Introduction

Alex Kracov: What's up everybody? I'm incredibly excited to welcome my friend Dini Mehta to the podcast. Dini was my partner in crime at Lattice. She was the CRO while I was the VP of marketing, and she helped grow Lattice to over $100 million in ARR. And now she's the operator in residence at a venture capital firm, Peak XV. We scaled together during the SaaS boom, and so I'm really excited today to get her perspective on how go-to-market works in the AI era. Should be a fun, kind of casual conversation as we pick her brain about everything, the craziness that's going on with AI. So welcome, Dini. Excited to chat.

Dini Mehta: Thank you, thank you. Always fun to jam with you. It just feels like we're working together again when we do these sessions. So pumped. Let's do it. Let's get into all of it. All the craziness.

How AI-native founders are building go-to-market

Alex Kracov: Nice. Let's do it. Okay, so you've spent since Lattice like the last two years at Peak XV, working with a lot of early stage startups. And I'm curious, like, how are the founders today you're working with building go-to-market differently in the age of AI versus like the playbook that you ran at Lattice?

Dini Mehta: Yeah, I mean, I think there's some stuff that continues to still be the stuff that hasn't changed, which is, you know, understanding a customer's business. How do you build trust through your people and the brand that you build, running great sales process and truly understanding, like where is the organizational chaos and how do you sort of unite a customer's org around a business problem and then proving value through your product? I think those things are still unchanged.

What is different is go-to-market is increasingly like being treated as almost like a design and systems problem versus like a people problem. I think in the past it was like, oh my God, it's like hire a bunch of reps, go get revenue. Now it's more like, you know, AI can reduce a lot of the cost around research and enablement and operational work, which there's a lot of. And so I think that is the difference — teams are thinking about where can I leverage AI and sort of make hiring or adding people the last lever in the business. Whereas I think in the past, in hypergrowth, it was very much like, you know, hire as many people as you can as fast as you can to try to get to the next milestone.

Alex Kracov: Totally. I mean, I remember like the board conversations at Lattice. We were always like, look at the spreadsheet. And it's like, okay, we need to hit this revenue target. And hire, hire, hire. Right? Like that's how you get there. And it's like, yeah, it makes sense on a spreadsheet, harder to do it in practice.

Dini Mehta: Totally. It's so easy. But now that I'm on the other side, I find myself doing it. I was like, you can go faster, what are you doing? And then I remind myself, I'm like, it is so much easier to be on the other side saying that. I also have more empathy for board members now, which maybe when I—

Alex Kracov: You're VC-pilled. Yeah, yeah. No, and I do the same thing at Dock. I mean, we're like a sales-led business. And it's like, okay, well, we got to hit our revenue goals, and it's like we obviously need more sales or increase quota and like all the things. And yeah, it's hard to actually pull it off. Are you noticing that these like AI-native founders are putting off hiring salespeople because it's such a system? Or are the sales teams just much smaller? Like how are they thinking about that team design?

Dini Mehta: I think the bar for talent has gone up significantly. So folks are like very picky about who they're hiring onto their team. And teams are smaller than they were before. So less delaying hires in general, more being super picky and delaying bad hires for sure. And I think the sort of the bottom has just moved up in a lot of these companies. And I think it depends if you're PLG versus SLG. PLG companies, you'll see that they're delaying hiring a lot longer than the SLG businesses, which makes sense because you need the capacity to like show growth. There's a real ramp on the SLG business versus PLG, it's less so. And you can do a lot with AI tooling. So that is like definitely a trend that is happening.

Are AI-era growth rates the new normal?

Alex Kracov: And like one of the crazy things about AI is like these companies are growing just so, so fast. I mean, I think Cursor got to like $1 billion ARR in three years, which is just ridiculous and like mind blowing. And so, you know, we grew up in like the triple triple double double era of SaaS, right? Which was like great growth. But now that's not even that good. So it's like, is this higher growth rate the new normal? Are they outliers? And maybe like how do you think about PLG versus SLG and hitting those growth rates? Because yeah, it's hard to do it if you're sales-led, I imagine.

Dini Mehta: Yeah, no, it is. I think all you read on Twitter or X or LinkedIn is about these companies that are getting $100 million in 2 months. And the truth is, a lot of those companies, you know, the retention and the showing value post sign-ups is still a thing that they're figuring out. And so in this new world with the AI tailwinds, getting revenue is almost easier than keeping that revenue, because you've got experimental budgets and people are very excited to test and run with it.

And I think for companies like — I don't know if that's the benchmark. In my opinion, companies like Cursor, I think that is an outlier. In my opinion, it's a mistake, I think, to expect every company, without knowing the mechanics and the economics of that business, to expect them to grow at that rate, because you'll just make decisions in building go-to-market that aren't good. And so I think a company like Cursor or Lovable — there's so many of these examples — I think they're like the category-defining outliers. And there's a lot more of those than there were in prior eras, because we're going through the shift. I don't think it's the benchmark.

That said, I think all growth expectations have permanently increased. Where, you know, it was like, oh yeah, you can get to like 2 to 10 is a great year. Now it's like, no, you got to go 2 to 20 if you even want to be, you know. So I think the sort of expectations all around have been lifted. I don't think people are expecting a billion in 3 years, but they are expecting — and part of that is because you've got a lot more tooling. So you should expect more productivity from your existing teams and resources. And so I think people are not defaulting to hiring more reps. It's more about, okay, how do we increase pipeline? How do we think about our conversion rates? How do we think about the tooling to drive more efficiency within the org? And so I think that the headcount being the last lever is something I'm seeing consistently. But yeah, the growth expectations are wild right now. It's a tough spot to be in.

PLG vs. sales-led in the AI era

Alex Kracov: Do you think like more companies just need to be PLG to like get that growth? Like is that like a big trend you're seeing with AI companies, or is sales-led still a valid motion? Or how do you think about that balance?

Dini Mehta: I think it goes back to your customer and the product you're selling. And so I don't believe in like — yeah, if the game you're playing is I just want to grow as fast as possible and get to a billion in revenue, then PLG is the path, and layering on SLG on that is I think the only path to get to that level of revenue. Unless you're maybe the labs, but even there there's a lot of stuff if that is the game. But if the ultimate goal is, hey, I want to provide value for my customers, then I don't think it matters whether you're — I think it sort of goes back to what's best for the customer, for your product.

And, you know, at Lattice, we had a PLG motion in the beginning and then we said, no, we're going to go SLG. And I think that was the right call. Even in today's world I would make that call again. And so I think you have to go back to your first principles, sort of ignore the noise in some ways, because in the long run — and all these companies are feeling it now where they're like, oh, they're bringing all the traditional sales-led motions into their org and realizing that they should have done it earlier. Sort of the same stuff we saw in the early 2010s. And now folks are feeling like, okay, we got to layer it on. And literally everyone, from the labs to all the companies that we know are scaling really fast, are building the exact same sales orgs that we know of in the SaaS era.

Alex Kracov: And like there's this huge trend of the forward-deployed engineer who's going to understand your business and do that. It's just so not PLG. It's such a sales-led motion. And those companies are obviously growing really fast doing that. So it's an interesting dynamic. And yeah, I mean, you know, it makes sense that like all the engineering companies, or people who sold to engineers like Cursor, were growing really fast, because engineers are like the perfect PLG audience. But then as you move to different tools, like what we deal with at Dock, it's just so much more of a top-down implementation than it is bottoms-up. At least it feels like.

Dini Mehta: Yeah, I think it's like if the tool is an individual productivity booster, which a lot of these AI tools that are growing fast are, then it totally makes sense to me and it'll scale fast. But if you're trying to go into an enterprise and do like organizational-level productivity, where like a group of humans — so there's a lot of sort of complexity in that, like governance, access, data security. And I think that stuff will continue to be SLG. And over time my hypothesis is that's going to become more and more important. I think we're in that like first leg of this evolution where the consumer individual productivity piece is key. And that's where most of the fastest growing companies are focused on today.

What AI can own, and what stays with humans

Alex Kracov: So it seems clear by now that at least in the short term, AI is not going to replace the sales rep. So I'm curious what AI is actually good enough to own kind of in the sales motion today. And what are the best teams deliberately keeping in the hands of humans?

Dini Mehta: Yeah, I mean, I think the thing that hasn't changed is people are still buying from people. So how you build trust with the buyer still sits with the seller, and the best reps are spending more time doing those pieces, which is trust, discovery, like figuring out the right stakeholder relationships, multithreading early enough, navigating the org and change management that comes with using a new tool. So I think that stuff isn't changing.

The place where AI is adding a lot of leverage is research, data enrichment, you know, how you do call summaries and CRM hygiene, providing that context layer across your whole go-to-market org. I think that is the place where you can add a lot of value, because when you think about sales, there's a lot of tasks in sales, but the job is only one, which is you're the human interface for representing your business and helping customers buy. And so the job hasn't changed. That is the job. But the underlying tasks that used to bog down a lot of sellers around, you know, you're spending so much of your time on admin work — I think that has materially reduced, and the best orgs and sellers are putting their salespeople in front of customers more, sort of hoping that that's going to drive higher win rates.

Who owns the AI transformation on the go-to-market team

Alex Kracov: And like, who owns this AI transformation on the go-to-market team? Is it the sellers building workflows and stuff themselves? Is it leadership? Is it rev ops? Is it the CRO? Like who have you seen kind of owns this tooling and stuff?

Dini Mehta: I think we're still very immature. I think the EPD side, the engineering side, has seen a lot more — has seen the transformation with these AI tools. I think we're very early innings in go-to-market, in my opinion. Like being in all these CRO circles, you talk to folks and everyone's experimenting, but I don't think anyone has a "oh, here is the playbook for this new world." And so I'll say that at the beginning.

But it seems like it's a combination, where a lot of times people are saying, we're going to have a centralized team called go-to-market engineering, call it AIOps, call it rev ops, that's going to own building these workflows or agents for folks. But then you've also — you know, the best orgs, they're also pushing it to their teams and enabling them to build their own agents, because if you're doing it right and you're scaling fast, you're probably not going to be able to get to all the things that the org needs. And so having a combination of both is probably the best bet. And finding the right tool that gives you that operating layer above the intelligence layer to sort of do both pieces is — because I think, in my opinion, the centralized approach makes more sense, because you can almost distract from like the job is to go spend time with customers and sell.

And my worry with, you know, getting every person to build their own agents is like, you know, you're turning into like AI theater of like, here's an agent I created and, you know, like, should you be doing that with your time? Like it's awesome that they're excited. So I worry about that if I was in the seat, like having AEs tinker. But on the flip side, I want them to learn. And so I think it's a delicate balance.

Alex Kracov: Totally. It feels super messy right now. I totally agree where it's heading, where it's like, okay, centralized AI tooling, more top down. But then like, I think about our own sales team at Dock and like Christian on our team is, you know, better at figuring out AI and has come up with like really interesting things that he's doing with Claude or whatever. And like without that sort of bottoms-up, I think we would be more behind on our own AI adoption at Dock.

And then like even trying to sell AI is a mess, you know, to go-to-market teams, because it's like, who do I talk to to sell this, you know? And like they don't even know on their side. It's like, is it the enablement person? Is it the ops person? Is it this random sales leader, this random rep? And like everyone has like AI ideas or thinks they know what to do. And it's like, really everyone has an opinion, but then who actually owns it and implements it is different in every company. Honestly, what I found is like there's usually like an AI specialist hidden at the company who's like the AI guy or gal, and like you got to just find that person and then you got to like befriend them and get on their like their workflow, whatever their random workflow is. You got to kind of fit yourself in. It's very messy right now, it feels.

Dini Mehta: Yeah, because it's not a specific title. And you know, there's some orgs where it's like starting to emerge. But yeah, I agree, it's not standardized at all. And it's super, super messy.

Alex Kracov: Yeah. We've seen a lot of our deals go to like this AI committee at a company, where we get vendor of choice, like we're in procurement, and then there's like, oh, there's this random AI committee that's going to like say yes or no and make sure — which makes sense, like going back to the standardization, is like they want it to fit into everything. But then, yeah, it's a new type of discovery question we got to ask, like, hey, do you have this AI committee that's going to review Dock or whatever it is?

Dini Mehta: And it's changing so fast. I feel like most tools, like even tools that didn't require POCs, are now like, I want to trial it, I want to run a POC before I can make a decision. So that is the complexity — it's real.

Cutting through the AI tool noise

Alex Kracov: Like there's so much AI noise. Like there's a new go-to-market tool every week. Every CEO is yelling at their company, go use AI, experiment with AI. Like how do you advise revenue leaders to adopt this without drowning in tools, breaking your own process? Like, I mean, I know from even our work together at Lattice, like there was only so much new stuff you could throw at a sales team without distracting them. So like what should be an organizational mindset when trying to roll out new technology to sellers?

Dini Mehta: Yeah, I think it has to start with — like you have to come up with what is your AI strategy for go-to-market. I think doing it sort of ad hoc based on like the last LinkedIn post you read or what your founder CEO forwards your way is not the way. I think the path is: okay, let's come up with what are the most time-consuming workflows in our organization that is really bogging seller time into admin work. So how can we now think about automating pieces of that, and what are the right tools that help us do that? So I think almost thinking about it in phases and getting buy-in across the org and then saying, all right, we're going to do this. And then sort of ignoring a little bit of the noise, because there is literally a new tool every week.

And part of the reason I think this is creating some level of like paralysis in buyers' minds, because they're like, well, there's going to be a new tool next week and maybe Claude solves that, maybe OpenAI solves this, maybe I'm just going to wait. And so I think it is a weird time in market. And so if I was a revenue leader today, I'd say, all right, I'm going to map out the go-to-market org, we're going to do this for the next six months, put our head down, drive some efficiency, learn from that, and then sort of figure out the phase two — versus sort of being in this constant shopping mode or constant paralysis of like, I'm not going to do anything because maybe the labs will just solve it magically.

Culture and team building with smaller teams

Alex Kracov: Yeah, we're definitely feeling that this quarter at Dock — people being very thoughtful with their buying decisions and how does it fit in. And it's like, oh, suddenly everyone needs MCP, right? And it's like, no one knew what that was six months ago. And it's like a really funny dynamic in the market today.

Switching gears a little bit — like one of your superpowers is team building. And like I believe, you know, we had like a 1% regrettable attrition rate on the sales team at Lattice, which is insane. Like your entire team loved you. But the AI era — like obviously we talked about pushing towards smaller, leaner, more automated teams. And now there's also this dynamic of remote culture too, that more companies are dealing with. And so what does culture and team building look like in this new era for technology companies? Is it the same? Is it different? How do you sort of think about that?

Dini Mehta: I think it's more important in this era, given that you've got smaller teams, remote teams, so leaders have to really lean into it. Otherwise you're going to turn into a mercenary transactional org that is just focused on how does this help me today? And if it doesn't help me tomorrow, I'm out. Which, you know, it is a strategy and I'm sure that works until a certain point.

But if you want to build sustainable orgs with good cultures where people are there for community, growth, purpose — I think those pillars still matter, in my opinion. And if anything, they matter more, because you've got smaller teams and you're putting sort of more onus on these folks. And so I think investing in the humans — and my hope is that managers have more time to do that, given that AI has taken some of the stuff off of the manager plates around, you know, let's do the admin work, the managerial work, so you have time to actually spend on helping folks become more confident, helping folks with their own growth journeys, helping them figure out their career path. And so I think that stuff matters even more in today's world. And it's a mistake for orgs that are not prioritizing that, because ultimately humans want the same things out of work. Nobody wants to just come in and be a machine, because you've got AI for that.

Alex Kracov: Do you think — I mean, you said it in your answer, but to go a level deeper — do you think that personal growth and development is like the key to building a good sales culture? If everyone feels like they're getting better at their job, that's going to create the winning culture? Or is it something else?

Dini Mehta: Yeah, I think in my opinion it's three things. One, it is growth. Like you want to feel like you're — and it doesn't mean promotion. It means am I getting better at the things I want to get better at? And that requires understanding who a person is, what is the path that they want to take, and what are the areas that you as a manager, you as an organization, can help them get better at. So I think that is a core tenet of making people feel like they want to stay at an org, because most people leave because they're like, oh, I've learned everything I could here, or I don't like my manager. Like typically those are the majority reasons why people leave companies. And so if you solve for like, hey, we're going to really invest in your growth and help you get better every quarter — and that's going to be an intentional effort from the top down, just because it helps the business, it helps the people.

Two, it's community. Like do people feel like they're part of this one team that's fun to be around? Because, you know, this is hard stuff. Things are moving fast. It's stressful. It's high pressure. But if you can also inject a little bit of joy and fun into it, and people see each other as, oh, we're part of this community, I think that matters a ton.

And then the third piece is like feeling connected to the mission of the company. Like what is the thing that we're doing here? Why am I working 60, 70 hours every week? And, you know, how does this connect to my — ideally it connects to your personal why. But if not, at least like have some level of a purpose for the business that gets you fired up. So I think that doesn't change. Like, yeah, commission plans, and yes, those things matter in sales. But at its core, all of us want the same stuff out of work.

Hiring salespeople in the AI era

Alex Kracov: Going to like the top of the funnel of how you even bring people into an organization — like is the hiring profile different in the AI era? Has it changed how you approach hiring?

Dini Mehta: So I think hiring, some of the stuff is still the same because, you know, I think the job is still similar, which is, you know, curiosity, being an ability to adapt, being, you know, being cultural. Those things have always been important. I think the new things that are very specific to this environment is your ability to be a systems thinker, like being able to think in a structured manner, linear way. Is your technical comfort with like using agents and tools and co-working with agents as coworkers. So I think the technical comfort.

And then the third piece that's super important is your learning velocity. The amount of stuff that is changing in the market, plus in most companies, as like codegen becomes easier with all these tools, the product velocity is like off the charts. Like that was a challenge — like we used to have it at Lattice, like, oh, we've got so much velocity, how do I train my teams to know what's going on? I think that is like an exponential challenge today. And so someone — you know, that would be a core attribute that I would want to hire for is like someone's ability to learn fast and then adapt quickly as things change, because you have to meet the market where it's at.

Alex Kracov: Totally. And to be like a good consultative seller, you got to understand all these new concepts and all the new AI things. Like, you know, we just did a training on MCP yesterday at Dock and it's like, okay, this is a brand new concept that customers are figuring out and we're figuring out and how does it all fit together. And yeah, no, it's a time of a lot of change, which makes it fun. And there's opportunity too.

Dini Mehta: I almost feel like the new rep is like a part consultant, part operator that's comfortable with tools, part product expert, and has to like be AI-pilled. Which is a hard — like it's hard to find all those pieces, but you can create an organization where you find people with the attributes to learn those things and become that new model. But I really think those are the folks that are doing the best work in sales.

What happens to the SDR role

Alex Kracov: One of the parts of sales that does feel like it's getting automated a little bit is outbound. And then, you know, in the SaaS era this was the place where you hired junior people out of college and trained them how to be a good sales rep. And so I'm curious how you think about this dynamic. Like, are there no more — like should we not hire junior stars anymore? And then if we do not hire them, or there's less of them, then how do we train kind of the next generation of sales folks?

Dini Mehta: Yeah, I mean, the traditional SDR role is under pressure, but I don't think it's going away. I still think there is value in having those teams, there's just much smaller. So I think the challenge of like, how do you build talent and what happens to your talent funnel, is a good one. I don't know the answer to that, but that does worry me — is like, how does that transition from the SDR to AE to manager is like now disappearing?

I think the activity-based SDR sending a bunch of sequences and emails is definitely disappearing. You don't need that anymore. But the value-based SDR that is making calls, experimenting with new channels, thinking about account strategy, knows how to navigate an org — I think that becomes more important in this world. And so again, like smaller teams, more AI tools to help make them more efficient. And I think that is the future of the sales development organization.

I've thought about like, does the whole thing collapse back to like when I came up in sales, there was no SDR. It was very much like you cold call, you find your deals, you close them, and you also sort of post-sale like help with renewals, expansions. It was very much like a full cycle, full stack role. And then over the last couple of decades we've like over-specialized, for good reason, partly because it drives efficiency and it becomes a machine. But now with agents, you could specialize agents and go back to this like generalist model where a person could do all of it. So I think that could — I think, yeah, that's like the hypothesis that I have is like long term that becomes the new model, starts emerging. But in today's world, I mean, the labs are hiring lots of SDRs and BDRs. So that gives you a signal that I don't think we're anywhere near sort of done with this. But I think the sort of rote "let me just send a thousand emails to get X meetings" — I think that is over, because AI can definitely do that.

How onboarding and enablement change

Alex Kracov: And I feel like — I mean, it's like any marketing or top of funnel channel, like we just abuse it as an industry and then it becomes less effective and then you change it. And yeah, I'm curious how onboarding changes in the world of AI. Like, I don't know if you've played around with like the AI sales role play tools. Like how does this all change? Like do you even need onboarding managers anymore who are going to help you bring in the next class and do the big cohort training? You just have AI train everybody? Like how do you think about that?

Dini Mehta: Yeah, I mean, I think like before we used to teach product, process, the messaging, the people. I think most of that stuff can be automated through AI. And so where you need folks learning now is like, okay, how do you research using our existing tools? How do you prompt our current AI tools that we've got? How do you access the internal knowledge system? How do you automate some of the repetitive tasks that you would be doing in your day to day?

So I think the role of onboarding is different now, because I think spending too many cycles on product doesn't make sense because it's going to change so fast. And that is the mindset that most of these orgs have, like, hey, we're going to build products and we know in a year we might have to fully rip it down and build a new one, and you can do that in this new world. And so I think sellers have to almost learn how to learn fast. And that's the job of onboarding, versus like, here's the things we're going to teach you.

So I think it's almost like all the tactical stuff AI can handle. But the judgment around what should I prompt, what should I look for, who should I go to — those are the things. The map of the organization, which includes agents, tools and people, is, I think, where enablement and onboarding should really be focused on.

Alex Kracov: I totally agree. And I mean, I think it's honestly across every single function — it's just how do we, like what is our agent stack? Where do you ask questions? How does this all work? And there's so much — I mean, we try and teach our customers. It's like, what's a good prompt? You know, because we see people just write bad prompts and then they get bad results and then they're like, oh, Dock doesn't work. And we're like, no, like come on, you know? And so there's like this whole evolution of how should you work with AI. And what makes it crazy is the AI technology is obviously changing really fast too. So that answer is changing as we go. So yeah, that's why — I mean, it's what you said. It's like you got to just be good at adapting and learning. And that's like the biggest theme I would say of our conversation.

The role of the sales manager

Alex Kracov: What about managers? Like, you know, you mentioned it before, managers should have their time back. Like, you know, Jack Dorsey is getting rid of all middle management at Square. So like there's these crazy thoughts around, okay, we don't need managers anymore. Maybe the AI can just be the manager. But you know, you talked about in your kind of culture answer, like managers are a big part of building culture on a sales team. So, you know, how do you think about this dynamic of the role of the manager in the AI era?

Dini Mehta: Yeah, I mean, I think the things that managers used to get bogged down with in sales, for example — it's like forecasting, looking at activities, call reviews, putting comments on calls, coaching recommendations. I think a lot of that starts to get automated through AI, which is great, because you've got tools like role playing tools, you've got comment tools, forecasting that can give you alerts, which is great.

But I think your manager's job — I think so much in sales, there is a psyche management piece which is critical, because the amount of rejection you're getting coupled with the pressure you feel to deliver revenue every single quarter uniquely sort of makes sales like a psyche management job for managers. And the best leaders do that extremely well, which is, how do you help reps feel confident even when things aren't going well? How do you sort of help them build executive presence, the career development conversations that keep people motivated and excited and feel like they're growing?

And then I think judgment. Like so much of the work is done through AI. Like you don't need to think a lot to do that. So now it's like picking — like AI's going to give you four answers. You still have to decide for your customer what is the right answer. And I think that is the thing where managers should be spending time — is like helping people learn faster, helping them grow, and then helping them develop better — I hate the word taste because it's overused, but, you know, I think develop a better taste and judgment around what matters.

The CRO as revenue architect

Alex Kracov: And like, what about your job? What about the role of a CRO? Is that changing in the AI era?

Dini Mehta: I think so. I mean, I think the shift has been happening and it's just accelerated with AI, which is the CRO is no longer just a leader of people. And I think the best CROs think of themselves as like, I'm the revenue architect that's building the system that optimizes and increases human leverage — which is, where can I put humans and where can I sort of take away all the tasks and the work to agents and AI? And so I think the CRO's job is no longer sort of, you know, just hire good people, get them to go get numbers. I think it is workflow design. It is systems thinking, and then figuring out where do people fit into that. So it is definitely like shifted to that direction.

Alex Kracov: It feels like rev ops has got to just be much bigger in this era. I mean, I think it's just AI and it's like the same type — like I think Shrubby from Lattice would still be very good at this type of thing, right? Like those are systems thinkers. And like, yeah, I feel like that's the—

Dini Mehta: They'll become the future CROs of these orgs, because I think, yeah, you're right. That is the most important part of like, how do you think of systems building.

Selling to AI-armed buyers

Alex Kracov: Super interesting. Okay. So sales reps aren't the only ones with AI. Now buyers show up having like used tons of research with AI, the category, they find vendors, they can even like pressure test your pitch before you even take a call. So how does that change the seller's job, with buyers being just armed with way more information? And where can a seller add value that AI can't?

Dini Mehta: I think it's building trust. That is the thing that humans excel at — building trust with the buyer and helping them make the decision across their organization, given all the pieces and all the sort of moving — helping them prioritize what is the business problem and what kind of value does this provide, where does this fall short. Because everybody's got information and it's no longer about giving them the most up-to-date information about your company. Because if you're doing that as a seller, you're failing.

In today's world, your job is to be the decision facilitator. And to do that, you have to be trusted by the customer. You have to come off as unbiased. You have to be knowledgeable about the market, about your product. And so that's why, again, learning becomes the core tenet. Learning and building trust are the two core tenets of the best reps. Because yeah, the buyers don't need — you know, there's two companies that do like the AI avatar sellers. Like you could get a demo, you can get questions answered by buying AI easily today. So the thing that is still unique and where humans have value, which I don't think goes away for a long time, is prioritization, judgment, trust building, helping them navigate the org on their side, and sort of making sense of like, hey, here's the thing we've identified, how do we sort of make sure all the people see it the way we see it? So that stuff still stays, I think.

Alex Kracov: I remember that being one of your mantras at Lattice too, is like people buy from people. And, you know, it's like it's not even just about the software selling. It's about the people on the other side. And do you trust them? Do you believe them? Do you believe that they're going to pick up the phone when you need something from them? And especially in a world where software maybe is getting commoditized, the people who manage that service for you are really, really important. It starts with the seller.

Dini Mehta: Yeah. I think that like my mantra, which I'm sure you remember from Lattice — like I used to say all the time — is the number one thing we sell is the fact that we care. And people have to feel that care in every touchpoint. And I think that becomes more important, because if you do that well, like the other pieces I think fall into place.

Is selling AI different from selling software?

Alex Kracov: Okay. You've worked with a ton of AI-native companies. Is selling an AI product different than selling traditional software, or is that distinction overstated?

Dini Mehta: Yes and no. Because I think people used to buy software for efficiency, mostly. I think most people are buying AI to like rethink and transform what they were doing. So it's less efficiency. I think there's pockets of where it's like just, hey, we'll make it better for you, but a lot of times people are buying for like, how can I transform how I'm doing this thing?

And I think the difference is, as you said, buyers are coming in better informed. They're more excited, more curious, but also more skeptical and more burned by like the promise of AI not having panned out. And so I think we're seeing that shift this year. Like last year was definitely more experimental budgets, let's do whatever it takes, like let's buy all the tools. And this year feels more, well, I want accountability and I want actual results coming in from it.

And I think the velocity change we talked about — like the product's changing, the market's changing, the competitors are changing. So that is definitely different. And a lot of times in AI you have to be closer to the outcomes, because you're not just selling like this feature and this product and this widget. Like you're talking about how does it actually impact revenue, you know. And good software always did that. But I think in the world of AI it becomes more important to like — the ROI comes in time savings and true sort of ROI for the end customer.

Pilots, POCs, and proving outcomes

Alex Kracov: Do you find that like in this era, people need to be more PLG, or if it's sales-led you need to do more pilots? Because you know every vendor's claiming these outcomes. Oh, AI is going to do X, Y and Z for you. But then how do you differentiate, back up the claim, build trust? Like can you just do the demo thing or do people need to get their hands on the product more?

Dini Mehta: No, I think you have to give people a chance to test the product. Like that is definitely a new thing. Like POCs and trials are the norm for sales-led organizations. I think people — these committees on the buying side where they're coming in testing. So I think the good news is there's some standardization starting to appear with how folks test and buy tools in the AI era, which didn't exist in software. But whether that's PLG or sales-led, I think getting folks to try a product is critical, because I don't think folks are comfortable enough to say, based off of a demo I'll buy. I bet that's the next phase of where we get to in this shift. But today it's super important to give folks a chance to test it.

Protecting teams from burnout and job-loss anxiety

Alex Kracov: And it's a tricky dynamic too — how do you structure that pilot in the right way. And this is always true, because especially if you need to add the data into there and the security concerns, it's an interesting new dynamic to navigate these AI pilots. It's fun. It's my life at Dock.

You know, there's a lot of talk in the market about how we all need to do more with less. There's this threat of job loss and like this whole kind of interesting dynamic with AI in general. Like how do you protect your team from this dynamic and not getting super burnt out by just the constant rate of change that's going on, this weird dynamic of job loss in the market? Like how do you keep people's spirits really high in this era?

Dini Mehta: Yeah. I mean, I think while there is job loss and, you know, we see the headlines — on the flip side, there's also new jobs that are getting created and hired for, which typically don't make the headlines. But that is the truth. And my sense is — my hypothesis is that's going to continue happening. Like I don't think we're going to go to a world where we don't — you know, we'll have very small teams and we're just going to lay people off. I'm an optimist. And so I believe that we'll find new roles for people in this world, because agents in some ways are junior employees. And so like, how do you — you still have to manage them. You still need people. You've got like higher span of control. But I don't think it goes away fully.

One of the things that folks get wrong, in my opinion — efficiency should mean like more impact per person, not more work per person. Which I think sometimes, you know, with the silly 996 stuff can just — I'm like, it's about the impact that you're driving and the outcomes you're driving, not the amount of hours you're clocking. The best people actually are really good at prioritizing, figuring out what's most important and getting it done and seeing really big outcomes. Like that's always been the core.

And so I think if I was a leader, how I would protect against that is I want to see the impact the person is having, but I also want them to have a life outside of work, because otherwise burnout follows. Like, you know, you're not doing yourself or your company a favor. You might think you are, but anybody that's been around the block knows how this ends. And so I think your job as a leader is to use AI to remove all the stuff that is painful, like all the admin work, all the stuff, every job — stuff you don't like doing — and so like use AI for that, but give people the gift of like being creative and figuring stuff out. Not like, oh, now you got to do like 19 other things because I took that away. That just seems silly to me.

Why sales and marketing worked at Lattice

Alex Kracov: I love that answer. I mean, no one liked updating Salesforce, right? And so hopefully we can automate that away with AI and no one's going to be upset about that. And then, you know, if you really like sales, you're going to get more time working with customers and solving their problems. And like it should be a win-win for everyone.

Okay. I think one fun place to end today's conversation is talking about sales and marketing. Because, you know, we spent four years working together at Lattice, you as the CRO and me as the VP of marketing. And so I don't know, I'm curious, like what do you think made our relationship so successful? Everyone talks about sales and marketing hating each other and getting in fights. We were never like that. Why do you think it worked well?

Dini Mehta: I think one, it was like a common understanding of like, we want to go build a big business. I think you had that, I had that. We weren't focused on like, does marketing hit their goals, does sales hit their goals. Like it doesn't serve anyone. I think this attribution challenge is overstated, even though it's not well understood. So people over-fixate on like MQLs and SQLs. And the first few years we didn't even worry about that. We were like, pipeline and revenue is all that matters. And yeah, like the lead sources matter for budgeting purposes, but in reality — I love that you managed your team on revenue attainment too. Like you'd be as worried if we're behind on revenue as I was, which sort of felt awesome, like I had a partner in crime who was worried about output just as much as I was.

And on the flip side, the sales team was grateful for all the work marketing consistently was putting in to bring in pipeline. So we had this sort of cohesive approach of like, marketing's providing air cover, but sales still has to go be the ground cover and have the conversations with customers, be aligned on what we're saying in our brand and campaigns with what sellers are saying to customers in one-on-one conversations.

And I think that sort of one team, one dream mentality was the core thesis, where I think everyone in go-to-market sort of felt like one large, one team, one dream. And I've been in orgs where I was like, everybody's pointing fingers at each other, and what you're missing is like you're letting the competitors win because you're too busy fighting stupid internal fights instead of focused on what really matters, which is we got to grow the pie. We got to build a big business, and optimizing for enterprise value is all that matters. So I think that was my favorite part of working with you and the team. And I think that's one of the reasons why we crushed it.

Alex Kracov: Yeah. It was great. I mean, as you're giving your answer, all I was thinking about was one team, one dream, and we beat that drum to death at all of our all-hands and things. But yeah, well, it was a lot of fun working with you. And it was really fun to catch up in this conversation. Thank you so much for the time today, Dini.

Dini Mehta: This is awesome. Always so fun to jam with you on stuff. And yeah, thank you for having me.

‍