[00:00:00] Kevin White: You can really get swept up in the news cycles because every day there's something new that's "game changing." So that is certainly not a way to operate and can't just be reactive at all times.

[00:00:20] Andrew Michael: This is Churn.FM, the podcast for subscription economy pros. Each week we hear how the world's fastest growing companies are tackling churn and using retention to fuel their growth.

[00:00:33] VO: How do you build a habit-forming product? We crossed over that magic threshold to negative churn. You need to invest in customer success. It always comes down to retention and engagement. Completely bootstrapped, profitable and growing.

[00:00:46] Andrew Michael: Strategies, tactics and ideas brought together to help your business thrive in the subscription economy. I'm your host, Andrew Michael, and here's today's episode.

[00:00:57] Andrew Michael: Hey, Kevin, welcome to the show.

[00:00:58] Kevin White: Hey, excited to be here. Thanks for having me.

[00:01:01] Andrew Michael: It's great to have you. For the listeners, Kevin is the head of marketing at Scrunch AI. Scrunch is the key into the AI first customer journey, monitoring brand presence, get actionable insights to grow it, generate an AI friendly version of your sites to influence LLM results. And prior to Scrunch AI, Kevin was the head of GTM strategy at Common Room, head of marketing at Retool and Segment, and an advisor to companies like Ashby, Deepnote, and Gretel.ai. So my first question today for you, Kevin, is in the context of your new role is, what does the future of the web look like?

[00:01:33] Kevin White: Yeah, it's a good question. Question that I don't know the answer to. I wish I did. I probably set myself up for being rich if I knew the answer to that, but I can quantificate or prognosticate about what the future of the web looks like. And I think this is kind of the thesis and what we're building at Scrunch is trying to foresee and go where the puck is going in that sense, which is reading about the big trend that's happening right now is that people are stop using Google, doing research on their own behalf. Like clicking blue links, looking through a recipe site, or some site that has just a ton of different junk because of an ad model and that experience to uncover a job to be done on the user end is not the best anymore.

[00:02:15] Kevin White: And so with LLMs like ChatGPT, like Claude, like Perplexity, these answer engines are really great ways or provide a better user experience than that old paradigm where you're doing all of this research on this one platform as a user. That platform is gathering and doing the research on your behalf by sending out AI agents out to go crawl those blue links and crawl those resources and then pull back the relevant info based on the prompt that you have. And so you can kind of see that new paradigm. You can see that happening in front of our eyes here where sites are seeing less organic traffic, they're seeing more users do zero click type experiences where they're researching the brand, they're comparing their products, all this kind of stuff on these new platforms.

[00:03:01] Kevin White: We can say what we will about that becoming a more boring internet if everyone is not going to different websites and stuff like that. But it is certainly a better user experience. And I think the users migrating to those platforms and having AI agents research and do the work on their behalf provides a much better experience. And so I feel like that is where we're seeing the future of the web, internet, AI, all that stuff that the future is on these platforms. And yeah, it's kind of exciting times. This is a new channel and lots of new stuff to learn.

[00:03:30] Andrew Michael: Yeah. It's crazy how fast things are changing. Actually yesterday, my wife was looking for something. She said, I'll just Google it. And then my son was like, "Mom, just ChatGPT it." My son's six years old. It's like, okay, I can see how fast these changes are happening in real time. I think one of the things as well as maybe in knowledge you're talking just triggered a thought is a lot of it as well geared by this instant gratification, I think that we get through like social and these fixes. And now, this is just accelerating that instant gratification whereas before you actually had to go through those links or you had to suffer through ads to be able to see the content. And now it's just instantly appearing there. So that user experience is just 10x better than what it was. And like, I think it's obvious that the market's going to dictate the way they want their experiences to work now that they've seen the lights as well.

[00:04:18] Kevin White: Yeah, I agree. But my concern is what it means for humanity if we're just getting answers fed to us, but we'll find a way. I think we'll figure out a way to learn. And I'm nostalgic for the old way of doing things, but also very excited about the future. So we'll see.

[00:04:32] Andrew Michael: Yeah. I think it's this thing as well of what is going to happen to websites? And I think we actually had a conversation about this around sort of like the theory of the dead internet where I saw recently a chart – and I think it was this year, where AI generated content has surpassed the human generated content on the internet. And at some point, do we just have an internet that's generated by AI being visited by AI and just like humans just really acting at the edge of it on the fringes?

[00:05:00] Kevin White: Yeah, well, that feels like a very dreary world, which we don't want to live in. So I feel like what's going to happen is these platforms, they're consuming all of this user usage and time will actually figure out a way for websites or journals, media outlets to create content. I mean, you've kind of seen this with… I think it was Google paid for the Reddit library. If you incentivize content creation that way in real authentic human content, I don't think that that is going away because otherwise, if AI is creating content for AI, at least the initial studies on it show that it becomes this kind of like doom loop of poor quality content over time, and everything just becomes like stale and generic in the same. And so I'm sure the LLMs, LM companies know this and they want to incentivize against it because it's like their business is at stake. So I'm confident we'll get it. We'll find a way through. I'm just not sure how right now.

[00:05:51] Andrew Michael: Yeah. I think it's very interesting to sort of think through these steps of what ends up happening there because when you're having an internet generated by AI that gets trained again by AI, just going to this a continuous loop of no progression at some points as well. But let's talk about the context. Well, obviously like Scrunch AI, you're trying to make sense of this now, as you mentioned, and trying to see sort of like where the puck is moving. I think definitely one of the things that companies are thinking about today is really like, how do we start to build experiences to be able to capitalize in this new paradigm and the new way of people seeking and getting information. So what are some of the ways that you're working with customers today to help navigate through this and take advantage now as earlier movers in this next wave?

[00:06:36] Kevin White: Yeah, we've kind of been taking a different approach to some of what you typically see or what typically comes to mind. And what comes to mind oftentimes is, "Okay, we want to track these prompts, our brand isn't showing up for these prompts. So then let's generate content to then show up for the prompts that users are putting into these LLM platforms." I feel like if you have a human in the loop, that is probably a good strategy. But oftentimes, a lot of people are like, "Oh, I can just automate all this with an LLM." And so where... I feel like that is a somewhat flawed premise is kind of what we are talking about before. It's like if the LLMs are generating the content, then that content is probably not being rewarded in the same way because they know that it's generated by an LLM.

[00:07:19] Kevin White: And so I feel like some of these workflows end up creating a lot of noise and a lot of content slump in order to rank for these prompts or show up for these prompts. And so our approach is a bit different, although we do work with a lot of agencies that help out with the content side of things, but our approach is more like, you have a library of content, you have a website, the LLM is using your website to return answers back to the users on those platforms. But maybe the LLMs when they're trying to retrieve data on those sites are being blocked or they're not getting the right information.

[00:07:50] Kevin White: Pricing pages is classic example of this where because of JavaScript and lots of different organizations of pricing pages, LLMs often get confused of like what pricing model means what and like what features are attached to that pricing model because one table is disconnected from another on the user experience and yada, yada, yada. So our approach at Scrunch is like, we know that these agents are coming to your site, working on behalf of the human prompting behind them. Let's take that traffic and let's serve up a parallel version of your website, of your different pages and make it so that that page is optimized so that the agent consuming it can consume it in the most optimal way.

[00:08:29] Kevin White: That typically means things like making sure that your markdown is all structured the right way, that there's not JavaScript's clouding things up, that you're using schema, that the LLM can consume fast, that the page loads fast. And then if you have JSON, is it also like the linga frinc? What's right term for that? Lingua franca? It's a Latin term and I'm butchering it right now, but that the LLMs can essentially take that content and consume it in the right way and then serve the best answers to users on the other end.

[00:08:56] Andrew Michael: Yeah. No, I think that's one definitely interesting angles or something I actually tried to do recently was exactly that was just take all the pages on the site, convert them to MD files, and then be able to serve those. And then, but there's just so many complexities that go into it. It sounds simple at the surface to actually do that. But then when you have to think about what happens when you update the content, how do we serve this content? How do we make sure that the elements are getting to this content? And so, and as you mentioned JavaScript and all of those issues that come up.

[00:09:23] Kevin White: Yeah. And it's also something that your typical CMS is not equipped for right now, especially like intercepting that traffic and serving a different version of the site. Like I think there's like headless CMSs that could, you'd get there.

[00:09:35] Andrew Michael: Exactly. I had to start working with Cloudflare. Like I went into this endless loop, then I'm trying to get it to it. Yeah. But I think that's also something – cause like I've noticed this is a trend now as well recently. Like I think Webflow does this, [inaudible] does this. There's a few other brands that have actually started to do this with their sites where they like generating these MD files across the board. And it almost feels like this is going to become a standard on the web now. It's like how do we enable LLMs to consume content effectively on our sites?

[00:10:00] Kevin White: Yeah, it's kind of a fun thing because in my traditional marketing experience, you get bot traffic and you're like, oh, you just dismiss it or ignore it because you're like, "Ah, this is just a crawler. It's not an important user, all this kind of stuff." But now the bots that are crawling your site, especially those retrieval bots are almost like your new VIP visitor. And so, you have to kind of adapt and adjust your website for this new type of visitor. And so it opens up a lot of opportunity for product or platform like Scrunch, but also for marketers and other people to figure out what is the best way to serve the bots so that you can get the conversions of your product.

[00:10:34] Andrew Michael: Yeah. It's weird though, because if you think about it really like they're doing the same job. They're like coming to your site, getting the information and seeing how they can rank it or share it back to the user in the most effective way.

[00:10:45] Kevin White: And that is why there is still a lot of similarities between how you optimize content for human or SEO versus how you do it for answer engine optimization or AI search. There's still human on the other end and you're still trying to best fit the needs or the jobs to be done of that human. And so it's just in this case, there's a AI agent going and doing that on behalf and returning data. But that data or that content is still the content that, you know, someone would have to research and do on their own. So it still needs to be valuable in a utility.

[00:11:16] Andrew Michael: Yeah. I think this stuff is fascinating and just to keep tabs on and see the way things are evolving now. I mean, I use Twitter quite a lot just to keep up to speed, but it almost feels like every single day you're seeing new advancements in AI. Like every day it's like something, this changes everything. Like I will almost need to sort of now filter out the breaking and this changes everything because things just rapidly evolve so fast. So how's the team thinking about keeping up with the speed and rates of change?

[00:11:44] Andrew Michael: And as you mentioned, trying to meet the puck where it is, like where it's moving is not an easy thing. I think it was like Hiten Shah on the show said this quote that stuck with me, which was like, "Product market fit is a moving target. And you can never really think you have it. You always need to be like trying to get closer to it." And so how is the team sort of actively keeping track and what's happening in the market?

[00:12:07] Kevin White: You can really get swept up in the news cycles because every day there's something new that's like quote-unquote "game changing." So that is certainly not a way to operate. You can't just be reactive at all times. I feel like we have just like this blend of what our core vision is of where we want to take the product and where we see the market going. And we really rely on our customers for being with us for that journey and also to inform us of what they need, what they're doing. And so that's kind of a more like long-term roadmap planning.

[00:12:38] Kevin White: And then when there's new things that come out, like this might date 

this podcast, but ChatGPT recently released their like app SDK essentially. So being able to plug into that is something that we need to support within our product and help our customers navigate to the ability to do that. And so whether that's, you know, us providing the right kind of like guidance there, or like actually building in the product, it's still kind of fits in line with like the helping our customers like get to this bridge, this journey, but it's something that's new that they're asking about and we need to be on top of.

[00:13:10] Kevin White: And so essentially, we'll just listen in. It's pretty obvious to know the big things versus the things that say they're game-changing but actually aren't, but really we're listening for our customers to help navigate them and then using them as a guiding light for all that stuff.

[00:13:23] Andrew Michael: Yeah, because was that ChatGPT gets launched, the web is dead, ChatGPT launches Atlas, the web is back and it's going backwards and forward motions all the time. And like you said, you need to not get too distracted and just say if you have a good vision and direction for you, things are going to this way. Let's talk a little bit about your previous experience. Well, because I think obviously Scrunch is pretty early as well to start talking about churn and retention. I think you've worked with some interesting companies like Retool and Segment. We previously actually had... I don’t know if… she worked at both of those companies also. I don't know if your time's overlapped at both.

[00:13:56] Kevin White: I was first at both. So Eleanor was following me.

[00:14:00I Andrew Michael: Following you. It was interesting. The episode when we're talking about Segment at the time, I still think it's one of the most popular episodes to date. And just because of the approach that Segment took back then, you were on the growth side and then Eleanor was on the customer success side. So I'm keen to hear from your perspective a little bit about like, let's pick Segment and say, what was it like in those early days, like in the growth team, how much emphasis actually was given to churn and retention? Because I think like segment was a hyper-scaler typically like churners like an afterthought in the early days until some point somebody has a growth ceiling calculation and like, "Oh shit, we better pay attention to this." So what was it like in the early days of Segment?

[00:14:38] Kevin White: I was there fairly early, but not so early. We were at a pretty good scaling. It was like a hundred employees are in like 10 million in revenue. So when I was there, it was actually this kind of interesting moment where I joined and we're like, oh, we're getting lots of pull from our self-service into more of an enterprise motion. I was actually starting and helping out the self-service team that Eleanor was eventually on as well. And we were seeing our ARR number just kind of flat line because we're hitting a ceiling, but essentially our ARR was being swallowed up and taken by the enterprise team.

[00:15:11] Kevin White: They would hit a cap of usage of events within Segment. Or I guess it was like, we would track on unique events, I think, or that's what we would build on. Once someone would hit like this ceiling there, essentially an AE would come in and be like, "Hey, I'll give you an annual discount on this." And so it was like eating at our churn numbers. And we're like, we were kind of racking our brains of like, what's going on there on the self-service team. But then we quickly realized like, "Oh, this is actually… when our self-service numbers, ARR numbers go down, it actually helps our enterprise numbers go up." And it was just like this whole exercise of trying to align on like, what is the right motion there?

[00:15:45] Kevin White: I will say that I've learned a lot from that experience and took that experience into Retool to not have that happen again, where essentially you create with the self-serve motion, you create this big gold mine of enterprise customers. And then you hire a big A team because that A team is able to take those self-service accounts and like 4 or 5X revenue from them by offering different features and different benefits and annual contracts and all that kind of good enterprise stuff. But then if you do that in a quick fashion, like we did it in like a quarter or two of Segment, we depleted all of that gold mine of self-service revenue. And it was really hard to keep pace to get it back to the place where we would feed that enterprise AE team.

[00:16:26] Kevin White: And so I learned that quickly and then actually bridged that over to retail where it's like, hey, we need to have a system in place where it's like, if someone has to hit this criteria or raise their hand in this way in order for us to say, this is a good account or good way to move from self-service to enterprise versus like a free for all of like, hey, we hired a bunch of AEs hit your number in any way possible. And then it just ruins your whole model essentially.

[00:16:49] Andrew Michael: Yeah. I think from the sales side as well, like in the past, we've interviewed, I think it was like Steli Efti back in the day. They had an interesting model where it was like the sales were not incentivized based on the first sale, but they would get like 20 percent of the commission on the first sale and then 80 percent on renewal or something like that, which like incentivize them to focus on like closing the right customers, the right time, not trying to come in too early in two deals.

[00:17:10] Kevin White: Yeah. I think HubSpot is also kind of, I remember them having a similar kind of study or incentive where the revenue or the commission was backloaded on the rep side of things so that like you weren't just like closing a deal, getting revenue and then like they would churn a year later. Those types of incentives matter and sales reps are especially incentivized because they're just being paid to come on this stuff. So that is an important factor in all of this self-serving enterprise motion, especially.

[00:17:35] Andrew Michael: One of the interesting things I think that came out of the conversation with Eleanor was at the time, I think she mentioned something like, notice that in the self-serve motion, just allowing users to get started without sort of like putting a good tracking plan in place ended up leading to churn on the other end because they just had this like free fall and something. And then at some point, like implementing a stage where you actually needed to sit and work with somebody to put together a tracking plan, which like in most cases would be considered like adding friction to self-serve experience. And like, in most cases, like a no-no. In this case, it turned out to have a big impact on actual retention in there. Because I think ultimately, like, if you think about Segment, value prop, it's like track one, send everywhere, have good clean data. But then if you don't start with a good clean schema, then ultimately end up with the same problems with any other tracking is like, is this something that you worked on similar timing?

[00:18:28] Kevin White: Yeah, we had different people within… on my team and Eleanor's team working together there. Actually, on my team, it was, more on the lifecycle management side of things where we were using email and in product notifications to nudge people towards the right things within the product. But then where Eleanor's team kind of came in stepped in was… today it's called like a scaled success org, where it's essentially like you have a inflow of self-service customers or users, and then you find those that seem like a good fit for your product. Hopefully, it's like a good percentage of them.

[00:19:02] Kevin White: And then you create this kind of white glove experience where you have a meeting with them, you set them up on the right tracking plan. And you can do that at scale if you have a team and do like some sort of like a webinar motion where it's like every week, you're just walking people through how to build a tracking plan. And it's a way to kind of take the full services experience and like scale it so that way when you have these self-service users, they're not churning so fast because they are lost with a technical product, we have to set up a tracking plan and all this kind of stuff.

[00:19:32] Kevin White: So I think scaled success wasn't actually a thing or a term during our time at Segment, but that was kind of like the kernel of team that we put together, which is marketing, supporting of meeting people where they are in the life cycle. And then Eleanor's team, essentially supporting and building out their resources and training that self-serve user based so that they don't churn five months down the line or whatever, when they're like, I'm not using this product anymore.

[00:19:55] Andrew Michael: Yeah. I actually, I think I went through that like scaled success motion at some points, and I think it was actually very, very valuable for us at the time. I remember like sitting down next to being deliberate and thinking about what you should be tracking, how you go about tracking it. And-

[00:20:09] Kevin White: At Retool, Eleanor and I as well built out this program where we would carve out those people who are signing up for our product. And if they look like a good fit, especially for enterprise motion, we would offer like this white glove service for them. And it is kind of like a biased audience because they're raising their hand and like participating. But we found that people who went through that program were like 3X more likely to stick around and continue using the product because we're helping them and building along with them versus just like letting them kind of figure things out on their own.

[00:20:37] Kevin White: And as much as the resources and videos and tutorials and all that stuff you can create, a lot of times people get more value, I would say, with like this like hands-on talking with someone type of experience. And so you have to have outlets for people to learn the way that they want to learn. It's kind of a learning that we have.

[00:20:52] Andrew Michael: I mean, it's exactly the conversation we're having earlier around consuming content and just getting answers. Like humans want to find the easiest path to get what they need. In some cases, it’s reading a blog post or watching a video and other cases having a conversation with a human. And so what did that process look like for you? We can pick Retool cause obviously it's a little bit more fresh and you've learned from the stakes than maybe in Segment. So coming in, seeing, okay, we have the self-serve motion. We want to find a way to sort of scale this so that we can onboard them effectively. Where did you get started?

[00:21:22] Kevin White: It's pretty classic type of do things that don't scale fast type of motion. Essentially, we had the hypothesis of the program. Essentially, the program was, hey, let's carve out some dedicated human resources to reach out to the people who are self-centered product and like offer this white glove onboarding experience, we'll build with you, we'll help you out. I think some people call it like a reverse trial nowadays too, is essentially like, hey, we'll have a meeting with you and we're not going to just pitch you our products, we're going to like build along with you and set you up for success right away in like a 30-minute session.

[00:21:53] Kevin White: And so we had that hypothesis at Retool. We essentially carved out like, what is the top 10 percent of 10 percent of people who are signing up and like, let's just start there and email, you know, 50 or so people and see who opts into the program. And then that also serves as like a learning mechanism for like, okay, when we talk about this or people are getting hung up on this. Then like, let's build that into our curriculum or our onboarding experience because we didn't foresee that. And so essentially, it's just like doing that thing that doesn't scale and then figuring out how to scale it. Eventually we got to a place where we would have 50 to 100 people on a webinar, we would sometimes pull in customers and like show different examples.

[00:22:31] Kevin White: And essentially, we would just took the learnings of this is what it takes to get someone like onboarded successfully in this 30-minute session. And then do that in a way where if you're watching and following along, you can get there on your own. And then also we have an outlet if you need to raise your hand or get stuck, like there's a Q&A section at the end, if you want to stick around and get actual like, like one-on-one type of help.

[00:22:53] Andrew Michael: Yeah. It's like, I think we're at Rahul Vohra from Superhuman early days, like onboarding every single customer one by one. But in this case, obviously like the business was already established. It's sort of like reversing back into that a little bit so you can gather the learnings and then figure out like what's improved during onboarding.

[00:23:09] Kevin White: Yeah. I think we had it at both ends where Superhuman was like, let's do this thing first and then we'll open the floodgates on self-serve. That's maybe the better motion actually, because you learn all the like intricacies and make sure people don't churn and then you can scale and automate it. But at Retool, it was like, we already had this Segment, so we had this built out motion and then just needed to like reverse back into that kind of like white glove way.

[00:23:31] Andrew Michael: Yeah. Cause I think it definitely is the most effective way. And I think on the show, like closing in on 300 episodes now, like at least 200 episodes, we've talked about activation being like the most important aspects of reducing churn. And so the most attention you're going to get from your users in that first week or even in that first hour, like to some degree and like... maximizing that time as much as possible is always going to be your best bet of like setting them up for success.

[00:23:55] Kevin White: Yeah. We just opened up self-serve. It's a soft launch, but by the time this episode goes out, I don't think it'll be soft launch at Scrunch. And so I get a kick out of building the onboarding sequences for emails and like building the content. And I'm like in the midst of that right now. And so I know it's fun for me to just try and think through it. Like, okay, what does this user need? What are their goals and their ambitions and like, how can I fit into like those motivations and get them to use our products in the right way? That whole puzzle to me is really fascinating.

[00:24:24] Kevin White: At Scrunch too, because our product is very self-serve, our sales team, when they have a call, they'll essentially build the whole workspace and environment so that you're demoing to the user, the prospects with the competitors and the personas and the prompts and all that kind of stuff that they would care to track or they would look at. Like that works in a lot more effective way than just demoing some like random brand environment.

[00:24:48] Andrew Michael: I think it's interesting though as well because like focusing initially on sort of enterprise clients and the onboarding experience that provide there, it tends to differ like for your self-serve SMBs, mid-market customers, their needs and the use cases and so forth. So like, it almost feels like you need both to some degree. You need to have like a really good polished, like high touch onboarding experience, but then doing this like reversing back into sort of like a white glove for the small businesses medium [inaudible] and one where you can automate it. And it can be done through like email sequences and onboarding experiences.

[00:25:19] Andrew Michael: I think it's definitely likewise also sort of like really enjoy this aspect of it. Cause like it's finding like who's most successful the product, like what makes them successful and how do we get more users to that state? And think that's like, if I had to summarize all my learnings in three or six years of this podcast, like that's probably like the one thing I'd always lean into and just focus on.

[00:25:40] Kevin White: Yeah. I'm a big believer in those product market fit surveys. And I feel 

like if you have some sort of traction, then you get those users who say like, I would be very disappointed if I use this product. And then you can get a pattern match there. And then it's like, oh, there you have your first audience that you can like start targeting and build and scale around. And like that whole thing is – or motion has unlocked a lot for me and like clicked a lot of like, oh, this is how you grow. And then you keep your product sticky and your net dollar retention, like going up into the right. That is like a cheat code for all that.

[00:26:10] Andrew Michael: And again, I think Rahul Vohra took that to the next level at Superhuman, which was interesting. And I think it's probably also relevant to Scrunch now as well as at the time they did the product market fit survey, but then they also asked like, what was the main reason or value prop that you signed up for? And so in that case, I think originally, they were focusing on speed. Like they just wanted to be the fastest email clients. And like those who mentioned like speed was the most important thing to them, they waited their feedback, like way more than just somebody who was saying like, I want an alternative to Gmail or whatever it was at the time.

[00:26:41] Andrew Michael: So like, I think at an earlier stage, it's important to like have both sort of like guiding posts. One is like, okay, like I care deeply like about this product. I'd be very upset if I didn't use it. And I'm like, what I use it for is really aligned with where the company wants to go and take the product because then at the earlier stage. Like, well, especially at this time where things are a little bit greenfield and you can go in any direction and like not having an opinion can also be dangerous, which just leads you down to like chasing the customer and then you end up building like nothing at the end of the day.

[00:27:07] Kevin White: We ran a survey at Scrunch, I think it's just because the space is so new. The answers were definitely across the board. Like some people are just like, I want to just stack rank how I compare against competitors. I just want to have some sort of tracking in place so that I can tell my CEO or board. So we didn't get as much really strong insights where it's like, we must go this direction. I feel like, you know, it was positive across the board, which is like nice and validating where it's like, hey, a lot of people like this product. But I think the nature of not having those like really clear insights about the products just because of the nascency of the stage and like every other–

[00:27:39] Andrew Michael: It's not established. Yeah. I think with email was much more clear from their perspective. Like everybody knows what email is. Everybody knows what to expect from it and like what they would prefer to have. So that makes a lot of sense. I see we're coming up on time. So I want to make sure I ask you a couple of questions now that I ask every guest that joins the show. The first question is, what's one thing that you know today about churn and retention that you wish you knew when you got started with your career?

[00:28:04] Kevin White: I feel like the whole Reforge School of Thought around churn and retention and how important it is not to feel a leaky bucket is something I learned through Brian Balfour and that line of thinking where it's like you have this long tail kind of retention curve. And if you can get that curve to be like smiley at the end, that's where it's like… if you see that type of dynamic working, that's like when you should invest heavily into acquisition and awareness because you're not filling a leaky bucket. I definitely did not realize that early in my career, but once I actually saw that and saw that come true, I was like, Oh, to me, that's when it makes sense of like, Hey, let's really accelerate and put the pedal to the floor and like really invest a lot.

[00:28:43] Andrew Michael: Yeah. I think like reforging the material they put together, it was excellent courses. [inaudible] checked it out, I think they've also redone a lot of them now, the retention and engagement one. I think I did like three times at some points and eventually was like teaching it at Hotjar to everybody as well, because it is so valuable, the information. I think they do a really good job of like taking relatively complex topics for most people and simplifying them into frameworks that anybody can understand.

[00:29:06] Kevin White: Yeah, he's heavily at Segment. And so I've been fans ever since we were exposed to that at Segment.

[00:29:11] Andrew Michael: Probably around the same time in the same cohort. Very nice. And next question then we'll take it in like the context of growth. Obviously, you've worked with a lot of good startups. You've advised a lot as well. And what's one question when it comes to growth that you wish more people would ask you, but they don't?

[00:29:29] Kevin White: This is a very niche thing, but I feel like a lot of people get hung up on like the next cool growth experiments or like, give me some cool tips or tricks. One of the things that I feel like is so critical to growth that a lot of people overlook is naming conventions. I know that sounds like really, really weird and very boring, but if you don't name things like your campaigns, your lists, your audiences, anything that you're doing, that's a marketing byproduct or growth byproducts, if you don't name it the right way, then other people who consume it or when you try and report on it and all this stuff, like it all falls down and becomes chaos. And so I'm really big on trying to keep rigor around naming conventions because I feel like naming conventions and clarity and organization are the key to scaling to venture scale, I guess.

[00:30:19] Andrew Michael: Yeah. I was laughing because I'm extremely… I know when it comes to this thing, like extremely pedantic and I think like as you say, like you want something that can survive, like anyone, anyone can just read it and understand what it is and not question things. And like, so often, I think going back to the earlier talk of like, dates and analytics as well, think that's also like one of my biggest pet peeves is like the event naming, just like how I wanna just have a simple system, like a crowd, read, update, delete, it's like same formats.

[00:30:44] Kevin White: Yeah. At Segment, it was like very action object-oriented naming conventions, very restrictive on that.

[00:30:50] Andrew Michael: That are good formats. And I think it's just like almost impossible to get people to follow it though. Eventually, like I've tried many, many times. Okay. I think at Hotjar, we came the closest, but ultimately you always get have one rogue engineer that comes along and like, oh, new events, and just doesn't take documentation, anything.

[00:31:07] Kevin White: New event, SSO TV 2 in office.

[00:31:10] Andrew Michael: And then there's a hundred things you're trying to figure out just to get one report. Kevin, it's been an absolute pleasure chatting with you today. Like before we wrap up, is there any final thoughts you want to leave the listeners with anything you'd like to share?

[00:31:21] Kevin White: I guess I'll plug myself and plug Scrunch where you know, I'm talking about this stuff a lot, not just about AI search and that space, but also just marketing and growth marketing in general on LinkedIn. So plug for my LinkedIn handle, just Google it and or AI search it and you'll find me probably. And check out Scrunch if you're trying to bridge this gap of like, where's all my traffic going? How can I get my brand to show up on these LLM platforms? Scrunch is a really great tool for that. So check it out. It's self-serve now.

[00:31:51] Andrew Michael: Amazing. For those who will make sure to leave everything we discussed today in the show notes. You can pick it up there or you can just ask ChatGPT to find Kevin's LinkedIn profile and Scrunch. But thanks again for joining today, it was a great pleasure. I wish you best of luck going forward.

[00:32:04] Kevin White: Awesome. Thanks, Andrew. 

[00:32:06] Andrew Michael: Cheers.