The Transaction

TT - 062 - Sydney Sloan - Full Episode
===

Sydney Sloan: [00:00:00] 80% of people are using, AI search as part of their research process. 80%.

Buyer behavior has radically changed in the last six months.

We get to throw away the old playbooks and create something brand new.

If you haven't yet started thinking about how do you show up in the LLMs you're behind,

Reviews and the voice of the customer and user generated content is really what the LLMs are pulling from.

It is all about eyeballs at the end of the day.

The most successful way to generate your reviews is actually in the app.

The move from SEO to what is now being called GEO, generative engine optimization, is a shift

We used to write our content for Google. Now you actually get to write your content for people.

More and more we're looking for the human connection. People want to be seen and heard and know that their feedback is valued and taken.

Matt Amundson: we are hitting our SNL moments where we're having our repeat guests. Uh, Sid is, uh, starting her second podcast with us. Well, on her way to being a five time guest, although Scott with what? Three or four plus your, your guest [00:01:00] hosting, you're, you're turning into our Steve Martin.

Scott Albro: It's an honor. I,

Sydney Sloan: I actually have my scale. I have my scale jacket already, but I take another one. It's like when I'm freezing in the morning. Yeah,

Scott Albro: Oh, well, that's nice. Sydnee. I've, I've never gotten

Sydney Sloan: have to get in.

Scott Albro: Thank. Yeah.

Matt Amundson: boo boo.

Scott Albro: yeah, No, it's great to be back. Has, has it been three or four times? I can't it, it hasn't been three or four times.

Matt Amundson: I think you've been a guest three times and you've been a co-host. Now this is your second time,

Scott Albro: second time. All right,

cool. Five timers, club. There we go.

​

Matt Amundson: Alright, awesome. Well, I'm joined not by Craig today, but I'm [00:02:00] joined by my guest host Scott Albro, who you all know and love. And if you're not following on LinkedIn, seriously, what are you doing?

Uh, because he is a, he is a rabble rouser, uh, and he knows how to get the people going on LinkedIn, which we love. and we are joined for our second time with Sydney Sloan, CMO of G2 crowd. So super excited to have Sydney on Sydney. Man, we were at Saster together a couple weeks ago.

You put together some amazing statistics in your presentation. I'd love to dive into that. Uh, and then you led the, the CMO session, uh, which was awesome. Love to dive into that. but let's, uh, let's kick things off here. So, Sid G2 crowd, it's amazing. I'm a customer. I've been a customer

Sydney Sloan: just G2. You're OG it. They, it was rebranded a while ago to G2,

Matt Amundson: My bad, my bad. Okay. All right. It's G

Sydney Sloan: some link somewhere, like on an inside thing. I see crowd every now and then, but it's just G2

Matt Amundson: I'm, I'm gonna, yeah, I'm

Sydney Sloan: now. You call it G2 AI if you want. If we're [00:03:00] like kipping, really?

Matt Amundson: Ooh, great. Well, I'm gonna call it G2.ai. So, uh, congratulations on, on joining and all the wonderful things that you're doing there. Tell us a little bit how, how's it been so far?

Sydney Sloan: Um, I mean, two-sided marketplace. That's the first time. Uh, so really thinking about like how to drive, you know, buyer traffic and engagement and also creating value for our seller side customers. Um. I think the best part in the last year, because it's been a year now that I've been here, um, is just being part of the explosion of AI and like trying to figure out how to harness that and organize it and create, you know, insights that are meaningful from the reviews that we capture.

So we actually are asking specific AI related questions and getting AI insights from that, and then being able to be part of. Hopefully the teachers, uh, that are inspiring, like we get to throw away the old playbooks and create something brand new. And, um, and the AI first companies are getting that by default, and the rest [00:04:00] of us have to force ourselves to do it.

And so, you know, being, being in the forefront of the, now what I'm calling segments to Signals revolution, uh, is, is, is, you know, once again, relearning everything that we thought we knew.

Matt Amundson: Well,

The Emerging Go-To-Market Playbook & New B2B Buyer Behavior Study
---

Matt Amundson: so much of this podcast is all about, you know, what is the new playbook? As we all started to recognize that the old playbook wasn't working. So maybe we can get into, uh. This revolution that you're talking about very specifically around signals and segments. what has changed?

What are you doing differently? And I know you started, you know, unfurling this a little bit about a year ago, but, you know, a year into the, into this practice. Tell us what's different and tell us what you're seeing as a result.

Sydney Sloan: Yeah, well, I'll drop some of the buyer behavior report stats in here because, you know, it's, it is, the buyer behavior has radically changed in the last six months. It is happening so fast. So now, um, 80% of people are using, uh, [00:05:00] AI search as part of their research process. 80%. So, you know, if you haven't yet started thinking about how do you show up in the LLMs you're behind, um, and the reason they're doing it, 30% of them say it's more productive.

And the most interesting stat is that it's being used in enterprise more than anywhere else. Which was astounding to me when we look at the breakdown of segments, um, and, and the three top and, um, you know, I can give you, uh, some slides if you want to, to flash up here too, so you can have them. But the, where they're going first is, um, uh, for short listing is the chat bots.

So 18%, 17.1% of people going there. Software review sites are the second, and then vendor sites are the third. So this is just a re orchestration of what was before third party. Second party, first party data, right? Where third party before was the internet and Google third party. Now are these LLMs and we don't yet have a way to, [00:06:00] to track or influence or buy.

From those signals, from from the LLMs. And, uh, and so the good news is, uh, for G2 is that we did not dome our site and we allowed the LLMs to crawl like Reddit, like Wikipedia, and so. You know, we are one of the top 10 influencers. Depending on what report you see, it's somewhere between seven and 10% of the influence on those LLMs.

So now more than ever letting people realize that the reviews and the voice of the customer and user generated content is really what is pulling the LLMs are pulling from. Um, and then first party. So it's kinda like, oh, I'll do the chat. You know, I'll do the chat. It'll give me my short list. Then I'll, I saw that was referenced in G2, so then I'll go to G2.

And then once I go there, I might do another double check on the person's website. And that's kind of the flow that, that we're seeing. So that's, that is the change. And it's, I think it a change even more.

Matt Amundson: that's a huge [00:07:00] change. Huge

Sydney Sloan: It's cha, it's

Scott Albro: So it's it Sydney. It sounds like in this new way of thinking about things, there are a new set of winners and losers who, who, okay, who are some of the losers in this new world?

Sydney Sloan: I mean, of people who aren't investing in brand,

Scott Albro: Mm-hmm.

Sydney Sloan: because I think this is back to the brand play. So when you look at what indexes the LLMs, it's, you know, the, the sites that source the leaders, um, it's uh, media. So they're still going to media sites. So back to like, you know, how do you get pulled from media or write contributed articles for Forbes and Fast Company.

Um, and so I think for the LLMs it's that, and then because people aren't going to. Spend as much in PPC on Google, then you have to figure out where people are also living outside the lms. So LinkedIn's gonna be a big benefactor for B2B buyers 'cause that's where they live. [00:08:00] Other communities I think can do a lot and then influencers plays a big role.

Um, and, and so, you know, if. If people before were spending, let's say, and Matt, we've talked about this a ton, right? How much do you spend on brand, um, uh, awareness? Uh, so demand creation versus demand capture, and maybe before it was 15, 20% on brand awareness and 80% on capture. Like how is that going to change. Now the good news is the people that are seeing people come to their site, I just was talking to the CMO of Iron Mountain and she's like, you know, our traffic's down.

But the quality of of traffic that we're getting is higher. So people are, you know, the people that want to find are still finding their way there. But again, it's gonna change, it's gonna completely change, um, uh, our strategy. So how does the small guy or girl stand out? And, I've been a couple of like, um. [00:09:00] Founder, uh, sessions and I'm like, well, get your founder game on like Scott Albo, right? Have, have a game, have game on LinkedIn, uh, and, and take the time to build up your brand and awareness like that. That will help, um, you know, do the contributed writing and then really think about. Where before we were, you might have like really cool advisors, like Matt, you were an advisor for SalesLoft, right?

Like, who are your ambassadors and how do you treat them and how do you activate them in the peer side? And then also, I see this a lot now, um, like user generated, promoted content. So you get your, you, your influencers, whether you're customers or paid influencers, they post and then you. You amplify their posts through the sponsored content on LinkedIn.

Um, and YouTube's gonna also play a big role 'cause it's yet to be indexed. Um, and so video and the transcripts from YouTube, um, so a lot of people are coming back to like [00:10:00] rethink their YouTube strategy. Um, and, and so yeah,

Scott Albro: Oh yeah.

Sydney Sloan: what, what we, what we did six months ago is definitely not what we're gonna do six

Scott Albro: yeah, yeah. Super interesting. What, what do you think Google should do here? I'm just curious.

Sydney Sloan: Hmm. Yeah, that's a great question. Um, I mean, I, I started thinking about it when Gemini first started appearing a year ago. I was like, Ooh, that's interesting. Like they're taking the top third of the page to lead their, you know, their Gemini summary. And I can imagine the debates they're having internally, uh, on, you know, how do we still sustain our core of our business, but not lose ground to the rise of the LLMs when they, when they have one. Um, and so, you know, just I think in the last conference that they just announced and how quickly they're adapting, I think they're just, they're, they're having to flip to the LLM game. Um, I, I do think that the LLMs are gonna figure out how to do, um, some [00:11:00] sponsored, like perplexity is already talking about it.

They've got like an early adopter. So I think they're, again, like internally, they're gonna. They're gonna figure something out because, uh, it's, it is all about eyeballs at the end of the day. And so, you know, as ChatGPT gets better and better, if the arrest aren't gonna catch up, then you know, that is going to be, that, that is the race.

Um, and so I think they, they have to,

Scott Albro: Totally agree.

Matt Amundson: Uh,

How to Use G2 to Rank in ChatGPT Results for SaaS Tools
---

Matt Amundson: I wanna get into a couple real tactical things just because of, you know, the, the, the, the seat that you occupy and, and, and your relationship with some of the best G2 users. You know, as things are changing, you know, are you seeing, Hmm? Are you seeing. Your customers, uh, try to have different types of reviews, encourage their customers to leave different types of reviews, just given the fact that like so much of, uh, you know, anytime I'm looking at a vendor, I'm asking an LLM to do, Hey, can you do some [00:12:00] comparison?

And inevitably what I see is so much G2, uh, content coming through. Um, what are people changing, uh, in order to, to sort of benefit from that?

Sydney Sloan: Yeah, I, um, so I thought I had the G2 playbook down, and then I got here and I realized there were some things that I was missing. Um, and, and so, you know, it all starts with review. Um, and, and how you curate those. And the strongest way, the most successful way to generate your reviews is actually in the app.

Matt Amundson: Hmm.

Sydney Sloan: So we have built integrations. So you can ask users just like you do in your personal life, if they're, you know, if they're enjoying, if at that moment of value, if you put your, you know, a prompt up, if they're a simple yes no or an NPS question, then you can throttle and say, I wanna you, would you like to leave a review on G2?

Um, and, and those are through connectors like Medallia or Qualtrics or Pendo, whatever application they're using. And they're in the app so you don't have to incentivize it. So it saves a lot of money. [00:13:00] Um, and they're automatically verified because they're in the product. Um, so that's the number one thing.

I'm like, if you want the easy button for reviews, convince your product teams that it's worthwhile to put the, the prompt in the app itself. Um, and then you've got it always on. Um, the other thing as you were just asking, like what are the people doing to like, make sure the voice of the customer on the reviews is reflective of the messaging.

They're trying to put out there. And so I think you have to be selective of maybe you do a specific campaign or run a certain product area. Um, uh, one of the things that we've just introduced is the ability to allow a customer to leave in one review process, to leave, uh, for multiple products, which has been a big unlock, um, uh, on value for a lot of the suites companies.

So just kind of looking at the, the levers to be able to maximize the impact of the reviews. And then the last thing that I would say is, um, this is interesting, so. Uh, to be, you know, sometimes it's [00:14:00] good to be in every category and sometimes it's not. And I'm kind of like, you, you pay G2 by the categories you're in, so like, this is, this is, you know, but, but if you truly want to be a leader, you know, concentrating your effort on your, on your leading category, like what's the door that people enter most?

And make sure you're a leader there. Um, and, and, and as it relates to the best software list, like companies that are in a lot of grids that actually hurt, it counts against them. Um, so you run it like the most, um, just because like the big guys were going in everything and then like the little guys didn't get a chance.

Um, and so, you know, they, they truly have to be generating leads or generating reviews in those categories at the volume. Um, uh, and so that's another way like Scott, for the little ones to, to stand. It's like. Pick, pick your, pick your best lane, but it doesn't mean you can't buy intent from the other ones.

So it comes to like still building your demand, end demand engine. [00:15:00] You can still buy le buy intent signals from categories you're not in. That's new.

Matt Amundson: Yeah. That's

very to not let you do that.

That was something that I always, always, always wanted.

Scott Albro: Hey, Sydney.

What Makes G2 Reviews Stand Out & Why You Should Always Respond to Customer Reviews
---

Scott Albro: What kind of reviews stand out on G2? Like how, what, one thing I'm curious about is there are a lot of reviews, right? And so how, what reviews stand out, whether that's in terms of where they're placed on a page, uh, whether that's in terms of, you know, just reviews that see more engagement than others.

Any, any tips there?

Sydney Sloan: well, I mean, I think when they're well written, succinct and a little bit punchy, um, they, you know, they tend to stand out. Um, the, uh, the way that the pages work is like, it has like the top performing and then the caution ones too. Um, and I think just from a, a review management, you always wanna respond to the reviews.

Fun fact, 4.6 is [00:16:00] actually the most believable. Like 4.4 to 4.6 is the most believable. I know everybody wants 4.8, 4.9. And then people start to go, I don't know if I trust it. 'cause it's only good reviews. And so you're paying your people and it's like you're, we're incentivizing 'cause that's human behavior.

Um, but you do wanna have some like constructive ones in there and then actually respond to show that you're listening to the voice of the customer and taking action. Um, so I think on those ones where they're not as. Glowing. Like then how do you respond to that and how do you manage that makes a difference.

Uh, 'cause that shows up too. Um, uh, and then Scott, I think like, and this is not a direct answer to your question 'cause I don't, we don't study like what's a best review like, on the page. So I'm just kind of thinking through it. I think then how your teams curate the reviews to repurpose them. So I've always taken G2 reviews and put them back on my website to, as the, like the proof point.

So before you did like case studies, now you can just grab a a a G2 review quote. [00:17:00] So it's also curating the voice of the customer and showcasing it in, in your website where somebody says something about a particular feature or capability. You can go grab that and put that extra validation in on your website.

Scott Albro: Very cool. Yeah, I love that. But by the way, one thing, it, it just, you know, listening to you talk about that brings to mind is, um. I've always been fascinated watching CEO founders respond in real time to complaints on Twitter X. Right? Like some, some customer will say, I'm canceling, you know, my Slack subscription because of this pricing change or whatever.

And Mark Benioff will come on and say, Hey, DMing you right now. Do you, do you, do you, you know, or Brian Chesky. Airbnb does the same thing. Someone will have a poor experience at, at, um, one of their listings. He responds directly on Twitter and it sort of, you know, it strikes me as one of those things that could have gone terribly wrong for the vendor.[00:18:00]

Just because someone's paying attention and actually responds, in this case, the, the founder, CEO, it actually turns into like this huge PR win that's a net positive for the company. It's actually made me wonder sometimes if founders and CEOs are actually seeding these things. Um, I have no evidence that that's the case.

But, but anyway. do you find founders and CEOs actually replying on G2? Does that, does that happen?

Sydney Sloan: um, I'm not a big user of, of Twitter X as you are, but I'm still waiting for the United. CEO to email me back because that's the only time I ever go is to complain about when I'm flying united, how much it sucks. Um, and they, and they never, they never respond. So, uh, take note United. Um, but uh, in, um, uh, um,

How Founder-Led Companies are Using Customer Reviews Creatively
---

Sydney Sloan: most of the companies I'm and lately have been founder led.

And there's a way where you can take all the reviews and create a Slack channel. So like Kyle Porter, like that was how he started and ended his day by reading, reading the [00:19:00] reviews, and he would take captions of those and actually put them in our all hands. Um, and similarly with Rada, we would start, uh, with voice of the customer and we would read like an email or, or a review, and, and the CEO would comment.

So I do think that, you know, for CEOs that wanna listen to the voice of the customer, like that is the best place and it's really easy to be able to create that funnel directly to them in Slack. Um, and then they can choose how they respond to, and by the way, they can, they can respond to the good ones too.

You don't have to not just talk to the, the other ones. And I think absolutely if the founder is. In there and like, thank you for that idea, or, I'm glad you love that feature. We just worked on it because like, it just makes it more human. And I think more and more that's what we're looking for is the human connection and the fact that people want to be seen and heard and know that their feedback is valued and taken.

So it's a huge, huge, um, benefit. And, and you can imagine as a buyer, like you, you're looking at two companies. You see one where. There's [00:20:00] comments and complaints and there's no reaction, and you see the other where there's comments and complaints and there are reactions. Who are you gonna lean towards?

Scott Albro: yeah. I mean, it, it signals so many valuable things, right? Trust, authenticity, founder cares, founder takes ownership. Like, yeah. So I, I love that. That's great.

Sydney Sloan: Yeah. Yeah.

Matt Amundson: Uh, I, I don't wanna make this all about, uh, what's the best way to use G2, but I do, I do find

Scott Albro: Or Do you

Sydney Sloan: secrets. The insider secrets.

Scott Albro: Yeah. Do you mean G2 crowd, Matt?

Matt Amundson: I don't mean G2 Crowd, I do mean G2 ai. Um, but, but it's like, it's really sort of, uh, it's, I, I find it to be, you know, not just because you're there, but like, because it's having this, uh, this real impact on LLMs, like it's. It's suddenly become like really, really important again. And you, you recall back from the presentation that I was a part of at SaaStr, like I was highly, highly advocating for people to get [00:21:00] back into it and really, not that they were ever away from it, but like, you know, if it's lower down your P list, like, like let's move it way up.

Right. Just, just given its impact on LLM search. Are you, like, one of the things that I've found just anecdotally, and this may not be even a trend, but it's just, you know, something, something that I've run into a few times is when I'm looking for something, oftentimes I'll look, uh, via use case and I'll find some review somewhere, and I'm, I'm guessing it's coming from G2, where the review includes a use case and so often.

In G2 reviews, you know, maybe somebody will leave a review. Like it's easy to use, it's simple. I like it for these types of things, but they don't necessarily go into the use case of how they're using the product. Is that something that we should be encouraging our users to talk about? Like, let's talk about your use case or like, you know, and, and maybe that's too big of an ask to talk about, like, Hey, look, like are you seeing x kind of ROI as a result of that?

But,

How Buyers are Using AI Tools In Their Buyer Journey & To Compare Solutions
---

Matt Amundson: but to [00:22:00] me, like I think that there is something. Really valuable to that in terms of how the LLM will, you know, sort of surface that information from you guys?

Sydney Sloan: Yeah. Well it's funny you mentioned G2 AI before asking that question. It's like a tee up. Um, and I swear to God I didn't, you didn't, this was not pre planned because that's what we're building, right? We're, if you can go, we just launched it last week in beta, so you can go to G2 AI and you can ask it questions.

And, um, and so we're on the process of moving. I mean, it's a beta, so we wanna get feedback, but the idea is that you could come with a problem to solve and not have to know the category. So before, it's kinda like you had to know the category to like orient yourself into the way that our taxonomy works.

And now it's like, I have a problem to solve. What do you recommend? Um, and, and, and then the prompt can like, ask you what, you know, why, like if you were switching, Hey, I'm not happy with this product anymore, I wanna switch to this product. Or I'm looking for an alternative, you know, then the, it'll say, well, why?

Why are you switching? Is it [00:23:00] price? Is it performance? Because those are other questions that we ask, and so it can make recommendations based on why you wanna switch. Interestingly, I think that's really cool. Then that data we're capturing as G2 would be super useful for a marketer to understand the why.

So we haven't built a seller side yet. We're just focusing on the buyer side to get that right, to get the buyer to the outcome that they want, which we think might be a, uh, a business case document or an RFP or some kind of document like that. But we're still like in the early stages of understanding how people want a query and search.

Um. And, and our goal is to be better than the LLMs. That's what G2 AI should be like. Uh, it it, if you get a better answer on LLMs we we're not done yet. We, we still work to do. Um, so, uh, it, it, it does come down to problem if you wanna call problem a use case.

Matt Amundson: Yeah.

Sydney Sloan: I think the other thing that I'm excited about is also like questions around, I have these products.

What are the best products that you would recommend I work with? So it looks at [00:24:00] integration reviews, which we've been able to start, um, capturing as well. Um, so we syndicate our reviews to a lot of partners. Like Gong is a good example and they run their partner network and so now they can get like not just the G2 review, but like talk about gong in relation to.

What it might integrate with. And so taking that, that, those data pieces and then informing like, like, this is my stack, you know, what else should I be considering, um, uh, in that. And there are some other vendors, like I think vendors looking at that too. So I think we're, we're, as an industry on a path to like being able to answer those questions as well.

Matt Amundson: Well, I think that's really smart because one of the things that I've really noticed here and like, it feels like everything's happening in the last six months, right? So I think you, I think that your observation around buying behavior changing in the last six months, like only for their supports, my thought process here, but people use different language in LLMs than they do in search engines, right? Like when you think [00:25:00] about, you know, for, for years, you, you try to get in the, in inside the mind of somebody who is doing a search. Whether it was like, from my days at Marketo, it's like, you know, best marketing automation solution might be a search query that somebody would run through Google today in an LLM, they're gonna say, my company looks like this.

There's probably options for marketing automation solutions that would fit my needs. I'm gonna give you some criteria. Can you create a matrix for me that would tell me the pros and cons, maybe the cost differences of utilizing, you know, the sort of top five most popular marketing automation systems.

And that's like, you know, it seems like a long prompt, but that's the type of stuff people are putting into it because they want the output out in like a grid that they can just load directly into a presentation or share with their peers or whomever is on the buying group with them, so that they can say, okay, we've identified the vendors.

We've identified what we believe are the pros and cons. We're gonna either shortlist or we're going [00:26:00] to evaluate vendors based upon some of the criteria that we've already gotten back. And all that is to say is like the behavior of the buyer changes so dramatically. You can't just go and use like these old search terms.

And quite honestly, people aren't using those old search terms. And so you need like a different way to surface that information to people. So I guess all that is to say is it sounds like you guys are really headed in the right direction and it does feel a little bit like you're on a bit of a collision course with some of these LLMs who are.

Who people are using today, but because you own the data, the way that you surface that data and thinking about that differently seems to be a really intelligent business choice.

Sydney Sloan: So I think there's two things there, Matt, like one, like I always wanna be helpful. So the,

How Generative Engine Optimization Helps Your Brand Show Up in AI Search Results
---

Sydney Sloan: the move from SEO to what is now being called GEO, generative engine optimization, um, is. It is a shift and, and yes, we used to write our content for Google. [00:27:00] Now you actually get to write your content for people. And, and so the best advice that I've heard so far is Guillaume G If folks know G he was, uh, intermittent ramp, but he's been at a bunch of different companies and he, he worked with a company called, uh, GrowthX and, um, and they, uh, figured out how to, uh, take the personas and understand the jobs to be done of those personas and write content around the jobs to be

done. So now you're like, okay, if it's a jobs to be done, that's a, that's a pain point. That's a problem, that's a, a responsibility that I have that maybe my product can help them with. And so, um, that was helpful in, um, in indexing, uh, on the LLMs. Um, and so I've started to recommend, like in your person. Uh, consider actually creating a persona called LLM.

And so you're writing for your personas, but you add another like gem or whatever you're using called LLMs. And it's like, how does that agent or [00:28:00] prompt give you feedback on what's the best process for writing to LLMs what we've learned at G2 listicles, um, or, or really, uh, highly, um, consumed currently.

Um, and again, going back to user generated content, 'cause you're human asking a question to an LLM in a human way. And so it's indexing human responses 'cause that's how that's created. Um, so those are some short term things people can do is like, okay, how do I. How do I think about that? And then there are products in that GEO category, which you can now see on G2.

Um, that will help, uh, you see when you are being mentioned in reference. So you'll have context of like, okay, where am I showing up and why? And then if you wanna look at a competitor, where are they showing up and why? And then change your content strategy around that.

Scott Albro: That's, that's super interesting. I think, you know, one thing that comes to mind with G2 AI or potentially the LMS themselves, is how you [00:29:00] prompt them to create the artifacts that you need during the buying and evaluation process. I think, I think, Matt, you were sort of headed in this direction, right?

It's like. I think about a, a typical enterprise buying process, right? You might wanna understand your requirements, then understand the vendor universe, then have a short list of vendors that you're gonna go actually engage that sort of map to the requirements. Then you wanna make a selection, you know, or maybe you want to issue an RFP.

Just feels like having prompts for each one of those things and then allowing the buyer to sort of run that prompt in the context of, you know, Hey, I work in marketing ops and I'm looking for this type of solution. Right? It's so powerful to have an LLM, like create your requirements list and then create a vendor universe list, and then maybe even score the vendors against those requirements and so forth.

I mean, doing that via prompts as opposed to [00:30:00] like. Hey, I've gotta go manually collect all this data, get it into spreadsheets that make sense, or manually write an RFP or whatever it is. That seems like, you know, when I think of G2, yes,

Sydney Sloan: So 2024.

Scott Albro: Yeah, yeah, yeah. But, but it's like, yes, you know? You know, as G2 looks to differentiate against the LLMs, yes, it's the data, but it's also sort of your deep understanding of how a buyer works through an evaluation process.

Sydney Sloan: and there's a lot of data that we capture that we don't share publicly, so. Pricing and, uh, other satisfaction questions. We have a whole bunch of AI questions, so we've got like 65 questions and we only share like nine of them. Um, so we'll have more insights that than, than the LLMs will have 'cause we haven't given them access to it.

Um, but in, in thinking of that, I think there's two, two things about the prompt strategy and, and you know, we tested out on LLMs and then we're testing G2 AI as well. Like, I would start with the [00:31:00] outcome. Uh, so I would upload an RFPI, first of all, I'd ask it to write the prompt for me. Matt, you were writing the prompt.

Now I'm just like, Hey, I'm a buyer for this. I'm looking for that. Like, write, help me write a prompt and then refine, refine, refine, and then would give it to me. And then I'd probably upload one of an RFP. Like, this is the process, especially in an enterprise company. This is what a typical RFP looks like for us.

My goal is to create an, uh, you know, a A an RFP in this format. For these business problems, probably seeded a couple of products that I think I'm using or that I'm switching out from. So I know that it's gonna get in that right category of like, uh, uh, of technology so it doesn't go off on stuff that you wouldn't want or need.

And, and it, you can do it in like an hour. Um, and that's what I was saying. I was just, I went to pull up the data, like the, the, um, the enterprises are the ones that are. Doing this fastest. And we were trying to figure out why the research, uh, leader on this was Tim Sanders and he and I co-presented and, and it's like,

Why Enterprises are Using AI Tools More Than Startups
---

Sydney Sloan: well, [00:32:00] everybody's being asked to do more with less.

And, and so, you know, these, there's more folks in these large enterprises, you know, they're having to get more creative. And so where we've always kind of thought, or at least I have like coming from big company, going to smaller companies that you know that, that. That startups and small companies, people are more adaptable and, and to, to using new things.

And that's not what we found. Um, that, you know, enterprise people are using it just as much as or more than even, which is, that was surprising.

Scott Albro: Sydney, what I, I mean, my gut there tells me that there's just more work to be done in the enterprise when it comes to an evaluation. So use the tools at your disposal. What, what do you, what's your perspective? Yeah. Okay.

Sydney Sloan: I agree. Yeah.

Matt Amundson: I think there's also more top-down directive to be using AI in enterprise companies. I think, you know, smaller businesses, you know, maybe there's an expectation that people are using it, but they tend [00:33:00] to be, uh, a little looser around, you know, sort of their, their. Top down, uh, sort of marching orders to an organization.

Whereas I think like, you know, you could pull just about any 10 k report of a public company and they're saying, Hey, we're, we're, not only do we want to be creating AI products, but we want to use, uh, AI internally to increase efficiency and, um, and, and reduce overhead.

Sydney Sloan: And they're smart people, so they figure out,

Matt Amundson: Yes.

Sydney Sloan: when I was at Adobe, I was in a room for full Harvard grads, you know, so figure it.

Matt Amundson: Hey, when you, when you were at Adobe, they were in a room full of USC grads.

Sydney Sloan: Oh, I was the one. I was the one and the only woman, but we won't.

Matt Amundson: Um,

The Meteoric Rise of Cursor & Which Categories are Ripe for AI Disruption
---

Matt Amundson: I wanted to maybe switch gears here for a second. There is a, there was an announcement, [00:34:00] uh, last week of a company that raised a round of funding. Uh, there's a new, new sort of, I guess, almost deca unicorn in town. I think their valuation is $9.9 billion, and this comes in the form of cursor. Uh, and cursor, they shared some numbers that are just staggering.

Staggering, right? So they, they raised, I think their a round in August or September of last year. They did their B round in December, and then they, they just did a, a c round and they are at, they're a 2-year-old company that's at $500 million of ARR. Um. And then I think, uh, that's shocking. That's very shocking.

I think what hit closest to home for me is in a lot of the dialogue around their growth, um, they have said they don't, they've never served an ad, um, they've never sent a marketing [00:35:00] email, and then I went digging through their LinkedIn. And they don't seem to have a marketer on staff. Um, what does everybody think about this?

Are we just talking about like, perfect product at the perfect time? Is there something unique about what they're doing? Uh, I, I, I'm sure you guys probably haven't spent as much time thinking about this, but I've been losing sleep, uh, thinking about this company, so I, I just love. To get your guys' take on this because I think this is one of the most interesting stories that we've seen here in, in, in the technology space in the last couple of months.

Uh, so I'll just open the floor. Who, anybody have any thoughts?

Sydney Sloan: I definitely do. Um, having. Marketed to developers.

So cursor's a developer, uh, tool, um, that absolutely hit product market fit at the exact right time. Um, but they also, developers don't like marketing. So you may wanna go back and look and see [00:36:00] like, are there developer evangelists and what's their community strategy?

And in like, you know, how, how that works. Like the GitHub models as well, like. It's not a really big marketing led, but it, it, it was like, of course right product, right time that that works. Right? And so it, it just grew like wildfire because it actually worked.

Matt Amundson: Yeah, I mean, they do have a community tab right on their website. It's like next to the blog. Um, they also have a Reddit, a subreddit community with 72,000 members. Um, so they're definitely doing, uh, a lot of their work through that. But Scott, I'd love your take as well.

Scott Albro: Yeah, I mean, I agree with Sydney. I, I wouldn't, uh, over complicate this. I just think it's product market like crazy product market fit, maybe the greatest example of product market fit ever.

And, um, the, the only thing that I think is a little bit different here is. I feel [00:37:00] like, you know, if, if you look at the Gartner Hype Cycle or whatever hype cycle you wanna look at, right?

Like that is, the hype cycle is very relevant when it comes to AI today. And I, I, I would say we are early in the, in the hype cycle like we are, you know. Uh, near, or, you know, we are still climbing our way to, um, the peak of inflated expectations, I think is what it's called. It's like, you know, that thing early in the hype cycle where everyone's

Sydney Sloan: trough of disillusionment.

Scott Albro: And the trough and the trough of disillusionment is coming. Right.

Sydney Sloan: Yeah.

Scott Albro: The reason I bring that up is because I don't think you can put all, all of AI generically into the hype cycle. I think you, you know, you were talking about use cases earlier. You have to look at it by category or use case or, you know, segment or industry or what, you know, problem, whatever it is, right?

And it [00:38:00] it, one of the things that just strikes me about ai, remember it's still early, we're still in the first few years, is like. There is this use case, market problem, whatever you want to call it, where AI just works and that's software development, and it's the one place where we can see the true magic of ai, right?

You can see that if you know these AI code editors, they work. You can see you guys were talking about outcomes earlier. The outcomes are really evident. You know, you can see big organizations, we were talking about enterprise earlier. I think Amazon, like every Amazon engineer is using Cursor now, right?

Like, it, it, it is just this market where like product market fit is just self evident and you honestly can't say that about many other AI segments right now. Right. So it shows [00:39:00] how powerful product market fit is, and it shows how powerful AI is, and hopefully it's a good sign for other markets right. To, but, but this is the one place where we're seeing like the, the true real magic of ai and that and cursor.

Is the ultimate representation of that, by the way, not to mention plenty of other tools out there that have gone, you know, not quite, not quite cursor levels, but you know, zero to 50, zero to a hundred in like less than 12 months.

Matt Amundson: Yeah. Yeah. Well I, you know, I'd be remiss if I didn't mention, you know, windsurf in that, in that space as well, this concept of like vibe coding, which like, come on, can there be a better category name than vibe coding?

Scott Albro: Yeah, totally lovable. I think lovable zero to 40 or 50 in a few months. Right? It's like now. One. One. Okay. So one other thing about Cursor is my understanding is that the retention numbers are really strong.

Matt Amundson: Mm-hmm.

Scott Albro: [00:40:00] I think one thing to look out for here is in like Cursor is not Vibe coding. Like Cursor is like a copilot that helps you do real coding, right? Lovable. Some of these other tools are more vibe coding. It'll be super interesting to see like, okay, it went zero to 50, but like what does retention look like in a year or next month? Right? Because vibe coating to me comes across as more of a toy that you use and have fun with. Right? Cursor to me comes across as like, no, this is like a real enterprise tool that once I start using it, like I can't imagine living without it.

Right? So.

Matt Amundson: Yeah, I think it was interesting. We were on a, i, I did a podcast earlier, um, where we were talking about so many of the AI use cases were this like kind of cut and paste, right? Where you'd go to an LLM, you'd do the work and you'd cut, you'd cut it out and you'd paste it into some, some, some other space.[00:41:00]

And here we're starting to see tools that are really embedded within a workflow. Right. They, they, they are trusted by the users in order to complete some portion of work that, you know, six months ago they were doing themselves. Are we seeing like the door opening to more of this type of. Uh, uh, of, of AI use case doesn't have to be amongst developers, but it could be in other spaces.

And what do we think like the next frontier of this is gonna be? Because to me, one of the things that I find most fascinating about the, the, the space that cursor occupies is like. I think Scott, like one way of saying what you're saying is like, this is truth, right? Like, this is real. This is somebody who's, who's really like the users of this, somebody who's really putting their trust into an AI tool.

Whereas before, we kind of wanted to be a little human in the loopy around it. But like, are we gonna see, you know, for, for marketers or for [00:42:00] sellers, are we gonna see people say, you know, I'm just, I'm gonna go hands off the wheel here and let you do your thing. Uh, does that, does that feel like reality coming in the, into the GTM space?

Scott Albro: I, you know, I'd be curious to get Sydney's perspective on this, maybe G2 has some data on this, but I, I, you know, the way I would look at it is I would look for other industries, markets, use cases. That show that bear a resemblance to some of the things that exist in software development that make Cursor and AI more generally real.

And, you know, some things to think about. There are, um, a large, uh, mostly deterministic body of knowledge that exists in text. That an AI can easily understand and thereby become an effective copilot. [00:43:00] And so, you know, if we think about the world of sales, for example, I'm not sure that really exists in the world of sales.

If we think about the legal profession, that definitely exists in the legal profession, right? You have case law, you have code, regulations, right? All kinds of stuff that an LLM can go learn and understand and then become a copilot or, you know, whatever it is. Right. And I, and I do think we see some of the, the legal specific AI companies gaining traction as a result of that.

I'm sure there are others, but that's one that comes to mind where it's like, well, what actually made it it, look, I don't think you're gonna struggle to find. Companies that want their employees to be more productive, whether they're software developers, salespeople, marketers, lawyers, whatever the question is, is like right now, where is that most technically possible?[00:44:00]

In a way that allows the copilot to be, to really deliver productivity benefits? The legal industry bears some resemblance. There are plenty of other industries that don't, where I think the LLMs are just gonna have to get better.

Matt Amundson: Yeah. Yeah. Sydney, what are your thoughts?

Sydney Sloan: Um, so I'll answer it a little bit of a different way. Um, and that is because like. We see at G2 like this graduation where, um, the, we call it the AI gradient. And so at the very bottom is chatbots and copilots. Like that's the found. That's the where you start. And, um, you could think of this as like ways to Waymo.

So, if, uh, Waze didn't come to be. Uber would not be where we are today, right? If we didn't have the ability to like help guide and direct people. So that was like even a copilot back then. Um, to the next level is like intelligent assistant. So it's taking on more and more of the task versus just write writing as a co-pilot.

Then a task [00:45:00] agent is the middle layer. Um, so you can assign the tasks that agent, they complete it, and then it comes back. So a little more like, uh, agent in the loop or a human in the loop than a process agent. So this is like where three or four agents might to, you know, do, uh, uh, a, a simulation and then you have a system of agents that's Waymo.

It can take all the data in, it, can existing data, new data, and, and ultimately drive the car. Um, and what we're talking about is like next year is going to be the year of AI orchestration, and that is where you take multiple agents that could be built in different. Areas and, and then they are chained together versus like a process which is just a, a multiple, multiple prompts in a, in a single agent system, if that makes sense.

And so in order to do that. Leading companies, and I'm doing this road show right now with Qualified Clay 6Sense and G2, where we're talking about putting AI into action. And so this is [00:46:00] what we're talking about is like how do you rethink and design this new system? And so, uh, I think I'm gonna write a paper on this like.

My, the recommendation that kind of I'm coming up with is you're gonna wanna hire an AI architect in your go-to-market team. So they're likely gonna sit in rev ops, like right next to the VP of Rev Ops. They're a senior level person that is gonna have responsibility for the design of your systems, the orchestration of your data, and the management of your agents.

And so they're like the top level, just like an IT architect would back in the day, right? They've designed the systems, they set up the processes, and so we just need that. But somebody that's looking at it from a new lens, um, and then you'll still have Go-To-Market engineers that are responsible for their different function.

So before we had ops people that did something, sales ops people that did something, and they talked about their systems. And I don't know if we need to like think of them in those silos anymore if we're looking at workflows. From a segment or customer perspective. And so that might, a Go-To-Market engineer might be engineering a process for a certain customer set all the [00:47:00] way through the process, which we don't do a good job of today.

Um, if you doing audit of your systems and see how many people, you know, are emailing your customer or whatever, like, so it could change. So we are doing a better job of. Servicing and communicating to certain types of customers through this single kind of engineer, and they can figure out the, what is the platform good for and where do I need prompts in that solution.

But it still needs to be managed. That's the difference. Of previous lives where it's like you set up, nurture, set it, and forget it, and it just runs like that. I don't think we're in that mode even close. Like I think it's gonna continue, learn, iterate, learn, iterate, adapt. Um, and so I like when I see like even in PE companies that have started to adopt.

The AI SDR R is like a qualified piper, you know, it still needs someone to manage it. You train it, you manage it, you, you know, you continue to check on it. And, and, and so that whole category is doing that. Um, so that's kind of how I see the future, um, in, in [00:48:00] Go-To-Market is like a re reorganization of responsibility and the way that we manage the work, um, and the determination of where agents fit and what role they play.

Um, you did say Scott on the like. I dunno if there's enough data for sales. I, I think there are, like when you think about. Calls, transcripts, emails, like all the data that a company would have internally that could train an agent. And then we actually are, uh, build, uh, G2 data sets to train agents on categories, objection handling, recommendations, comparisons.

Um, I'm seeing the birth of synthetic, uh, I think it's synthetic is the right word. Like, um, uh, like one mind where, you know. Uh, they, HubSpot just like built an agent that basically closed business and we're looking into that as well. And so, you know, eight months ago I was like, no. As you know, people still wanna talk to humans, but guess what, um, uh, if the, if the agent [00:49:00] has a better answer, they actually might be okay talking to the

Scott Albro: yeah. To, no, to people. People. There will be a segment of the market that does not want to talk to a human right by, by the way, my point on sales wasn't that there's enough, I should clarify, wasn't that there's not enough data. the, the point about the coding space, the software development space, is that all of the data and information is deterministic.

Right. You know, whether that line of code is correct and, and similar in the legal industry, right? Where like, you know, look, this is what the case law says and you can see, you know, how a judge ruled. And, and whereas in sales. Agree. There's a ton of data that actually might be a problem, right, because it's

Sydney Sloan: We only wanna pick the good

Scott Albro: Yeah, yeah. Right. It's not Now. Now we, we, we, and I don't want to come across as a pessimist. I totally believe we're gonna figure that out. It's just gonna take a little bit longer than like, Hey, [00:50:00] here is how you as structured Here's how you write this line of Python to achieve this. Right. And that's. It either works or it doesn't.

Right. Like LLMs like that stuff right now. And, and that is, I think one of the reasons Cursor and other tools in the development space are, are working Right. You know, real results. So,

Sydney Sloan: Makes sense.

Matt Amundson: Geez. I mean, we just blew through an hour. Holy smokes. This was

Scott Albro: well, it took us like 10 minutes to get started because Sam couldn't figure out the recording button.

Matt Amundson: But that

Sydney Sloan: Oh, but we have to blame Craig for that.

Matt Amundson: that

Scott Albro: that was Craig's fault, I forgot. Yeah. Yeah.

Sam Guertin: this moment to, uh, yeah, throw Craig under the bus.

Scott Albro: Good. You should, Craig is, has a little, a lot of Teflon on him when it comes to like mistakes, owning mistakes. So, uh, yeah. We'll, we'll give that one to Craig, Sam.

Matt Amundson: Well, Sydney, thanks so much for being on the show. This was fantastic. A lot of great tactical. Uh, a lot of great tactics that some of the listeners can take back and [00:51:00] start implementing right away. I think there's a lot to think about. There's obviously just a whole new crop of these AI companies that are exploding right now.

There's a lot to learn from them as well. Um. We're just gonna, like, it's, it's crazy. Six months ago, the world looked completely different. Six months from now, it's gonna look even further different. So I think, uh, you know, it's, it makes it both a fun and really challenging time to be in Go-To-Market right now, because, you know, it's, it's a little bit challenging to make long-term bets, but I think one of the things that we're all sort of consistent on here is this, this idea of GGEO is, is real.

And it's having a massive impact on the way buyers are buying.

Scott Albro: Yeah.

Sydney Sloan: it will be the way forward.

Matt Amundson: It is the way forward. This is the way. Alright, well that was the transaction everybody. Thanks so much for joining Sydney. Thank you for another incredible podcast. And Scott,

Sydney Sloan: on your five timer,

Matt Amundson: the five [00:52:00] timer, Scott. send you a plaque.

Scott Albro: Yeah.

Matt Amundson: all right. Thanks everybody.