Dust x Clay - GTM in the Age of AI - full conversation

Hi everyone. We're gonna get started. Thank you so much for coming. We are super, super excited to be here. I'm Tess, I'm Varun's assistant for those of you who don't know, and this is Varun and this is Gabriel. And they're gonna tell you a little story about why they're dressed like they're dressed.

We're wearing bathrobes. Yes. Why is that?

Because I thought we were going to the sauna.

Well, yeah. I was curious.

Skinny dipping.

I think we—

Yes, we went skinny dipping last week.

We just went skinny dipping last week.

Yeah. There we go. This ridiculous outfit.

40 degrees out, flip-flops on the streets of Helsinki.

Hey, this is Gabriel, founder of Dusk. Very happy customer over here. This is Lars, he literally invented Google Maps. And we are going into the Baltic Sea. All right, let's go, let's go! Wow, we're leaving! But it's happening! All right, we're back. This time we have slippers. Walk from here to the cold water. It was too long. By the time you're there, you're already freezing. All right, let's go!

Breathe.

All right, it's time. Yeah, I did longer than 2x better.

That was, that was, that was, uh, firing cold, right? That's what, that's what the star's for. Um, content. Good content.

I thought pretty good content.

Memorable content.

I don't know what the audience thought, but I thought it was pretty entertaining.

Do you count your views on LinkedIn or YouTube for that kind of stuff?

Um, I don't count the impressions, but somebody does for you. But, uh, I hired someone who does count the impressions. Yes, attracts them. Yes.

Yeah.

But I do it for the love of the game, Gabriel. I do it for the love of the content.

All right. Well, Varun, welcome to Paris. Uh, we have a lot of people excited to talk about go-to-market in all of its new shapes, forms, sizes, and flavors. And let's do it. Let's go. Let's start. So I think we should start. I'm doing well.

Yeah.

Yeah. Finland was a bit cold.

How are you guys doing? Feeling hyped? Feeling energy? Lots of people out there, people down here. Wow. Let's go.

At some point these are going to get warm and we're going to take them off and it's going to feel like a real sauna.

Okay, great.

But I'm waiting to get a little, a little toastier. Great.

Great. Let's go. I'm hyped.

Okay, so what are you doing in Europe? Why are you here?

We're launching in Europe. That's, yes, we're launching in Europe. That's why I'm here. We're on a month-long Euro tour. Gabriel, we didn't actually have Paris on the agenda originally, but Gabriel—

Can you believe he did not want to stop in Paris on a Europe tour? Yeah, exactly.

No, no, no, no, no, no, no, no. It was just that, you know, we're trying to be focused, you know, disciplined. And it's like, let's start with the English-speaking countries where we have a chance. But then it was like, wow, the momentum is crazy. The community was really great. Like, everything was going so well. And we were in the sauna, and it was so compelling. And I was like, fuck it, let's do Paris. You know, let's go. So we tacked it on, and now we're here, and it's happening.

We're really happy that you added it on. And I think the room will appreciate just how disciplined and focused you are from the way that decision was made.

Yes, yes, yeah, very disciplined, very—

Yeah, but it seems that it's a pattern with you. We— I remember the first time we met where we were grabbing CitiBikes and pedaling down New York for a dinner that we had just decided to organize, which we got there just on time, roughly. And so there seems to be a pattern of great just-in-time content.

Well, actually, it's funny you say just in time. Did you get that from Kareem at all, or did you cut independently? Just out of my brilliant brain? No, no, because Kareem and I have, 'cause I know I hadn't mentioned G, but Kareem and I actually describe a lot of our decision making.

Kareem is the—

Is my co-founder. Co-founder, okay. We describe a lot of our decision making and hiring as like just-in-time decision making and hiring. I'm not kidding about this. Like so many of the hires we made in the early days, like, you know, like let's say you're hiring some leader, right? The smart thing to do, the disciplined thing to do would be to interview, I don't know, some real number like 20 or something and then be like, okay, have some calibration. Figure out what feels good and then go for it. A lot of our leaders were the first people we met, actually. How many of them are still there? All of them. In fact, no one has left Play except for being fired or for starting a company. That's great stuff. So they're all still there. And it was like just in time. And I was like, wow, if we didn't hire like Jess Bergson, who leads our CX team, and, you know, She just appeared kind of, and we interviewed her and it was like, and we had like a couple of concerns, but mostly it seemed like a really strong fit. And I was like, man, we would've been so fucking screwed if we didn't hire her at that precise moment. 'Cause then everything started blowing up. And then like, if we didn't have her at that moment, it was crazy. So a lot of just-in-time decision-making. And then this Europe thing, just to like maybe just take an actual decision and unpack it for a moment. 'Cause actually all the decision-making happened on the Strip, right? So the original decision around Europe was like, okay, so why come to Europe?

Well, you say you launch, but you're already available. What you're launching is a team, an office.

Where is that?

Where are those?

Yeah, so we're already available from self-serve. And so 20% of our overall revenue is already coming from Europe, but mostly from self-serve revenue. And we have like, you know, a dozen customers on enterprise plans here, but they are very notable customers like Mistral here in Paris or Adyen or Miro. Or Lovable or Lagora, but it's not like a real presence. Like we don't really have people on the ground. So the launch is an office in London, people, a team to really commit. And we were like, okay, let's do this methodically. Let's do this plottingly. And like, let's start with UK and then the Nordics and then like, and then, and then I, and then actually, but then coming here, I just being on the ground and meeting everyone, meeting the communities, like doing all these dinners. I'm like, the opportunity is unbelievable actually. And it is—

who's looking for a job in Paris at Klei?

Not actually in Paris, unfortunately, because I have been, I have been convinced to moderate a little bit. Like, I think basically— so while I was here, I started talking to other, like, people on international expansion and stuff. And I was like, it's so— because literally just today I was talking with Karan, who's our CFO, and he was like, you know, why don't we be a little bit more methodical about this? Like, there's so many downsides to rushing. And he's like, you could— like, what are some downsides? He's like, Hey, what if you fail at the expansion? Like you saying it's just one person or two people, like it doesn't matter, which is like the cost of it financially is not that much. But he's like, what about the morale? What about the— like, Clay's on this like amazing trajectory. And if you don't and you fail in a market, like it might feel shitty, you know, and it might like— you can't, you can't put that genie back in the bottle. And my response to this— literally this conversation literally happened today, by the way. My response to him was like, It's kind of hard to argue with him sometimes because every point he makes is very logical and methodical and, uh, and well-reasoned. And so it's like, you can't really argue with that. But I oftentimes think in startups and in go-to-market in particular, sometimes you have to be a little irrational and you have to be just like aggressive. And, um, he's like, what's the worst thing that happened? We lose like 2 or 3 months. And I'm like, the 2 or 3 months is the difference between like us doing something. And I, it's hard to like, The other thing is like when you're here and you're meeting all these people, you, it's, it's just instinctual. Like you see the opportunity, you see the customers and you see the lack of competition and like, it's just, you just need to execute. And it's hard to describe that emotion.

And I think doing it from the US, I feel like maybe also the, the scale of the mistakes coming to Europe might, might feel more moderate. We're going the other way right now. We just opened San Francisco.

Let's go.

It seems like a big mistake.

Are you moving?

Yeah, not yet. No. Okay.

Okay.

No, no, no, not yet.

You're not losing him yet.

Well, traveling there very often.

Yeah.

But like when you start the US, it just seems like a huge effort. Yeah. And everything, I don't have a CFO yet, so I guess I don't have to have those conversations.

Yeah.

Have them with myself. But like the cost of trying something and trying it well versus rushing it, because I think maybe I'm slightly more methodical on some aspects of this. And I wonder like you have to do it well enough to give yourself a chance. Like there might be something as doing it too rushed and then just not even giving yourself a chance.

I don't think they were completely disorganized. Like this tour was like fairly well organized T-shirts. We have T-shirts and fire extinguishers and, and from Tess and, and like a whole social media campaign and like rigorous posting about every single event that we do and, and like 20+ events. And so like it is very well coordinated in many ways. And, and, and we are going to like hire a couple of like marketing and stuff like that. So there are things, but I think it's just kind of like the scale of ambition and not only just like, because I think from the beginning of this tour 3 weeks ago to now, it's gone from like something we're doing to kind of like a top 3 company priority. And I think that's like an insight you can only really get if you're here and meeting people and like getting that.

So basically you should get— it's not called go-to-market, it's called like jump-to-market. So that's basically— so you're going to be hiring jump-to-market engineers or plunge-to-market engineers. So one of the things that obviously is amazing about Clay is its brand, really well recognized. Tell us about the beginnings of that. Was it always the case? Was it always something that you and Kareem focused on a lot? Do you feel like it has sort of gone beyond your wildest expectations, is now sort of out of your control, in the hands of your community? How do you think about the Klai brand today?

I don't think it's gone out of control. I do think that it's always been kind of core to us. This is, by the way, another thing like I love Karan and actually we do these videos now where he analyzes like marketing decisions.

That's how many— that's how many times you hit on the table where he analyzes marketing decisions.

But like Early on, we made a lot of somewhat irrational marketing decisions on brand, right? Like we bought Clay.com before we had revenue. And let me tell you, it was expensive. Like we, we hired a full-time claymation artist, but before that we were contracting with him and he had a very innovative pricing model where he would be like, hey, here's this blob, okay? And that blob, it costs like $4,000 and, and it's like a beautiful blob. Let me tell you, it's like purple and orange.

It's very nice.

It's like this big. And if you— you can only rent it though. You can rent it for a year and then you have to renew your license to the blob.

To use the blob.

To use the blob. Yeah. So it was like a SaaS business model for blobs. And like, it was kind of irrational, obviously, but the blobs were beautiful, you know? And we hired a head of brand and we were like 10, 15, 20 people, you know? And so anyways, these are all things that you—

So head of brand, not brand marketing, just head of brand. Is that— does That person's role just managing the brand? What are their responsibilities that that person has?

This was more of a plunge, I would say. Yeah.

Okay.

But ultimately, shortly after him joining, he ended up becoming the head of brand and people because he had the soul of Klay. And so, and this was Kareem's insight. Well, it was like, it was Kareem's insight that, hey, wouldn't it be great to have someone who could connect to the external experience? With the internal experience and make that consistent and have a really tight feedback loop there. Because that's oftentimes when like difficult things, that's when people leave a company where, where they feel like there's dissonance between what you're saying externally to what's happening internally. And so anyways, these are all—

I remember you telling me you do a lot to pick up on all the signals that you collect from candidates as you interview them and sort of decide what the best ways to make them happy and make them useful and make them have impact to the company might be. Is that something you just feel out? Is it something that everybody who's interviewing at the company takes part in? Yeah.

So we, we're, I mean, basically the whole point of interviewing is, and just to finish the last point, basically the, like, you, these are things that are core to us and we really care about and they don't make sense in the moment. Obviously if the business works, it becomes a differentiator and a moat, but if it doesn't work, it doesn't matter. But if it doesn't work, at least we did something that was beautiful and that mattered to us, right? On the interviewing point, and actually, by the way, when we were at Europe 100, you were like, showering me with like, uh, with people inside. We should come back to that one. But, um, but I look, I think the whole point of interviewing is just, so Kat, actually we just interviewed and, and she's joining Clay, um, to do events and, uh, coincidentally is at an event. So, um, uh, let's, let's, let's see what she thinks after. But anyways, the point of interviewing and to interview Kat as an example is you're just trying to kind of extract as much information as possible about her, about the human who she is. Um, and then kind of take that information and compare it to what you were looking for, right? Or what you want. Um, and so we try to do interviews that do that. Um, and, and one of them is like a deep dive interview where we just like try to methodically go through every kind of decision you've made in your life and things like that. But I did want to come back to you because I felt like— and so last point on that is I think our intent is how do you find unique people who can, um, who can give you something different. So for example, Kat's colleague, Mustafa, who is the hiring manager of Kat and is our head of events. We really care about doing events in a way that feels authentic. Mustafa went to Carnegie Mellon and obviously comes from a technical background, ended up organizing Daybreaker, which is kind of like a Burning Man-esque kind of event in San Francisco.

It's a sober morning dance party that is amazing and colorful and vibrant.

Yeah, yeah, yeah. And then also like organized events for VCs. It ended up not even working out in some ways. And so it was like complicated where like, rarely do we hire people where like references in certain places don't make sense. And so you have to like really be intentional about the fit of the person because some like, like that type of personality is not going to maybe work well in a VC environment, but could thrive in a creative environment like Clay. But I wanted to ask you about the tenure, because I feel like there's such an innovative concept that you've implemented at Dusk that I feel like more people should know.

We haven't implemented it yet, but we're thinking about it.

I thought you had. There was like a few people—

We're prototyping it with the engineering team.

Okay, okay. So let's hear about the concept.

So the idea is that we— people familiar with the concept of tenure at universities? Yeah. So you basically, as a professor, you join the university and at a certain point you're given tenure. That means you're sort of impossible to fire.

Or just like any teacher in the United States.

In France, you're impossible to fire anyway. So it's just really, we're not inventing anything new. But the idea is that to make people really excited and happy in the long run of a company that's going to go through many different phases, you kind of have to really make sure that they're perfectly aligned with what they're doing and not just a sort of hamster wheel of new trophies and new carrots and new levels because you're just going to be adding them. So the experimentation is to say you're on the team, and then at some point, we tell you you're tenured.

And how do you tell them that? Like, what do you—

We add them to the Slack channel of tenured engineers. That's a ceremony. It's really— and then they get to make a lot of the decisions about how that—

And is this the first time everyone else is learning about the tenured, or do people know that there's—

I don't think there's any engineers on our team in the room, so I think it's definitely—

It's still a secret.

No, it's a test.

OK, OK. It's low-key. It's low-key.

What's exciting about it is that it does seem to align well with the people that we think will represent the brand, the spirit of the company, and a bunch of the decisions that I think are really hard to make alone as a founder.

So what happens when you're, um, I mean, you do have Stan, obviously.

And what happens, what happens when you're a lone founder? Yeah.

Yeah.

So what, what happens if you're, what happens when you're in this channel? Like what do, what, what does that look like?

It looks like a, a lot of conversations to try and make sure that new people joining the team get set up correctly. There's sort of this group of people who are trying to watch out for you in your first months. Also, when things aren't going correctly, that we communicate quickly to that person, find the right sidekick.

You have a good, like, setup too for people who are starting in their first couple of months, right?

Do you want to tell— we have pretty formal check-ins at 2 weeks, 40 days, and 100 days.

Yeah, but then they come and do like a presentation to you, right? Yep.

Yeah.

And they're like, here's, here's what I did.

Here's what I did. Yeah. Yeah. And then you're like, well, the first 2 weeks we just go through their interview process when they join, before they join the company. We say, this is, this is what made us excited about you and this is where we had doubts.

Yeah.

And we sort of give them the detail.

That's really good transparency.

I think it's—

Kat, we'll have to do that too.

People are excited to hear where people had doubts. I think it gets them excited about what they might have to prove and also where they may not have gotten to shine during the interview process. Yeah. At 40 days, it's like, right, you've been here more than a month.

Yeah.

Do you feel like there's something that you feel, you feel proud of? Yeah. And then it's the first time that we can start looking at signals and saying, hey, by the way, we, We thought this during your interview process and we told you at 2 weeks and we're still seeing it. So we're wondering like, is it something that we're not helping you with? Like on the, on the, on maybe on the constructive feedback side. And then at 100 days, 100 days, 100 days is a bit of a symbol in France because that's the number of days that Napoleon tried to get back in power for and he failed miserably. But the 100 days were pretty glorious. So the first 100 days—

really the inspiration for the Prohibition period, the number of days.

I don't know, actually, we should look that up.

But the Prohibition period is 100 days. The probation period?

The probation period.

Oh, sorry.

I thought you were talking about prohibition.

I was talking about the probation period. The probation period is 100 days.

The probation period in France is 4 months.

4 months. OK. So not quite just Napoleonic history.

Not quite.

OK. No.

A lot of the rest of the law is still Napoleonic.

OK. OK. OK.

But 100 days, it's like, all right, you've been here more than 3 months. If you're a salesperson, it gives you a full quarter. Anybody, you have more than a quarter of information. And so at that point, you're already sort of— it's the end of your spin-up. You're set free. Yeah. Best of luck.

You do this in the US as well?

Yeah.

Okay.

Yeah, yeah. Prohibition periods don't really matter. Doesn't matter. Yeah. No, it's not the same thing. Yeah.

But you still have the same structure and stuff.

Yeah. Yeah. So I think for hiring, I was going to ask you about go-to-market engineers.

Yeah.

How many do you have?

Well, we have 2 teams of go-to-market engineers. The normal role, which is the role we've kind of tried to pioneer around—

Did you invent it?

We did invent it. You didn't? Yeah.

Yeah. Is it trademarked?

No, we don't have that. Even Clay isn't trademarked actually. So we're really well behind on the— this is the happening. This is the trade-off with plunging into decision making, just-in-time decision making. You don't catch up to a lot of these things. But, um, yeah, so we kind of have two teams of go-to-market engineers. The first one is, hey, the, the, the, the role that every, a lot of these companies have, like OpenAI and Cursor, where, um, these are people who are using tools like Clay and Dust and other products to build revenue systems using AI and automation and scaling creative ideas and building really cool systems. And we have a team of 7 doing that across the— doing all sorts of automation work across the whole business. And then we have those people doing sales for Clay too. Yeah. So it's kind of like if Figma hired designers to sell, but if designers were more charismatic, basically. And so that's the approach, basically.

Do they, do they pass a proper engineering interview? Do you consider them as software engineers? What's the distinction and how do you help to explain it to people who are like, we're gonna also hire GTM engineers. Do we just hire them as engineers?

No. Though I will say there are multiple people on the team who can code and maybe could pass our software engineering test. So look, I would say generally what is the team's mandate? The team's mandate is to maybe write code, maybe use products like Clay and other products, to build like systems and automations that can generally grow the business, right? And to drive way more efficiencies and scale. And they do that in lots of different ways. So as an example, it's like, hey, they actually— events. So I'll keep running with this metaphor. We are kind of bad at like rigorously inviting sales prospects to events. So they actually ended up building a system where every kind of salesperson can actually go through all their deals programmatically kind of invite people. The invites will— well, another risk is another risk in events, by the way, is people invite the same person over and over again. And so it's the ROI is not as good, obviously. And so the system is built so that like you can't invite the same person within a certain period of time.

And when they come up with a great use case, does it automatically get like pushed into the product, shared with the team? How do you make those use cases then shine to your customers?

So this was a real problem because the feedback I give Osman, who who leads that team, brilliant mind, amazing, wasn't great at the feedback loop to product. So we hired someone basically to do that job who sits with that team. And this is also a very interesting hiring profile. I was like, someone posted, Samira on our finance team posted her profile on the channel. I was like, stop the presses. I like, I need to talk to her immediately because I haven't seen a profile like this in such a long time. Like she went to Carnegie Mellon, studied computer science, but was also the president of the mock trial team. Was also like interning at Blackstone and finance firms, which is kind of unusual. Then like went and was like a really successful multiple promoted engineer at Airbnb during their growth trajectory, but then quit all of that to go be the COO of a nonprofit in San Francisco. And so, um, and when I met her, she talked really fast, which I really liked.

That's surprising.

Yeah. And, and so she had it all, you know? And so, uh, and so we took kind of a bet on her. To do that, and she's like crushing in this. And that, that her job is basically this, like trying to make that feedback loop as tight as possible and build those best, the best things they create and innovate on. They're basically kind of, or their, their north star is bleeding edge of, of go-to-market and pushing the boundaries as much as possible and then pulling that back into the product.

What's something that you think is truly bleeding edge in AI-infused go-to-market engineering today that you think is like, if you have a company where you're doing that, you are in the top 1% of companies that are acquiring new customers.

I actually do think the event system they built is pretty sick. But another example of something that's like pretty sick that they built recently is they use, they use Clay to basically every customer success person. When you are meeting with your enterprise customers, you will create like a deck to analyze, to just like review, like a quarterly business review deck. They've actually automated 99% of that process. Because normally that takes hours and hours and hours to create and do. And they've automated it fully because they're pulling in data from Salesforce, they're pulling in all product data from Snowflake, and they pulled in data from all these different places to like fully create like a 20-page highly branded and customized deck. And then the person just has to come in, edit a couple of things, and it's ready to go.

Yeah, that's a pretty popular use case for us too on Frames and the ability to sometimes actually plug it into the live data.

Yeah.

So that if you send it and people only look at it the following day, it has that day's data and not the stale data from the day before.

Yeah, or just like, um, other examples would be like, how do you use, um, Clay and other products to kind of like build a, build an engine around support and like constantly, um, tracking the, the, uh, the support tickets that are coming in and like analyzing them for what, uh, what could be potential upsell opportunities. What are the questions they're asking that like could tie to something else and then like routing that. And basically you kind of like building an orchestration system on the backend to across every part of your business that can drive revenue.

I want to talk about Clay agencies.

Okay, great.

How do they work? How do you chat with them? Did it come about because people were saying, we've got an agency that's focused on Clay? Did you come up with the idea? How happy are you with it today? What would you advise people think about when they think about that kind of ecosystem?

I think with some of these things like the ecosystem, A lot of these things we didn't like come up with. They were happening organically and then we like noticed it and then poured a lot of fuel on that fire. Right. So with the agencies as an example, they, they were our first customers. Cold email marketing agencies were our very first customers. And that was kind of because they felt the pain most acutely. They had a lot of different customers, so they felt that pain a lot and they were entrepreneurial and technical so they could use the product. And then they would post about us on LinkedIn. Uh, and, and, and then that's how we kind of started growing actually. That was, and then we kind of enabled that. Um, over time it became really clear that because Clay, similar to Dust, very open-ended product. So people needed help kind of imagining the art of the possible and implementing it. And so we, and, and people were kind of pivoting their business, their, their agencies around Clay. What we were able to tell the story of and, and, um, is Unlike Salesforce and HubSpot, Clay's implementation can be driven, tied directly to revenue. And so you can charge $10,000, $15,000, $20,000 a month for Clay agency work. And that is such a revenue, like, case that you can go to so many other agencies and be like, hey, you should stop what you're doing and build an entire new practice just on this. And, and that's how we recruited way more agencies. And then we have a team that goes and enables them. And then we have a team that sends them leads. I'm curious, actually, it's such an open-ended product, like who your first customers were. And also, let's start with that and I have a couple of thoughts.

Very similar motion in the ecosystem. I think what we're seeing, not necessarily as focused on go-to-market engineering, it might be focused on operations internally, just general productivity, but a lot of people processes where people are really excited to see. Actually, I'm always amazed at the use cases where Dust agents are used as personal coaches for people joining a team.

Oh really?

Yeah.

Like, what's an example of that?

So people will have an agent.

I need tenured engineers when you have Dust people coaches.

Exactly. But for our customers who don't have tenured engineers, everybody gets a performance coach that will stock the last performance reviews and then challenge your goals, your calendar, and your plan for the week based on what you got told by your manager you are spending too much time on or too little time. And I think that continuous feedback loop and the ability, especially for junior people, to have a small, friendly tap on the shoulder on a regular basis like, hey, you were going to post this, but you've been told that you're sometimes a little long and not particularly concise and precise. Why don't I help you on that specific aspect of the internal messaging that you want to post? I think that agents to quiz people on product knowledge are always a case that I find really creative because, you know, traditional onboarding software will have a one-size-fits-all quiz that is just about as boring as you can imagine it. And getting people to get tested on a regular basis on the thing that is going to make them rich if they get to sell it well is a way for them to move.

Are we doing that use case?

No.

Yeah, no, we should do that use case.

We should do that use case. Yeah.

I think it's a really, it's a really powerful way for people to have a tailored to them experience.

Yeah.

Of how they can do better at the company.

Yeah.

Yeah.

No, I really like that because, um, obviously we're hiring a lot of people and, and learning Clay is a core element of that role. And, and we are thinking about, we have like a, bootcamp that everyone goes through, but how do you like enforce people and make sure they're learning?

And make sure it stays over time, because the product is going to continue to evolve and there are going to be things to learn about it every week and every day.

What are the other like cutting-edge kind of internal operations use cases where like you were asking me, like if you see it, you're like, wow, this is a team.

I love the transversal stuff. So, agents that will listen to call transcripts and directly feed feedback from customers to the engineer that is literally writing on that part of the codebase. And saying, by the way, 20 minutes ago, a call happened where this person said verbatim, I don't want the API to work this way. I think it should work that way. Click through, get to the data source, be able to talk directly to that person. Like just cut the middle people in the company that would meant, that would have meant it was going to be a meeting. And then the following week, product marketing was going to say last week, somebody told me that maybe the API endpoint isn't quite— no, it's real time. It's happening because you're coding on that part of the code base. Your agent is aware of you as an engineer and of all the calls that make that happen. The other way around, I think, hey, you're on the marketing team, the code just got merged, this is what it does. You don't need to bother an engineer to understand which segment of the customer base is impacted. Here's suggested messaging for them to be able to know. And by the way, this is the list of customers that had mentioned they were really excited about this feature getting, getting into GA. So again, all these small connections that aren't always by function and by team, but actually really represent how much of a mesh a really high-functioning company is.

Yeah. Did you, how do you feel about, like you guys have obviously stayed really horizontal and I'm sure over time you've gotten a lot of pressure from investors or whomever or yourselves to verticalize into one thing or the other. How have you thought about that?

I still think about it today. The struggle is there's so much potential in building an operating system for companies. And I think that's the dream. And I can't let go of that dream just because it seems like the technology technology is ready and the customers that we've worked with are ready and listening. And it says more questions.

Okay, great.

So that I think is what keeps me on the horizontal path.

Yeah.

It's harder to say.

Has there been moments where it was like, this is a disaster? Like obviously we should, and if there, what would you verticalize in if you were given a choice?

That's a good question. I actually don't know. I don't think that I would necessarily take the most used use cases of agents today. Because I think there's still a lot of faster horses versus cars going on in agent land. People are taking their job today, trying to automate bits of it, but doing it as if they were asking interns or associates or very anthropomorphized versions of agents to do it. I think that the future of better work at companies is essentially getting a lot more of that transversal and ambient stuff happening. So it would be hard for me to verticalize on a use case. Because I think the use case would be if you're starting a company from scratch, that would be the use case. That would be like you're 2 people and you want to never grow to the size where you feel you don't quite have the unity and the sort of single motion that you had when you were 4 people in a room. That's the use case that we have.

So how does the go-to-market work? Like when you— the sales motion work, not the self-service.

That's why I've been asking the questions on go-to-market. We have so much to improve. We try and ask people how excited they are about rewiring their company. And if people blink or sort of skirmish, we know they're not customers. But when we have founders or leadership team like light up and say, that is exactly what I want to do with this company, we know we have a deal. And then the product is built in a way that it diffuses quite broadly across teams. So when we check in a few months later, it's actually used across functions with best practices getting shared in a way that I think—

and how much enablement do you feel you have to do right now?

I think we're still doing a little more than we will as the product gets better, but we're learning a lot and also able to make the product better. So it's like it's worth it right now. We don't have product managers. Because I think any enablement is by people who are essentially product managers in the field.

Yeah, I think same. We waited really long until we hired any PMs. I think we're getting a very, very nice cue to shift over to Q&A.

Are there questions? And I'll give the mic.

We just have two mics, so I'm just going to go over and give you the mic if you have a question. Not all at once.

Any question?

Amazing. Thank you so much for— it's been nice to see your chemistry that you both have. But I guess my question is around automation. When do you know to automate a task? Because sometimes I know that there's certain rules that people say, like, you shouldn't automate something unless you've done it 5 times because then you know it's broken. Or sometimes it's feeding the loop of like, okay, we don't have all the information so then it's broken and then it's repeating. And then it's kind of just doing that loop. So I guess my question is, when do you know to automate something?

Yeah.

To keep it simple.

Well, I think actually, um, to, to make it a bit more controversial, I thought maybe let's use the 3 use cases you just told me about Dust. Yeah. And I, I kind of wanna give feedback on one of them. So 2 of them, the, um, basically the, the Gong product feedback loop. Yeah. And the learning, uh, and development use case. I was like, no-brainers, let's do this. Like that makes a lot of sense. But actually for the one around the, the feedback to, on, on, uh, performance feedback, uh, this is not a fully formed thought, so I'm self-conscious about sharing it. So, like, cuz it's very like, uh, but I'm, I'm like actually feeling what you're describing around should that be something that's automated? Should that be something that, because there's like, cuz I'm just imagining when you were saying that use case, I'm just like imagining me being like the, the person who is like, got this feedback. And I'm just like, I mean, I feel like it makes so many assumptions. First, it assumes that the manager's feedback was correct. Then in like, it assumes that the manager's feedback was written in a way that like makes sense and like is comprehensive and thoughtful, which, you know, I don't want to say is rarely seen in performance feedback reviews, but it's certainly not the standard. And then like, I feel like that type of loop could also have these trade-offs of really constraining people's creativity and forces people into a box. And I'm just worried about the— if you automate something like that, I'm worried about a lot of the unintentional consequences.

I think the unintentional consequences bit is exactly the element that I would worry about. I think humans are really bad at modeling complex systems. And when you automate a system that you don't fully understand, it has unintended consequences. And I guess climate change is maybe one of the bigger forms of when you do like the Industrial Revolution without thinking about some of the unintended consequences. To me, there's a difference between automation and complete autopilot. So I think that automating the suggestions, the nudges, and getting a system that is going to remind you to do your breathing exercises every morning or—

I need that.

Or your stretching. And that automation bit is fine. Now, if it's a black box that you can't contribute to, or it's your manager's feedback and you can't actually modify the agent's memory, or you can't challenge the agents and say, you know what, actually you've been telling me to act this way and I, I don't agree. Can you stop giving me that feedback in my, in, in, in, in the weeks to come? As long as you have an interface to make sure that you understand how it is going to change or be editable, then I think it's not a, it's not software that is imposed on you. It's something that you have a co— additional relationship with. So I think for us, what we try and ask is like, how well do you think you understand that system? And the, the support automation use case is often one where I feel many teams are automating too quickly. And they were like, well, we got rid of tier 1, uh, we, we're deflecting 90% of tier 1 and that's great. And I was like, but do you feel like the signal that was coming in from tier 1 is still felt as pain by the product team? Because the fact that it was easy to deflect with an agent doesn't mean that your product is right. And maybe it is cheap pain that you should be focusing on. The fact that you've sort of— you now have a way to no longer listen to your user's pain is actually removing the, the prodding that you should feel. And so maybe it is not something that you should automate right now. I think that— so, so those are my two points. One is automating complex systems is quite complicated. And then I'm all for automating the activation energy that gets you to do stuff, but I wouldn't necessarily automate the entire brain and, and put it in a box and then decide that The software is in control.

By the way, we see the same thing in go-to-market. There's all these AI SDR companies that automate a lot of the email writing as an example, but then take away a lot of the human insight. And we try to combine that at Clay. So maybe an SDR has a really creative idea and then that is the carrot or the thing that jumpstarts it. And then Clay can take that creative idea that's already working manually and automate the execution of it as opposed to constantly coming up with the ideas autonomously.

Thank you. So you told us before that your go-to-market team don't have, don't hire really some software engineer. You can have some people that can code a little bit, but they are not fully software engineer. My question is, in my team, basically all go-to-market engineer are software engineer too, because we think that if we using Clay and using some code, open source tools, etc., we can push the boundaries of the using that by using Clay or anything else. Do you think that it's easier to transform a software engineer into a go-to-market engineer, or do you think it's easier to transform a growth marketer into a go-to-market engineer?

I think that it is probably just as easy to do both for the thing. But the trade-off is— so for example, by the way, one reason it's not imperative for us to have full-blown software engineers doing all of this is if there is something that actually needs a certain level of technical expertise, why aren't we productizing it in Clay? If we hit some barrier that someone can't meet without being a full-stack engineer, that is a signal that should be productized in Clay, then we build that. So that's kind of why— it's not like why we do it, but it's one reason why we're not so opinionated about that. But I would wonder, basically what I would wonder is What we find is we actually don't interview that many full-blown, like full-stack engineers who want to be go-to-market engineers. And they actually find the problems that they're working on, maybe on different team, maybe on the backend or frontend or whatever that they're doing, they find them to be more interesting or more relevant to their career or they're more passionate about it. And I think it's more from us, like what we've noticed is more of a recruiting thing and where people are like, feeling fulfilled. And also we find more people who are coming from growth and ops backgrounds to be more passionate about this type of work and then giving them more resources and tools or whatever to do it. So I think it's more of a recruiting comment and then the product comment.

Hi, thanks so much.

Would you be able please to describe your typical sales journey and in which step of the sales process, which AI use cases drive the most productivity or impact?

You want to start?

Sure.

So we tend to focus on fast-growing teams, and I'd say our ideal customer profile is more of a mindset than it is a segment. The three elements that I think are really important are the leadership team is fundamentally optimistic about generative artificial intelligence and they, they, they, they need to rewire their company. It's imperative for them to, to, to build the company in the future. There's a lot of growth or ambition in the company, so they think a lot about upside rather than cost reduction, firing teams, letting people go. And I think that mindset is very different. And the third point is they have talent density on their teams, which allows them to see their, their team as a source of ideas and implementation. So they wanna build with their team. This seems to be on off, but, uh, they wanna build with their team, not against their team. And, and so for us, the sales process is then really talking to leadership very quickly, even if the first Workspace user for Dust is somebody that is not part of the leadership team, getting access to the beliefs the company has about what that future holds and how they're gonna build it and understanding them. Um, pretty much every step is AI.

Try to build.

No, pretty much every step is, is AI enabled for us. I think one, the one that I'm getting the most, uh, value from is, is on deployment and on getting from 10 to 50 to 100 to 200 users in a 300-person company. Because I think the granularity of the data that we can see in usage and the connection between the use cases they've found, the use cases they haven't found yet, the jobs they have, is sort of an infinitely explorable space to come back with it to them and have a great thing to, to, to, to tell them. So our lands are generally relatively modest compared to the number of active users we'll expect a few months down the line. And it's that post-sale— I'm sort of cheating on your question, but to me it's that post-sale period that is magical in, in how AI and the interpretation of unstructured data about how our users is using us that gives us a lot of, a lot of idea and feedback for them.

I think actually the models are, despite us being a usage-based business and you guys more seat-based, like I think we also have like a modest land and expand and we kind of set it up that way. And actually the team incentives are set up that way too. And so we are like, No one is at— like, we are not— that's why, by the way, we have non-salespeople doing a lot of the sales, because it's not— we're not trying to maximize the revenue on the land. We're more focused on skill sets like product expertise and enablement and things like that. To answer the question, um, you know, when we started the sales motion like 2+ years ago, there was a lot of skepticism about AI in sales. And so we, uh, actually originally The original sales motion was me saying, hey, look at these amazing things we can do for you. Look at how much we can transform your world, rewire your company, you know, like, like rewire your go-to-market. And it would kind of fall on deaf ears more or less. And, and, and people were like, yeah, yeah, yeah. But then it was like, it didn't like go anywhere, you know? And so we kind of pivoted a bit and we were like, okay, let's focus on a very specific problem that you have, which is data enrichment. And let me tell you, every big company has a data enrichment problem. Every company has a data enrichment problem. And, um, and we can like quantitatively prove to you that we will improve your life there. And we will also consolidate the vendors you're already using. And all these people had like an existing pain and an existing kind of budget line item for it. And so that insight and, um, wedge completely transformed our sales process. And we, we do tell them about the other stuff, but we are, we spend most of our energy on the data enrichment. And like the POC is set up around a free data test and audit we'll do for you, which we almost always win actually. And, uh, and so that is what the go-to-market is centered around. So we'll, we'll have a land just mostly on data enrichment. And then when we expand, like we have an NDR, you know, well north of 200%. Because you start with data enrichment and then we're with you, you have trust, like we're helping you, and then we can expand you to all these other use cases more methodically. Okay, I think we can take one or two more questions. Let's do the two in the back.

Where are you?

Hello. Thank you so much for this conversation. I have one question regarding the go-to-market in the age of AI. So you discussed obviously about software engineers and PMs, but what about the rest of the teams? I would say, especially sales and success managers, how their role will fundamentally change with the rise of AI, especially because some people would say that SaaS is dead. I don't think it will be dead, but definitely there will be some major changes going there, especially with the rise of agentics. So yeah, I just wanted to get your thoughts regarding this kind of roles and how they will change in the next few months. Thank you.

Yeah, I mean, I think that we actually see that our, I mean, again, so we have non-traditional people doing sales who are a bit more technical than your average account executive, but we actually see them using Dust, using Clay, Vibe Coding apps and Lovable. And we see that the skill sets that are prized are shifting. So it's not necessarily like, is someone exceptional at discovery or maybe even like multi-threading or something like that. It's actually like product expertise is getting more and more valued. And that, and then by the way, product feedback loops, like salespeople as mini PMs is like getting more and more valued. And so those are the skill sets that we see. So that, and then a lot of the manual work and research that salespeople do, tools like Clay and Dust and others are kind of automating away and centralizing to a centralized ops team so that people have more time to work with customers?

I think the, I think the skill that's still incredibly valuable is, is understanding the mindset of the customer and navigating the path of least resistance to bringing them value in a way that they understand, are ready for. And, and will really help amplify internally. So I don't think that's getting automated away anytime soon because actually the larger the companies, the more, um, the more important these decisions are to make, the more human-driven and almost irrational and trust-driven they tend to be versus the purely data-focused decision that almost looks like statistical arbitrage or real-time bidding arbitrage from ad tech. So increasing the portion of time and effort that sales or success or deployment or forward deploy engineering teams spend to understand and communicate that they have understood exactly what customers need is I think the bit to keep and the rest should be valued at almost zero. So I, I think in the same way that we've had generations of companies that communicated on what we came to call vanity metrics. I think the vanity metric, uh, world is, is pretty ripe in the workplace with people who are taking pride in actions that should actually have been completely automated when almost all of the focus is on driving the ability to understand what the customer needs and making sure they understand the language you spoke when you said you could deliver for it. The number of calls, the number of emails, all these metrics that are essentially upstream or downstream, the number of follow-ups, how up to date the CRM is, uh, how many meetings you took with products to share the feedback. All these things are not going to be where you should spend your time. And so I, I, I'd argue that the best people in those functions are already pretty good at that, and they may actually shine through even more visibly because we're going to get rid of some of the equalizers that we had in companies. Where the fastest-moving team members were not able to show just how much faster they were. So I think we, we had this expression in software development of the 10x engineer, a bit of a myth, but I think a useful, a useful concept to say like two people typing on keyboards, that the amount of impact that you might have from one versus the other is a 10x ratio. And I think we're going to see 10x people in just about every function. Because we're going to shed away the amount of rote, mundane, not useful work that was imposed on teams that were less technical.

Should we do the last question though? Yeah.

Okay.

It's just for the next one. What are the bad— the bad lunch? Hi everyone.

I thought I'm going to be forgotten. Sounds good.

So my question is more about how you guys see the organization of AI in the future and how it looks like in your both teams today. Sorry, say that one more time. How you see the AI organization in the future, meaning your next hires, what are the next jobs, and so on. It's here for you. This is part of the Rewire.

I'm trying to come up with a smart answer. I think it'll look surprising. I think it's just going to be a distilled version of what highly functional companies today look and feel like. Right now, you know this expression like the future is already here, it's just not uniformly distributed yet. Some companies today—

Microphone is on.

Maybe it's just my head. Some companies today work in teams where there are no useless meetings, there is no useless hierarchy. People actually get to focus on work because they don't have to deal with politics. And in the same building, in another company, there are people who spend days and weeks preparing for a slide share to look smart at a presentation because that's the only way your boss gets to see your work. And then they live, they're in the same building and then in the same street in Paris in 2025. So I think the organizations of the future are just a lot more like the organizations that are highly functioning today, where you've cut away at a lot of what was essentially a bit of a waste of time and focused on what are we exactly, exactly building? How do we find more people who understand exactly what we're building and how quickly can we make the product better so that the people who understand what we're building value marginally more what we've built for them? Uh, so I, I think that. It will look like fewer processes that are essentially information pushing, because information pushing isn't going to be needed in a world where the software adapts to your needs. The software remembers what you forgot. The software tells you what to pay attention to. If you can imagine the best possible dream version of software that just makes you feel like you're like in a cockpit in Star Wars instead of feeling like you're a typewriter in the 19th century, like all of a sudden you get to not have to understand how a piece of software works because it automatically adapts to your needs. It's not asking you to fill in the same form as the person next to you because not all of the information is useful. And so that sort of goes away. And what you're left with is just pure, unadulterated customer focus, customer acquisition, and the feedback loop to make more of them happy.

The only one other thing I'd add to that is that just a little bit of caution. Around AI-first organizations, just because, I mean, you even brought up one earlier, right? The unintended consequences of support tickets and deflecting all of that and then not feeling that pain. You were just talking about coding and not understanding the software that you build. And we notice that where people use Cursor for something and then don't understand the code they just written and it's harder to debug. We ask our marketers to not use AI for writing because the whole point of a large language model is to synthesize and give you an average, basically. Marketing is all about not reverting to the mean. It's about standing out. So I think your question around what is the future of an AI artist, what does this look like? I think there is a real kind of push right now where everyone's like, use AI, default to AI or die. And there's these CEOs of Opendoor and Shopify and whatever, they're like, especially at bigger companies, they're like, you have to do it. You have to default to everything. And I understand why they do that, but I just, I just think there's a lot more caution on this. And you'll lose a lot. Yeah.

I also think it's, we're just going to stop calling it AI very quickly. We'll just go back to calling it software.

Yeah.

And so it's just like the software just got better. Now let's get back to building great companies, having a story about our brand and our company. And understanding how we differentiate from the others. And so we didn't quite pay as much for the.tt as you did for the.com, I think.

Uh, although, uh, I think it's like we were clear on run for a long time and it was like a very playful brand element.

Yeah.

So it's good.

But the reason we didn't want to be dust.ai is because I actually don't think that the AI is, is really important. I really don't think that people care if it's AI. What they care is that it works and that it allows them to do new things. And so that phase is probably going to get over quite quickly. Growth hacking.

So we did have a thing on growth hacking, but I do just want to like, I can share what we were thinking, but also want to just keep, I don't want to like ruin the flow of the questions that we have. So we'll make an open offer and feel free to take us up on it, but also feel free to just keep asking questions if we can keep engaging. But we kind of had in mind, you know, Gabriel, obviously an expert in AI rewiring. And, and so if you, if you wanted like feedback on your own business on like how you could use AI to rewire your business and, and through the internal operations or, or, or go-to-market and like you had a growth problem that you were really struggling through, we would be happy to like live hack a solution with you and thought that would be entertaining and fun and hopefully instructive and useful. So that's going to just be an offer that we'll make. Feel free to take us up on it. Also feel free, we can just keep engaging and have more questions and that could be also fun too. So yeah, so we can keep going or not.

Would you like to do something with an example?

Yes.

Yeah.

No, no, no, no, no.

You guys would ask like your actual business, your actual problem, and we would try to help think it through with you live. That would be the idea. So like for example, it's like, Hey, I have this business and I'm struggling with people aren't showing up to my demos and why not and what can I do to solve that or just coming up with something. But yeah. Okay.

Wow.

Okay.

Let's just go back because I think people ask questions.

Thank you. Hi, um, can you hear me? Yeah, yeah, perfect. I'm in the manufacturing space. We're building foundational models for factories, and over the last couple of months, the challenge for us was a lot of our customers, they're not active on LinkedIn or some of these platforms. So in terms of AI tools, how can we utilize them to, to reach out to some of these decision makers? Because The way we are go-to-market at the minute is trade associations, trying to partner up with them, getting them to set up a showcasing, like a private one, get some companies out of their network to come. It is a long process. It's taking us 6 weeks, 8 weeks before we get the first conversation. So anything to accelerate that?

No, keep the mic. Do these people only come to trade conferences? Like, they're not on LinkedIn profiles, which makes sense because they're in manufacturing. So that tracks. But I'm curious what else they do where you can see them and find them and engage with them. Is it just conferences or anything else?

Well, they have LinkedIn profiles, but they're not active, as in, you know, they would see the message, but that's not their space.

Sure.

But they presumably would respond. Would they respond to an email or phone or something like that? Or—

They do, sometimes they do respond, sometimes they don't. That has been a slow process. I mean, we did the same when we started LinkedIn ads.

How big is this? How many people are we talking about? Like, what is the quantum? Like, is this tens of thousands of people, 100 people? Like, what's the quantum we're talking about?

So I'm based in Ireland, so we are just sort of like using the companies in Ireland.

How many are there?

Within this, within our ICP, there are roughly 800. So the specific ones that we are targeting. Manufacturing in general, it's hard. So 800.

How much would you be willing to pay to acquire one qualified meeting?

Well, with the trade associations, it's free at the minute.

Imagine I could get you a meeting with a decision maker of one of those 800 companies, but I'm putting a price on it. How much are you willing to pay?

On average, if you do like a trade show with one of the big events, you're looking at £300 or €300 to get one of those conversations. Which might lead into a potential meeting. So that would be 300,000.

So, um, yeah, so I think this is just like a different go-to-market approach, right? Um, in this situation, I would use products like Clay or Dust, not to like automate the outreach necessarily, because it's just 800 companies. It's very small. Um, but I would use tools like this to, um, help you come up with something very custom, much faster. Uh, and, uh, and instead of it taking days to come up with some very unique app or whatever for using your large language model, your foundation model that you've created. You could create it very quickly, like in an hour or something using these products. And then I would spend most of your time and energy, now that that's done, in making that really land in a very special, authentic, creative way. We're just talking about 800 decision makers. So that is something that you can do very manually. And you could go to all of them, You know, you could just visit them. You could come up with some very physical item or gift or something that really stands out. Um, you could curate an entire community just for the 800 people, you know, and like, uh, they only show up to conferences. Maybe you create like a, I'm just making things up now, but you could just create like a, well, maybe not that, that, that could work, but they're not that online. But like the, um, but maybe you could create like a way for them to meet up with each each other at the conference and you become known as like the matchmaker. 'Cause all these people, uh, human connection is a connection is a human thing, right? So they're going to wanna network and do stuff when they're at the conference. Maybe you, you're the person who helps bring that all together, right? Um, you know, maybe, uh, you know, so I'm just saying like that is where I would spend your energy and then I would use like AI software to just like have the feedback loops be tighter.

Okay.

Makes sense.

Thanks.

Thank you. Physical, physical, physical. Product samples to each and every one of them. I think that some of these businesses have— some of these more traditional businesses have, um, like the manufacturing function just doesn't see itself as like needing to interface with the world that it doesn't need to interface with as much. But what they do interface a lot with is vendors, like vendors and providers, quality assurance companies, companies that are in the, in the flow of making them make sure that they get the right ingredients to build the right products that are at the right quality and certification level. And those types of functions end up being easier to tap into that are not necessarily the engineering floor or the manufacturing floor, but you can raise awareness of your company by taking part in, um, a roundup of the Chinese vendors that are under-indexed on for this particular industry, or like articles and content that essentially get you in the supply chain conversation. In front of eyes on the legal team or on the quality assurance team that isn't necessarily just the manufacturing team, but that are going to talk about it at the meetings where these people that you might have a harder time reaching are going to be present.

Hey, thanks for the great conversation. You had a great point on the convergence to the mean in what companies look and sound like as AI is now in everyone's hands. And I think like overnight all companies became AI native and at this point all of the companies' marketing starts to look and feel a lot like the same. You guys have done things quite differently. I would be super curious to hear about two things. One, your inspiration about behind the campaigns with like joint rolling and kung fu movies with your customers to then the second point, maybe more also to the both of you is like, is there a way to find paths to rewire almost your organization to be able to think and be more comfortable with kind of trying to be out there in a different way that you don't sound like 90% of your competition on the market at the current time.

Yeah. So, um, two things on this. One, on the second point first, it, uh, it has to be top down because let's just take the joint rolling thing or the kung fu thing as an example. Um, so actually the answer to these are both very connected because it all comes down to the people you have in the company and, um, 25% of the company and half of the business teams are, are, come from very non-traditional backgrounds and, uh, are like, are doing the current job for the very first time. So that is what enables the very creative thinking. And I'll give you some examples of this. I'll give you two examples of this. But to your second thing of like, how do you enable the organization to do it? It has to be top down because otherwise people won't be able to think creatively and freely. Like I have so many people at the company are like, wow, this is the first time anyone will listen to my crazy ideas. And like, like, let's take both of those as an example. We hired a more seasoned and traditional sales leader for the very first time. And, you know, they were upset actually, both of them. And this is the first time because like, remember, the whole sales team is non-traditional salespeople. But like now we have— and like the blunt rolling thing, it's like we hired Snoop Dogg's like blunt rolling guy to teach like all of our customers how to, you know, roll a joint. And like, and then we like videotaped it. Or, and then we hired like a Marvel filmmaker to come and do a kung fu fight scene with all of our customers, like OpenAI and Canva and stuff like that. And, and, you know, when we posted about like, especially the, the, the weed one, like it definitely like was spicy, you know, and it definitely, people were definitely like, and, and, and it's specifically a lot of our salespeople, even though they're non-traditional, were really scared about inviting their prospects to it, really scared about the pushback. And so despite them not coming from traditional backgrounds and their leaders, so you really, you need to be top down. It's like, no, no, This is who we are. This is what's going to happen. And like, you need to be convicted of that. The first point, I think it comes down to the people you recruit. So I'm going to just give you two examples of people who are doing this because these aren't my ideas. These are their ideas and how their backgrounds can help inform that. So Peter, who runs our social media, you know, doesn't have a social media background. Peter actually was an investment banking analyst at Barclays and did finance at, at Clubhouse, the social media app. Um, but on the side, he had a TikTok that had 200,000 followers about spreadsheets. And so when I found this, I was like, wow, imagine finding someone who has the analytical and operational brain, but also the same brain that can get 200,000 followers making TikToks about spreadsheets. I don't even wanna watch a— look at a spreadsheet. I run a business about it, you know? So like, um, so, so that was crazy. And, and then I'll take, Jessica is another example. She comes up with a lot of our crazy ideas. Jessica— this is going to be a particularly American example, so just pardon me for a sec. But basically, Jessica, she went to school in Texas. And in Texas, they have a law that says that you can have guns publicly. This is the American bit. But they had another law that said that you can't have sex toys publicly. And so she thought the dichotomy of that was ridiculous. So she started a protest and a campaign called cocks not glocks. And she basically had thousands of people marching down the streets of Austin, Texas, carrying dildos and like saying cocks not glocks, you know? And then she ended up having a career in political organizing actually for many years until we met and kind of pulled her into viral marketing at Clay. You know, when you look at that in political organizing and stuff like that, ultimately what that is is marketing, but not in a B2B context, you know? And so how do you— and this is what the interview process is designed to— like, how do you understand who a human is, who a soul is? And like, you know, and I've known Jessica for a while and she actually felt a lot of pressure to kind of be more conforming. And we actually almost hired her 3 years ago when we were just like 8 or 9 people. But even I thought that was a little irresponsible to have a full-time person who just to just schemes, you know, when we were like 8 people. Even I thought that was a bit of a limit. But, and then she was like, she really wanted to perform more. So she was like, okay, maybe I should have real skills. Like maybe I should be a product manager. So we, we talked and I helped her get a job, like to be a PM at a company, boring. And, um, and, and, but she, but she wasn't, she wasn't happy, you know? Um, it wasn't like really fulfilling her creative self. And so we talked and I was like, what if, like, what is your dream job? And she was like, I just wanna like come up with fun ideas and scheme. And I was like, what if you could do that full-time at Klei? And then you need the top-down conviction to be like, we're going to have a full-time person who just comes up with wacky ideas and executes them. And you need top-down backing to make that happen because she runs into a lot of pushback even today. Kat's going to be working with her, and I promise you there'll be pushback from Kat. You have to have that. But you need the people with this background and top-down enforcement to kind of make it happen.

Awesome.

Thanks. She has been waiting very patiently though, so we should come back to her. We can't forget her.

Hi guys, pleasure to listen to you so far. My question is about buying signals. So we're a software company, we do real-time tracking visibility, and we sell to big enterprises. The challenge of, you know, long sales cycles and then the deals, they slip from quarter to quarter and so on. So we, I was thinking about buying signals, like we, we often, we don't know how hot are those companies. We know that the topic is hot, but we don't know when exactly is their, their buying decision or when, when exactly is that, how to capture that buying signal with its that's, that's when they want to buy it. I mean, you can receive a tender invitation, that's, that's obvious.

Can you say what your business is? So your product? So I, um, maybe I said at the beginning, but real-time tracking visibility.

So we, we're a software company that does real-time tracking visibility. So we aggregate a lot of data in real time and we help people visualize all their assets around the world, uh, they're being that are in transit and give people ETA on where's all their stuff.

Like, kind of like, like if you're like shipping companies and stuff, like where—

Imagine, imagine you're building a car and you have 3,000 pieces around the world all being in transit going somewhere to your factory. And we can tell you exactly where is each of those pieces and when is it going to arrive or not arrive.

So the buying signal for that product is like a moment when something, some disruption happens. So for example, would a buying— I'm just making it up— but would a buying signal for this be when an earthquake happens in some part of the globe? Or some— like in Venezuela right now, you can't fly there anymore. I think it's the boats that are dangerous. All right, whatever. I feel like I saw a Trump tweet about airspace. Yeah, the airspace is closed in Venezuela. So that would be Wouldn't that be like a buying signal? Because if some company was relying on something in Venezuela, that would be kind of, they would be kind of like—

It could be, but this is where I'm curious, like how would, how could Clay help with that? Like we have 100 customers, we have, let's say, 100 prospects. How to know which one is now, you know, in the moment to buy? Like how would you, could you enrich that data?

Yeah.

To actually help us. Know which ones are, you know, they're serious about it and which ones are, well, we're going to chase them forever, but eventually we're going to get lost in the board meetings and so on.

Well, we just have to define what these moments are. And so if we— let's just play along for a moment and say that you agree with me that like that an earthquake or Venezuela's airspace being closed or something like this is like a material disruption event in the globe that would affect a company's ability to get parts out from there or whatever. And it sounds like, okay, it sounds like there's some alignment on that being like a disruption event. What I would do is like, okay, so you have 100 companies who are your prospects, and then you could use Clay to track them. And for each one, we could try to understand like what matters to them, right? And so for example, the Venezuela thing would only matter if they have people in Venezuela, probably, right? Um, or like with the war in Ukraine, it would only matter if they hired, they had engineers or people in Ukraine. And so you could use Clay to like look at their employee base and be like, where do all these people work? Where are they from? Like what's happening? You could scrape their job listings and their job postings to be like, is there unique data points about locations or something like that in there that gives you some insight? And so you can use that. And then once you have that, okay, here are the hotspot areas. You can use Clay to constantly track those areas. And when news comes out or a tweet comes out or something that says disruption is happening in some way and you can use AI to analyze that to determine if it meets your guidelines of what disruption really means, then that's a trigger for you to reach out and do something.

Thank you. I'm trying, I was trying to understand if there are assets that you can track before signing the customer, because it seems like a lot of the, the dealers are going to be tracking all their assets. And if there's a way for you to track some of their assets without having sold them the software yet, they get a lot of value that it's hard to not click on.

I mean, this is something connected to, yeah, to what was just being said before, because we could track, for example, any issue in the Suez Canal. We can track that. So if there is a problem there, we can track piracy alerts, we can track all kinds of ships that go through or port issues, specific port issues. And then we could link that to specific companies that are— have a lot of operations there or have a lot of distribution or supply chain there.

Because I think that, yeah, upstream from buying is intent. And so the intent is probably like, I— in being interested in, in the kind of signals. And that seems— the content that I would click on if I was responsible for knowing all the pieces were—

I would look on the con—

I would look at the content that is telling me that there's a chance they aren't where they should be. That seems like clickbait content for a person whose role it is to track those assets.

It's just complex because the people tracking the assets are not the people making the buying decision necessarily. So we need the C-suite. We need to get to the people that actually decide then to improve their supply chain visibility.

Let's just do a couple more questions, then let's come back to her.

Hi, so we are, we are— the pressure is on. We are a B2B SaaS company where we give visibility to IT managers into on and offboarding and like license optimization. So our ICP is an IT manager. This is a very saturated market because they're constantly being solicited as an IT manager because, yeah, all the time. So it's been really hard to break through the ice of like cold calling and like email outreach because it's kind of like the same thing. People are doing the same job signals, like the job posting, like the HR signals and other signals like that. And we've been having a really hard time because it's the first time that the sales team is scaling. Like it used to be 2 people and now it's 5.

Yeah.

And like, it's the first time that we're like structured, like making structures or operations or processes as in sales team. And I guess one of the big problems at the moment has been differentiating ourselves with the data. When you have somebody on the phone, how do you know to make it as personalized as possible, but through volume? And actually, we've been using Clay for data enrichment just to make sure that the data is as precise as possible. So that's one of the problems. And then the second problem is our buyers, they're usually like 40, 30, 30, 40 years old and they're not like on TikTok, on Instagram, sometimes on Reddit because we have some inbound from Reddit. But like, how do we differentiate ourselves with our marketing to make it seem like we are a fun company? And it's like talking about IT is fun, even though that's not really ours.

So a couple things, just to be clear, like I don't think that So a couple, a few thoughts here. First, I don't think you have to market yourself as a fun company. You have to market yourself in a way that resonates with your buyer and, um, um, or in a way that feels authentic to you and you don't care what, like, you know, like, like it doesn't, it doesn't have to be a fun company. It just has to, um, resonate with, with your buyer. Right. And I think by the way, Dust and Clay both have like self-serve motions that are, uh, very mass appeal. And so very colorful and playful brands are more necessary, or not necessary, but certainly have a longer impact in that. But with IT, I'm not saying that doesn't necessarily have to be the same prioritization. Going back to the first question, I'm just curious, basically my mind is thinking about product problem or sales problem. And my question is, what is your differentiation of the product? It's a saturated market, like you said. Makes sense. Why do you win deals?

At the moment, we win deals because we are cheaper than the other solutions.

Okay, so that is, I would say, a problem, uh, in that, um, it doesn't have to be a problem, but I would say the— if that was why we were winning deals, I wouldn't scale the sales team from 2 to 5, um, because that doesn't seem like it will be sustainable, uh, in the long run. Like, I think you will need a clear value prop because you're probably not going to be able to generate the pipeline, which you're seeing, by the way, right now to support 5 salespeople on the quotas and OTs that I'm imagining they have, if that is your distinct value prop. Um, and so this might be more of a product question, um, than it is a sales question. I'm curious if you have other thoughts.

Assuming you haven't tried already, I think that community tends to communicate very freely, especially because of the security aspect of their role, on very dedicated forums. And so I think the meta problem that I would try and crack is how do you get into the WhatsApp group as a brand and name that people talk about on these super high velocity WhatsApp groups and Slack channels of IT managers by, by, by city and community, by stage of company or by industry. They, they, the one thing they will get fired for is if a security vulnerability that they are responsible for gets noticed by the CEO and not them. So the one thing they are incredibly good at paying attention to is what are the networks that are going to tell me first if something I should know about is about to happen. Those groups tend to be really tight-knit industry groups. And so what can make you a thing they would want to talk about? And is it price or is it something else? I kind of agree that there may be a differentiation that has to come back to why they felt compelled to mention it or why they feel that if they ask the question, has anybody seen— I don't know what the name of the company is.

Korma.

Korma.

Yeah.

So if somebody asks in the WhatsApp group, hey, I just got a ping from Korma, should I check it out? Has anybody heard about it? What is the thing that the WhatsApp group is going to respond going to be that is in your interest? Because if it is, oh, it's cheaper, you're actually sort of compounding the problem that you have to differentiate based on value. But if it's like, setup was incredibly easy, support is amazing, I keep finding like they are on top of the aspect of IT management that you think is an edge. Like everybody has a vulnerability. And I think the emotional vulnerability for IT managers is fear. It's fear that they're not going to be the one to spot the thing that's really, that's really a big problem for the company. And that's a really strong marketing emotion to land on because it's one that keeps people sort of anxious. And awake. So how can you positively hook into that by providing either, again, content that associates you with being incredibly good at finding the thing, or to some degree, something that is taking a thorn out of their side on the fear that they have that they're not going to be the first one to know. So I'd really go deep on the emotional state of your buyer on top of the differentiation of the product.

Super helpful. Thank you.

Okay, one last question.

So we offer an observability solution more geared towards mid-market enterprise. And so our buyer persona is going to be quite high in the hierarchy, meaning CTO, head of engineering, head of infrastructure within an enterprise company, which means someone who's pretty hard to get to. And the playbook for years in observability has been outbound. It's enterprise, so it needs to be outbound sales-led. The issue we're seeing, or I'm seeing at least, is that these people are getting harassed every day and it's getting worse and worse. I was speaking to a CTO yesterday who was telling me he gets anywhere from 10 to 50 emails a day. He never reads them and barely reads the title. And yet when we end up speaking with them, they're usually interested. Like he himself, once we got him IRL at an event, was super interested in what we had. And so my question is, when you do outbound—

why, why was he interested in what you had?

Because Datadog is crazy expensive as you grow.

Back to price is the differentiator.

But so how do you get to those conversations? The problem is outbound seems harder and harder. Even when you do qualified outbound and you reach out to the right people, they just do not open emails, they do not open LinkedIn messages. Do you have to go back to IRL or are there any other growth hacks that you can do to actually initiate those conversations with those really hard to get people who everyone wants to speak to?

So I can give you growth hacks. I can just give you one I learned yesterday actually that I've been told works every single time, but it's not going to be a— I'll give it to you, but I don't think it's going to solve your problem, right? Because it's a deeper problem you have. So like one hack is like, we're in Europe and so everyone's maybe in a different country here and maybe they're in France or Germany or Greece and you could do your outbound in Greek and you could do your outbound in Dutch with a Dutch signature. And I've been told that that really works and very few people do that in a very tailored way. And so that is, as an example, like a growth hack that probably will work for a period of time to maybe get meetings. But I think that is probably just a temporary advantage and not a permanent advantage. I think more broadly, one other issue in this market is obviously it's very competitive. Obviously, Datadog is the dominant player in this market. Another reason this market people are probably not, people are probably not responding is, is I'm just also guessing that the brand presence of your company is not that high. And, you know, brand and performance marketing are very related and you kind of need a great brand so that people have the awareness to then respond to you. Not just the value prop makes sense, but the awareness of the brand to respond to you. And so So, you know, there are like situations where price in this market could really work, right? Like for example, there was a period of time where all these companies popped up around like helping you with your AWS bills, right? And they grew up like Antimatter and others, like they grew really quickly, Vantage, others, because they were so expensive. And so it might be interesting to study like what they did that kind of resonated. I'm also just trying to think like what are other ways that you can kind of stand out with these buyers kind of originally? Like CTOs in particular are a notoriously hard kind of group to penetrate in some ways. And usually we find that people, that email is a poor channel for that market. And something much more human like dinners or in-person events or or again, WhatsApp groups or things like that are much more connected. You could obviously try to simulate that with, with videos or text messages or WhatsApp messages or something like that. That could kind of simulate that. But yeah, those are kind of like a few thoughts.

Are you trying to get your next 10 customers or your next 1,000 customers? Like just an idea of scale.

Um, let's say 100.

100. Okay. So I always like, one of the things that I feel is interesting to think about is like when you can reach people in a situation where they're not the, the, the sort of apex predator in the situation. So at relatively non-technical conferences, technical speakers end up being tier 2. Generally they're not on the main stage, they're on the technical stage, but in the speaker's lounge they are like as present and as easy to reach out to as any of the, the, the, the main speakers. They're much more amenable to conversations because they're not generally as the, the, the, as mobbed. And so I'm wondering just to look at scale, the, the types of events that some of these people will actually speak at and getting speaking engagements at those events to actually have a founder-led motion for a little longer than you, you would've naturally felt inclined to do. So I think there's just a ton of trust in the founder-led motion right now that actually cuts through a lot of what people are perceiving as at-scale marketing where they just literally don't know, am I talking to a bot? Or a salesperson, right? But, um, I, I think there's a lot more to do in how you could actually scale and extend the founder-led motion in ways that feel like you're not getting pitched when you're obviously getting pitched. But it, it is the curiosity thing. If there's, uh, also parts of their repos that are either open source or open to contributions, I think that CTOs cling on to pull requests and comments on their codebase a lot longer. Than they tend to admit. And so I know for a fact that anything that is published or that's commented on our repo, like my co-founder will immediately latch onto and read within seconds. If you send us a vulnerability disclosure, a bounty request by email, probably doesn't care. But if you actually post something on the code with a pull request, he'll be on it within 10 seconds. And so just finding people in the channels where you're talking about craft rather than talking about business is just an angle that I think gives you a bit of an edge.

That's, uh, all the time we have for questions. However, before we break for networking, we, um, are going to play a little game, and I hope that you all will— Varun has no idea what's about to happen.

I definitely know what—

Yeah, Gabriel does. Um, so Gabriel, if you could help us by putting this on our lovely friend Varun, we're gonna do a blind taste test of some of France's favorite offerings.

Wow.

Take it, guys.

Varun famously—

Yeah, I'm going to be sipping. Okay, great, great.

It's not just me. I'm going to be serving the tasting.

Wait, you're going to blindfold them too?

No, no, no, he's going to feed you.

You're going to feed?

Yeah.

Okay. Yeah.

All right, great.

Okay, we're going to do something to eat and something to drink.

Okay.

I know that you, I know you know how different European countries are, and so you're going to obviously nail this one. So we're going to taste macarons.

Okay.

We're going to taste two flavors of macarons. Okay. And you just have to guess the flavors, and then we'll have one sip of wine and you're going to have to tell me if it's red or white.

Okay.

Okay.

That seems, that seems easier.

That seems easy. Okay.

All right.

We're going to, we're going to start with the macaroon here. Just like, there you go. Just give me the flavor of that macaroon.

I think this is going to be too easy, so I think you guys should just shout out random flavors that it's not to confuse him.

Okay, let's go with pistachio.

Pistachio.

All right, great.

Honestly, I was gonna guess chocolate and pistachio, so I, I'm sure I got— I was gonna get that wrong.

Okay, now the second macaron.

These are not quite macaroons. They're almost the same as macaroons, uh, but not quite the same. So just have to give me the flavor of that.

But at least I'll try to be vulnerable with, um, yeah, wait, what's this? Yeah, no, what is this? What am I putting in my mouth?

It's definitely edible.

A similar type of cookie.

Yeah.

This is not good. Really bad.

Really bad? Really, really bad. But— I don't want to eat that. Okay. Well, you don't have to eat more of it. I'll take care of the rest. All right. So that was not pistachio. Do we agree? The good news is we can flush it down with a glass of wine. And it's gonna— it's either red or white, and so you just have to tell me which one it is.

Red.

Yep, absolutely, definitely red. All right, Varun, congratulations! Thank you so much, well done. It's definitely not red, but you really nailed one. So, uh, but when Varun talked about his Tour de France, he was very excited to come and he was very excited to not be stereotypical about France. And so we obviously, we take things very, very seriously here. And so in the name of the go-to-market audience of France, I hereby name you French. You're welcome back anytime. All right. Thanks for coming, Varun. All right. Thanks, everybody, for hanging out with us. We're going to have nibbles and drinks and Varun and everybody else. So please— and bread and cheese, goat's cheese, delicious goat's cheese. So yeah, thanks for coming by and let's see you around.

All right, thank you so much.