the emergence of we see the enablement is transforming into revenue activation in terms of how do I activate all this assets that I've created over time and how do you get that into reps workflow as they're executing on a deal. How do we get them to adopt the right information at the right time and in their workflow without context switches. We call that revenue activation.

Now, An example, making it concrete. So ⁓ in sales, it's very common that the very first call that a rep takes, we call that a discovery call. And it's a near universal requirement from most of the customers that we work with. We want to improve the quality of the discovery.

Right. So it's possible that ⁓ reps can qualify a bad deal without sufficient information. which means that I will keep working on on a deal that did not have the right ingredients to begin with, or I could be disqualifying too much. I could be disqualifying a good deal.

Right now, the companies provide training on what's running an effective discovery call. They provide training and effective discovery call includes, okay. I'm meeting with a prospect and ⁓ it was an inbound prospect with a large company. and my product I sell it to, let's say a cross-functional buying team.

Today we're not selling to a single buyer anymore. It's a buying team, almost always. And the buying team comprises of someone from revenue operations team, from enablement team, another from sales team. And I need to take all of them together with me on the buying journey.

Let's say I have gotten an inbound lead and the person who submitted the form is a revenue operations leader. Okay. Now, if I'm a good rep, what am I supposed to be doing? Okay.

At least spend some time looking up their website, understand what the company does and think about, okay. And then do the person's person research. Look at what was the person's profile, where the person has worked and so on some background about the person that helps you build the business. The third thing that I need to remember is about, okay, the for revenue operations persona.

What are their pain points? It's not about my feature that I want to sell, my product and features that I want to sell. It's about them. What is the pain point that persona has in that particular vertical?

Right. So this is again part of training. We provide persona training, there are persona decks that get created. That's a very elaborate process that's conducted by product marketing teams.

Okay. Now, rep is supposed to remember that information. And okay, for this person, this is what I suppose I'm supposed to do. Then the fourth thing that I have is okay.

I'm running a discovery call and we have a homegrown sales methodology. And as per the discovery call, I'm supposed to ask this five questions that are really important for us. Okay. And I have to do it call after call.

already and I'm only scratching the surface here. You're talking about information overload. And we expect the reps to be doing all of this in their heads. And great reps do it.

But is it easy? It isn't. Right? A revenue activation ⁓ with AI, what it means is that, okay, we can have a very easily, we can create an agentic job.

You could use GTM buddy for this or you could create a cloud or charge a PT agent or whatever. Feed in an account and you give you a research parameters in terms of what areas of the company that you want to research. can give the person information. If the person has social presence, you can extract some information about the person and you feed in all of the content that you have about the buyer persona.

You do all of this. You can build this as a AI workflow. And some. AI native reps are already doing that.

Okay. What platforms like GTM Buddy, what we do is we create this into an automated workflow where before the person is getting on a call, the seller gets a digest. This is what you need to know going into the call. Okay.

This is just one example. Then what happens during the call? What should happen after the call? Follow is critical.

A good follow up A great follow-up is ⁓ a concrete deal mover. Now, how can you leverage AI, especially in the world where a rep is jumping from one call to another? There's not enough time in between sometimes. So how do you be deliberate in terms of, okay, based on the conversation, based on the pain points that I was able to uncover, what is the good content that I want to share?

What's the personalized experience that I can share? Can I create a presentation that... leverages the conversation, not just a generic, ⁓ this is my company deck. This is my product deck versus something that is personalized.

Can I create that content with AI? It is possible earlier. You need to have a lot of PowerPoint or Google slide shops and, sorry about saying this, but a lot of people do not really know how to use templates. Well, you see here.

All the fonts are all over picking the wrong designs and all of that. Every company has a slide master. Not everybody is consistent in terms of following the template. But with AI, you can automate that stuff.

the other thing that's coming in, sorry about the long-winded answer here. This requires, you know, certain level of extensibility that we have always taken for granted in SaaS. Okay, a SaaS platform is something that you don't write custom code for every tenant. which means that your platform will build SaaS products in a way that they're extensible and configurable for different customers.

I'm not writing custom code for everybody. If I say, this is my way of doing it, there's a single configuration and it works for everybody, that's theory. What does extensibility mean in the world of AI is becoming a new question. It's a new frontier for B2B applications on how do I take, provide, You know, level one is prompting.

What are the system prompts? Do you want to go and change them? The next level on top of it is the skills framework. Right.

Is there a is can you customize the skills that you want for creating your presentations, for example, or building a business case? Right. A lot of these things are changing. The last thing that I would say is that from an administrator perspective.

Again, the pressure of, you know, there are, I know of people, a single revenue enabler supporting a GTM team of 200 people, one person supporting 200 people. How does a person do it without leveraging AI? So all of these use cases that we are talking about, they are not addressed in the traditional way of enablement. It does require us to unlearn and relearn.

It's not just about the technology. It's about cultural blocks that, you know, not everybody learns by doing and with things changing so fast, it's very easy to get overwhelmed. I'm overwhelmed with the pace of change. Right.

So I would not. I would sympathize with anybody who feels that sense of vulnerable. How do you deal with it? Yeah.

Utsav Bhatt: No, I think ⁓ you covered a lot of ⁓ points and it reminded me ⁓ you know, for for a decade I was doing ⁓ IT consulting, digital consulting, and was part of several large deals. And in the large deals I was playing the roles of say role of sales enablement, and I could figure out how difficult it is to talk to like five different ⁓ you know, sort of horizontal service units and a couple of verticals, talk to different stakeholders, get everything together at one place.

And it's again it it's it's just not skill. ⁓ skill is an important Aspect, but what actually fails is if you find certain people who are really good at that, it's very difficult to replace them because you need to have so many different skills to first understand the organization deeply, understand the way in which people work in that organization to that extent, and after that, when it when a deal comes, look at that deal, figure out what needs to get done, and then work almost like a machine.

So you have to run a system in the in your mind to ensure. Sure, you're able to get out that deal on time and stitch it all together. AI is a is a very good way to ⁓ make that so that you can consistently keep on delivering ⁓ at a a level which is required for cracking those deals ⁓ and it's left to people in in terms of their individual ⁓ understanding of company, their own ⁓ nuances in terms of how they handle such deals, ⁓ it becomes more coherent and you can you know ⁓ sort of drive ⁓ revenue growth via that.

So we we covered a lot of ground. Here, in terms of if I can quickly capture what's broken today, ⁓ we we then you sort of explain the term of revenue activation, what it means, and how it is beyond sales enablement, which was which is always looked as a support function. We talked about an example, and then I think one point which I want to ⁓ talk about is several AI tools which people are using right now ⁓ essentially have a simple Google like chat window. You write something and ⁓ output comes out, and then Sreedhar Peddineni: .

Utsav Bhatt: Is so much of information to process there. And your view and GTM Buddy's view is quite opposite to that. You don't want an interface where again a sales rep comes and starts writing something and gets insights. You actually want that intelligence to be delivered directly, ⁓ you know, through the flow of work, if I can use that analogy.

So tell me what was the point where you realized, hey, we don't need another simple ⁓ chat window, we need to build it into workflow. And the reason I'm asking that is Sreedhar Peddineni: Yep. Utsav Bhatt: Revenue activation even more so requires requires that function. If you think like you know logically, you will have a lot of information.

So you need a wrapper on top of it to get insights. But you sort of designed your product in a different way. So please walk us through what what made ⁓ you take that call. Sreedhar Peddineni: It is.

Yeah, we also have a chat window. So don't get me wrong on that. I have nothing against the chat window. But if you have a 500 people revenue organization and you give them all access to a chat window, what are the assumptions that we are making here?

We're expecting that all 500 of them will go to that chat window. And step two is that they're asking really good questions. Okay. And it is generating a lot of content, right?

Unless you have this, this generation generating a, a driven content is made. You can get, create good quality content, lots of it in minutes. So how do you separate wheat from the chaff? What's really the meat of it?

Are we expecting that all of these 500 people will be able to use it equally effectively? That that's anybody's guess. And we probably know the answer to it. Utsav Bhatt: Hm.

In in fact that's why ⁓ ⁓ in in technology we always had big applications and there was something called as best in class. ⁓ Like this is best in implementation Of SAP, this is best in class implementation of Oracle. And the experts and consultants used to come and teach what is that best in class. Similarly, when you talk about how you use AI embedded in a workflow, if there are great revenue activation champions who know what great looks like, what are good questions, and you built it as a feature in the product rather than leaving it to people to ask, you might be actually ⁓ and sort of getting to that training aspect of let me now train your employer.

employees or how to use my system, right? Because that's where AI adoption also fails sometimes. So that's actually quite a smart move and ⁓ a way potentially how these products could be designed going forward. Sreedhar Peddineni: Yes.

Yeah, absolutely. So the prompting expertise, it's a life skill, but probably we have moved beyond the world of prompts now, as you know, I'm preaching to the choir here. prompts are now almost replaced by skills. In terms of any work that requires any level of sophistication.

Utsav Bhatt: Yes. Right. Sreedhar Peddineni: you're talking about skills and you're talking about synthesis of data, not from one sources like, uh, okay, go on, help me write this email or fine tune this email or, uh, ask generic questions based on the internet. It fetches the information.

It says these are like, uh, important, useful, uh, know, capabilities of AI. But when you're talking about a business context, you're invariably talking about how are you, how am I synthesizing all of this data together? We are getting into a world if I am working in cloud in that same chat window, the chat window can be cloud. For example, I need to know how to figure out, figure out what connectors I should be using, what MCPs I'm connecting to.

That's becoming a life skill. If I'm going the route of build versus buy and with these systems, people can build stuff. It's not necessary that you're buying GTM body. You could be building it.

Utsav Bhatt: Yeah. Sreedhar Peddineni: It costs its own money and if it is for you, if you think building is the right thing for you, go for it. But you need to train the 500 people on how to use your connectors. What is an MCP?

What are the tools available? And how do you construct a skill that synthesizes data from three or four different sources that you have, oftentimes more than that? You're looking at your CRM data, you're looking at buyer engagement data on your digital sales room. looking at your call transcripts is like a minimum subset of information and you're adding the web research.

Already I talked about a very simple use case that's pretty significant in terms of you know combining all of that data. So that is a skill unless people are we are training people to that level of data synthesis it's going to be Utsav Bhatt: Yeah, I I've been part of some AI adoption projects where ⁓ the first step which people talk about is hey, let's have a lot of ⁓ training sessions on on Claude, on Chat GPT, on Copilot. And and you're like, Yes, but there have been training sessions on several skills for a long time.

People know Excel for a long time, but how many people know how to make a macro? ⁓ so i i it's just that ⁓ you know, everyone has a different level of understanding of the tool, usage of the tool. So ⁓ I think going to the skills ⁓ route is is much more relevant in the AI world than just keep on learning prompting. I think we are beyond that.

But but that makes me actually move to the to an important point. Great, we have a great tool, we have an understanding of how AI can be used for revenue activation ⁓ But whenever you introduce such an AI system in your ⁓ it's not easy. The technology ⁓ looks very difficult, but ⁓ in the larger scheme of things, it's usually the easy The harder part is how you Make people to change their way of working because one thing which I keep on repeating endlessly is it's not that people are you know resisting AI, it's just that they are protecting how they know to succeed.

So if you teach them a new way to succeed, they will absolutely you know it's a common sense for them to use that ⁓ product or tool. So, in your experience of implementing AI system, ⁓ what are the big barriers which you have faced? And I'll break it down into two pieces. One is more structural.

Sreedhar Peddineni: Even. Utsav Bhatt: Barriers like incentives, metrics, reporting lines, how you govern a company, performance management of employees. The other is cultural, which is habits, perception, fear, ego, you know, identity clashes which people have and a reluctance to learn new of way of working because they are fixated in their traditional ⁓ methods. So in your experience, what are some real bockers to AI adoption inside you know the kind of projects you are doing?

Sreedhar Peddineni: So systemically, what we see is that ⁓ with a lot of our customers and prospects that we talk to and ⁓ one common theme is that from the leadership, there is a mandate from the board. There is a mandate that you should become an idea. OK, now a lot of people are scratching their head in terms of what does that really mean? What areas are we leveraging AI for?

What's the priority order? Because you're talking about all functions have to become a native is the mandate. Now, how do companies do that? Okay.

Let's give cloud license to everybody. Anthropic seems to be making some smart moves. Let's give access to cloud to everybody. And a lot of people that we talked to, they use cloud chat.

A lot of people have not even installed the cloud desktop and without installing the desktop, you don't get co-op. And if you're trying to solve, forget cloud code, that's for somewhat of geeks who really want to get to that. So I'm talking about the most common usage is that a lot of companies have given access to plot, especially in the last few months. And 90 % of them don't use cloud desktop.

What does that tell us? You're not really using the biggest, most significant development that cloud or Anthropic has brought to the market. namely cowork, which in turn is a layer that's built on Cloud Code. Its ability to execute on long running jobs, manage context and accessing data.

are things that sets it apart. So now you have this mandate of AI transformation. I gave you access to Cloud, go build it. People are struggling with that.

along the way, in every company, there are some really geeks, if you will, who are really interested in AI and there are a lot of projects that are being run. So ⁓ in some companies, the challenge that we see is this extreme preference towards build. That's because there are a few people ⁓ a core group of three four five people come together Okay, they are enormous by technology in terms of what's possible with AI These are typically GTM people now have never written code before but they figure that I can wipe code stuff Utsav Bhatt: Yeah.

Sreedhar Peddineni: and they are into their thing. And so they become like you're talking about tribal wisdom, right? Wisdom that's ⁓ in people's head. It's not like that group is aware of the entire organizational business processes and what are the best areas of investment.

So there you start seeing that, okay, there is this specialized packets and those are the people that you go to for everything. So there is no method to the madness. That's the systemic problem that we see from a people perspective. It's about people are excited and scared by AI.

And honestly, I am as well. I am at the same time, excited as well as scared by AI in terms of where are we headed in five years from now or 10 years from now. I don't know the answer to it. But what I am betting on is like any which way I look at it, I might as well learn this as much as I can.

Utsav Bhatt: Yeah. Sreedhar Peddineni: not worry about where it's going to go, which I don't control. Culturally, the number of people are probably discounting the level of learning, the investment of self investment of time that is needed. I could bring in consultants to train, but by the time they provide training, it's outdated.

The pace at which things are changing. So the only. Utsav Bhatt: Yeah. No, I I think I think two points that you mentioned ⁓ quite resonate with me.

I think the point on the structural blocker ⁓ around people have this habit of building on their own. I just had a recording ⁓ with ⁓ John Willis who is one of the f founding sort of personalities in the DevOps movement, fifty years of experience and we discussed a lot around how ⁓ twenty, twenty five years back, ⁓ K Loc ⁓ like how many thousand lines of code was a metrics that people used to measure and compare. Now Token usage. So if you build a lot and you start saying, Hey, yeah, my token usage is going up, it might be just bad usage of because you have built a bad system, it doesn't do much.

⁓ so so that is something which you know in next maybe four to five years, somebody would write a paper and say, you know, we are measuring it all wrong. You know, you should measure it in a different way. So that's on the system side. On the cultural side, I I couldn't agree more with you.

I think people who are building companies around ⁓ AI are are really scared because they are very close. Sreedhar Peddineni: Mm-hmm. Mm-hmm. Utsav Bhatt: close to how it can actually change your work.

Myself, like you know, three years back when I started looking into building a new consulting firm, I ended up writing a whole book on future of consulting because I want to be very clear what's going on as a consultant. Just think about all possible options. But it it is writing on the wall. You know, you have to do something about it.

So why not just go and build something? So that's that's it. Sreedhar Peddineni: That reminds me about your book also. I meant to purchase it.

Can you tell me the title again? Utsav Bhatt: It's alt consulting. ⁓ A G alt consulting. Yeah.

Sure, sure. Yeah, you'll you'll get it there. Sreedhar Peddineni: or consulting. Yeah, yeah, yeah.

I'll find it in on Kindle, right? OK, awesome. Utsav Bhatt: ⁓ I'd like to close with a question that I ask many on the show, which is a lot of CEOs, and in your case maybe chief revenue officers, are currently under pressure to accelerate AI adoption. You know, everyone is investing, everyone is experimenting, everyone feels the urgency, there's a board mandate, as you said.

So if you were advising a CEO ⁓ today, what is the one AI adoption mistake you would tell them to completely avoid? Sreedhar Peddineni: I would say whenever such a massive shift is happening, the world always moves in a pendulum in terms of software tech stack. There is a consolidation wave and there is a fragmentation wave, you know. So at times there is everybody wants to consolidate everything.

Then you start seeing innovation from point solution vendors is fragmentation. Right. So if you are a CEO, you're probably getting pitched by a hundred different vendors offering the moon. with with AI.

My number one suggestion would be to, it might be better to take the build approach if you are willing to invest the time, money on people. But avoid adding to the tech proliferation that already exists in sales tech, especially sales and marketing tech. And be judicious in terms of picking platforms as opposed to point solutions. A platform that's built to be a native, Look for more of platforms opposed point solutions.

That would ⁓ my number one advice the CEO. Utsav Bhatt: Great. ⁓ so thank you for joining me today. This has been a fascinating conversation.

We covered a lot of ground and for I think listeners, there might be a whole new area which ⁓ they might have now being more aware of in terms of revenue activation and how it can be completely be transformed using AI. So thank you for everyone listening to Alt Consulting AI adoption conversations. If you enjoyed this episode, please subscribe, share it with your network, and join us next time as we continue exploring what it really takes to drive AI adoption inside organizations.

Thank you. She once more for joining me. Sreedhar Peddineni: Thanks a lot for having me, what's up.