Ep. #2, Charging Customers was a Mistake with Jamie Davidson

Christine Spang: Welcome to Platform Builders, the absolute best podcast out there to learn about the titans who build the tools you use every day.

Isaac Nassimi: We have with us today Jamie, who's the CEO at Vitally, the absolute best customer success platform out there, in my opinion. Welcome to the show.

Jamie Davidson: Thank you for having me.

Isaac: Let's jump right into it, Jamie.

First question for you is, can you tell us what a customer success platform is? And you know how it's different from something like a CRM?

'Cause every company probably does have a customer success platform, but they might not know it.

Jamie: I would assume or hope at the very least, everyone's using something to solve the problem that a customer success platform does.

I mean, customer success platform at its core, you know, organizes your disparate customer data into one place, so that the team that is working with customers, typically your customer success team, can understand customers and can proactively engage them and be more strategic with them.

To help them onboard, implement the product, renew and intervene if at all necessary during the customer journey.

Isaac: So in other words, what would happen if a company like, you know, let's say a SaaS company didn't have a customer success platform?

Jamie: Typically, you know, it's offloaded to a support workflow where it's on the customer to reach out that they have a problem or that they need something.

And that is, I think, fine at certain scales, perhaps in certain products, but the more complicated your product gets, the more that a customer invests in it financially or from a resource perspective, the more that they're certainly, I think, looking for a bit more guidance to ensure that they get value essentially from what they pay for.

And that support workflow where you're relying on the customer to tell you that they're experiencing a problem, that they're potentially going to churn, doesn't help you get ahead of what typically customers is trying to get ahead of, which is churn.

They're trying to prevent the customer from canceling the subscription, and ideally they're trying to grow and increase, you know, your retention and your expansion through, again, that understanding of data and the workflows that they're building in the customer success platform.

Isaac: Fantastic. Overall though, I want to ask, you know, how have you seen this space change over the past few years?

'Cause I know you guys started as something a little bit different, something like a customer data layer.

What made you make that change and how have you seen this space change?

Jamie: Yeah, the products evolved a good bit, 'cause we've been around, I mean officially I guess eight years, but not in market for probably more than like five I would say.

But originally the problem that I was really interested in was, giving customer success teams a better understanding of how the product was being used by customers.

I thought it was a pretty clearing weakness with the market at the time. A lot of the players in the space weren't really doing much with helping customer success teams truly understand, like, product usage and that the customer's actually seeing ROI and where they were struggling.

And I'm a former CTO who kind of accidentally and organically stumbled into customer success, and so product usage and analytics was part of my history and my training.

And I saw a good opportunity to kind of help customer assess teams better understand how customers were using that product.

I learned that was a real problem, but I learned pretty quickly in the journey that we had to evolve into more of a workflow tool.

Like the data filled in a piece of that puzzle, but teams wanted the puzzle. They wanted to be able to act on that data, to execute workflows, to send emails.

An integration strategy was not going to work for us to kind of offload that to other folks. We had to bring those workflows into our product, which is when we then decided, yeah, we need to evolve into a full customer success platform backed by still really rich product analytics, which is still fundamental and foundational to our product, but we had to bring those workflows in into our product to really solve the problem.

Isaac: Wait, so rewind there a little bit. You said you were a CTO before and then I know that you ended up becoming, you know, chief customer officer of that company. That is a really unusual leap.

Jamie: Yeah.

Isaac: Tell me how that happened.

Jamie: I did it to myself. I mean we were a pretty small company. I think whenever I made the transition, we were like maybe 25 employees.

We work with big, big enterprises, so like Qualcomm, T-Mobile, HPV, so they're all like customers of ours, we were like as enterprise as it gets and we just kind of got ourselves in a bit of an interesting situation where we needed somebody with a better foundation, like understanding of the business and the product to really engage and like work with our customers.

And so I thought it was an interesting kind of opportunity. I had really close, like, coworkers with me that were like strong engineering leaders as well.

And so I saw a really good opportunity to kind of like pass the torch, solve a new problem. And so I was like, you know, I'll move into customer success and start to understand our customers and understand how perhaps I can provide a more strategic arm of our business too.

Like these big, big enterprises that were paying us, you know, always for the most part six figures plus to use our products and, you know, well I am, yeah, I've been a CTO a couple times, but at its core I'm a problem solver and to me it was just a new problem to solve and thought it was an interesting opportunity and that introduced me to the world of customer assess platforms and here I am now almost like nine years later.

Christine: That's super cool. As someone who also came from the former CTO track.

Obviously I started Nylas and I was CTO for a long time and today I'm the CEO, which is a wild thing for me to think about sometimes.

But I think that engineering gives you like a really great perspective to come into a lot of other different technology functions, especially since you have like a very sort of like, systems mindset towards things and really kind of breaking part problems, and debugging them, and being kind of structured in the way you kind of tackle solutions, I think it's a superpower in many ways.

Jamie: Yeah.

The best engineers are really good problem solvers. And so if you present a technical problem, or yeah, business problem or anything else, if you can break down the components of it and come up with a good solution, you can solve a lot of things outside of the technical challenges.

Chrstine: Yeah. And I think that like, you know, you see this within kind of the engineering function as well of like people going from like IC's to managers.

You have to like learn to apply your debugging skill to inputs that seem a lot fuzzier than what you're used to working to.

And then from my point of view, just like going beyond that, it's kind of just like an extension of that path of like, well it's a different set of axioms and different types of problems, but you can still kind of use the same approach at the end of the day.

Jamie: For sure.

Christine: I'd love to just go back a little bit to, almost like the architecture of the product that you were talking about a little bit before.

You know, Isaac mentioned that you started out as this sort of like data layer. How did that kind of like fit in people's tech stack at the time?

Was it, so you were kind of like, this source of truth collecting some set of data alongside a CRM. How did people interact with the product in that form?

Jamie: Yeah, like it originally was purely, actually integrated just with the product. We didn't even actually integrate with the CRM, this was before pre-revenue.

This was like us trying to figure out like, you know, what product market fit was going to look like for us.

Christine: Okay. So kind of early iterations and not necessarily something that was like used in production by customers?

Jamie: We did have like, you know, beta users using it and you know, giving us feedback.

And so we definitely had like, I think it was a few dozen like customers using it and some that were, you know, actually using it quite consistently and really liked it.

But others, yeah, wanting to be able to, you know, implement and execute on the workflows that the data was educating.

'Cause that was the, again, the missing piece. It's like okay, we tell you that, you know, this really complex KPI that you need to do and you need to track to understand customer success and the customer's like RR with your product is like bad or good.

Christine: I see.

Jamie: So like, and we didn't really connect that. So we rather than offloading it to like Notion or like the CRM, we had to bring those into the-

Christine: So the first iteration was kind of focused on insights or just like measuring things.

Jamie: It was more analytics, like a data provider, more than anything else.

Christine: And then it sounds like the takeaway was that, like the analytics are not that useful without action.

Jamie: That's the unique thing about customer success is that everything that they do, whether they're sending an email, whether they're working on like some new internal process to like build their health scores, or redefine their onboarding like workflow, it's all educated by customer data.

It's all educated by like, you know, the history of customers, what's worked, what hasn't and whatnot.

And yeah, it was a very, very quick and frankly easy lesson to learn by talking to customers to be like, okay, we know the starting problem that we can solve, but it's not the only problem that we need to solve.

We have to continue along and now learn from the CRMs and from the customer success platform at the time, what they're doing to actually help educate that next step and then bring that next step into our product.

Christine: Was it a challenge to get people to really kind of commit to buying another separate tool alongside their primary CRM in the early days?

Like I feel like it's pretty standard now, but definitely wasn't.

Jamie: We didn't charge anybody. We didn't even have pricing. It was just like, hey, use this.

We actually went through an incubator or accelerator, I always get those two mixed up but, called Techstars, which is a big one, you know, and here in New York.

And so we worked with a Techstars network to like get early feedback and whatnot. But pricing came I think a year after that and actually was organic in and of itself.

We had customers reaching out and being like, hey, you know, we want to use this now. Like, what do we pay you? And we had to figure that out.

Christine: You'd definitely be a little scary to start depending on a tool for vital workflows that is free.

You start to wonder like, how long are you guys going to be able to sustain just giving this to us?

Jamie: Yeah.

I get asked the question of like, you know, if I could do this all over, what would I change? Like what was like one mistake, I think we made along the way, I would actually say charging customers was like a mistake.

We should have done it later. Like our hand was forced, in some ways, which is a good problem to have your customer's wanting to pay you, 'cause they want to know you're going to be around, they want to know they can make that investment.

But like, you know, I still wanted to solve the problem in a little bit more of a different way.

I wanted to be more confident in what we were bringing to market and like our product market fit.

And so I actually wish we had been like, you can use it for free right now, you know, don't worry about it.

We're not going anywhere. We're not going to charge though right now. It's a later problem for us.

And yeah, I wish we had probably given it another six or nine months before we had actually started pulling that trigger on revenue generation.

Isaac: That was probably a pretty hard thing to sell to the team, to the investors that you're not going to charge anything. How'd you get everyone bought into that?

Jamie: Oh, I mean even if you're not generating revenue in your early days, like there's a metric that is indication of like your potential to generate revenue, right?

Like Figma is a great story. Where Figma didn't generate revenue for like a long time and then they started generating revenue, I think it was like literally like five years ago.

And they were, you know, within a year or two, or at like some insane number.

You know, the revenue generation, at least from an investor's perspective, at least as somebody that of course has raised venture funding.

Like there's a different perception for a company that is revenue generating versus one that is specifically and purposely not revenue generating.

Like the moment you start generating revenue, that's a metric that you are held accountable for and like you must report on.

And so to directly answer your question, it was more, you know, while we're not looking to generate revenue yet, like we're looking to make sure we go to market with the product we're most confident with and we're going to measure ourselves from trials and like, you know, just the funnel in general.

The flip side of that narrative was, I didn't want to start the clock on, you know, being held accountable to revenue generation until I was confident that that number was going to shoot up into the right consistently.

Christine: Honestly that's pretty strategic and kind of smart in retrospect because I think when you don't have revenue numbers, you get to really kind of drive the narrative with investors and then when you get to a point where you have numbers like, you know, these people typically coming from a finance background are like, ay, I understand this and this is what it should look like at this point and this is what will make me happy.

And it can be harder to align that with the kind of the evolution that you see of kind of how your product is maturing.

Jamie: I even tried to like, you know, to our early investors again, like the first, I dunno, a couple dozen customers organically kind of paid and I would try to communicate that like, you know, while we are engineering revenue, it is not a number that I think we need to fixate on right now.

Like that wouldn't land. Like the moment you have revenue, the moment it's like, well what's a now, and what's a now, like what's this growth rate and whatnot?

And it's like, it's kind of interesting to play it back. It's like, well if I just said no, I don't want to take your money, it's interesting that that changes the optics, to like the investors and like they don't ask those questions anymore, but the moment you accept one person's money, it just changes the narrative and it changes what you're held accountable for.

Christine: But you had a clear view of what was the level of value you wanted to see people getting from the product before they started paying you.

What was that? How did you decide?

Jamie: We were entering a market that was already, you know, around, right? There was already competitors.

We weren't the first in market, and so for me it was, am I confident that we can repeatedly beat the competition?

And so that was more the indication. I, you know, of course we measure that through our customer success.

Like is the entire customer system using the product, are they relying on other tools to get other parts of their job done and not.

So we of course measured that, but it was somewhat of an external measurement against like what the other players in the space were doing and trying to understand like, is the competitive differentiation here obvious defensible and like, you know, can we actually market it and like draw in like repetitive business?

And so it was kind of a little bit more of a hard thing to quantify at the time for us, it was more or less just continuously looking at the market and ensuring that we were confident that, you know, no matter who was going to be in the deal that we could beat them.

Christine: So using kind of like a product management style simulation, you know, looking at sort of the competitors and your differentiation, because like you weren't in deals right now.

You couldn't measure like how often do we win and how often do we lose?

Jamie: Yeah, I mean well we were working with customers who had experience with the market, right?

And like they were unhappy with it and so yeah there was part of it trying to figure out what was that.

I don't like the term MVP, actually my designer calls it minimal lovable product, which I like actually better, like MLP. Minimal viable product makes it sound like it's too bare bones.

But we were trying to figure out like what the MVP was to go to market with that also had the differentiation at the time, the differentiation was the product analytics and really the also ease of integration.

It's historically plagued our market with like just it taking too long to get the product implemented.

And so we strategically invested in like easier to implement integrations from day one paired with product analytics.

And then it was, so the differentiation paired with like what's the MVP version of what people who have experienced with their market expect, what's that intersection where we actually will now somewhat, you know, beat the competition on a fairly regular basis.

Christine: I think it's really cool that, you kind of like demonstrated a lot of value to customers, such that they were organically wanting to pay you and that's sort of how you bootstrapped the revenue generation at Vitally.

How long did you feel like you could kind of stay in this pre-revenue state? Did you have to like go out and sort of structure your financing around that?

Jamie: We raised a bit without any revenue, yeah, for sure. I mean the clock was on.

Yeah, like we, we didn't raise a huge round by any means. I would say that out of all the skills I have as a CEO, fundraising is middle of the pack.

I'm not terrible at it about great at it, but I wasn't the type of CEO that was going to go raise, you know, many millions of dollars just on like a vision and with no revenue.

Like I'm an engineer, I always want to proof that what I'm saying is true, that I can actually execute on what I'm saying.

And so we raised, I think it was like a few hundred thousand or something like that coming out of Techstars without any revenue and that was enough to get what was a very limited team at the time.

Myself, my co-founder and one other engineer who had actually worked with me at my past company enough to continue to along that journey methodically, carefully even making sure we invested heavily in the product, right?

But the clock was on, we didn't have infinite amount of time, we had to do it within, you know, I think it was roughly like a year before the clock got close to midnight essentially.

Isaac: Now I remember reading, and correct me if I'm wrong here, that you landed some awesome logos really early on, right? Like a Segment, Zapier, Productboard.

It really actually some of the best products out there, like actual industry benchmark products and companies.

At what point in your story did that happen and how did you make that happen? I mean that's so rare.

Jamie: It was organic for all of them actually.

Segment was the first of the ones you mentioned and I think Segment was like maybe like our 10th customer and they actually reached out to us for a totally different reason.

They were launching a new integration marketplace and to my surprise, I had did not know this.

They said that we were one of like the top five most requested integrations that Segment did not have.

And so they were going to launch this new integration marketplace and they wanted to choose five new integration partners to build the integration with and launch with.

And so they went through all the most requested and we were one of those. And in that process, the CTO got familiar with our product and was impressed by it, and was impressed with how easy it was to implement.

And they had some not so great experience with the market themselves and he had forwarded it to their customer success team to be like, hey, these people really seem to know what they're doing.

I'm impressed with the tech, you should check it out. And so then all of a sudden, now we had two conversations going, oh, where's Segment.

It was how do we launch this integration and how do we actually become your official customer success platform? We were actually able to do both. That kicked off a whole, a new world for us.

Actually the integration launch in Segment was probably one of the most like critical points for our success, because they put a lot of their marketing firepower behind it to make sure that it was success.

They promoted, you know, here's the five partners that we partner with. We were one of them, we were front and center in a lot of their marketing materials in their integration marketplace that led to I think the Zapiers and the Productboards, but it was all organic.

That was done with like no salespeople. Inbound requests, to see what the product was capable of.

Eventually, of course it led to a full-fledged sales team, but I had to put on the sales hat and try to get those deals over the finish line myself in those days.

Isaac: Well when those kind of leads come in organically, I think people consider that the equivalent of just receiving a check in the mail randomly or winning the lottery. But there's something behind that, right?

You said Segment was your 10th customer and at the same time you were their most requested integration that logically tells me that all of your existing customers were demanding integrations from Segment.

Jamie: Oh yeah, yeah.

Isaac: It really just appears that you focused so much on value. I always think that the sign of good product thinking is what's technically called long-term greediness, right?

Where instead of being short term gradient and trying to take everything you can in order to build the value long term, you have to focus on value and helping others.

Not everyone can get in that mindset and very few people can stay in that mindset for a period of time as the clock is ticking on runway.

How would you recommend people, both get and stay in that mindset and hold the line as their team, or their investors are demanding that they start to collect as much money as possible?

Jamie: I mean, it's just business, right?

Like if you're going to be successful, you're going to attract competition. The competition is going to build off of you, you know? Everybody builds off of somebody, right? Like something iterates, something leads to something else, leads to something else. And so the best businesses are the ones that iterate so much that the competition can't keep up.

That, you know, you are iterating and then you don't rest easy and then you iterate again, and then you don't rest easy, and you iterate again.

And so what I would say is like you want to make sure that when you go to market, of course you all want to be successful, but if you are, you're going to draw on competition who's going to have the benefit of being able to look at what you've done, reverse engineer it and figure out how to do it better.

And so you want to make sure there's some defensibility there of what you've done so far when you bring it to market, so that the moment that you actually have some success, you aren't vulnerable to like that immediately going away by a competitor who just does it better.

And again, you just got to iterate and continue iterating, because the competition will learn from you. They'll figure out how you did something, and you'll figure out how to do it maybe slightly better because it's always easier to solve the problem the second time and not the first time.

And you just got to keep your formula, the guess that way.

Isaac: That's absolutely incredible. And I think that there's a lot of hustle culture in startups, right?

Like the word iterate is used a lot. And I think what often people hear is like, do everything. And if you can think of another thing, just start doing that too.

How did you, you know, get your team to choose the right things to iterate on, because finding product market fit is more than just putting the time in, it's putting in the focus as well.

How did you drive that focus? How did you achieve product market fit truly?

Jamie: It happened with a different version of the platform that actually even exists today.

And so the original like product market fit was found with what I would say is a more traditional take on like the customer success platform.

Sure, we differentiated with like product analytics and ease of integration, and with like UI and UX and things like that. But it was an opinionated platform.

And by that I mean like, we sat down and tried to understand what truly does like the CSM need to see, or the VB customer says need to see.

So we would present our data and our workflows like, here are your customers, here's your turn, here's your revenue dashboard, here's your renewal dashboard.

And we would have to like one off, again, we'd be presented with a problem and we would have to solve the problem.

That found product market fit in the traditional customers as platform landscape. But again, iterations I didn't rest easy.

And in fact like the motivation was a series B raise in, was it 2025? So in 2022, Fincher like basically shut down.

Like, you know, it was hard, it was almost impossible to raise a growth round and we were needing or wanting to raise a series B.

I was at the time, like again, we were doing well and generating good revenue, but I was like, I don't know if I can go raise a growth run in this market by just being like, hey, we're a customer success platform and we're better with integrations like product analytics and these things.

And so I wanted to find a new narrative, a new vision, which would help my team think about the product differently. And that's what you see now. We instead pivoted away from like, how do we solve the problems for our customers to more of like a building block, like a toolkit kind of methodology.

Like almost like a Notion, or like a ClickUp kind of methodology.

Like how do we actually though, instead of trying to figure out how we solve the problems, how do we reverse engineer what we've done, create the building blocks that our customers can put together to solve not only problems in their ways, but solve problems maybe they're not even thinking about.

So we went with a more of like a flexible, almost like low code, like no code kind of approach to our product with a thought being, it's going to be heavily focused on flexible project management, flexible reporting, flexible automation, just all powered by your customer data.

And that narrative of like your data and then the work you need to get done on top of that data is of course important to customer success, but it's not just customer success that can influence support and sales, and even product teams and things like that.

And so we wanted to branch out and come more of like a cross-functional, like collaboration platform just with your customer data and workflows at its core.

That's kind of what we raised that series B narrative on. We did it in the fall of 2022, got the series B done and then completed that product.

And that's our vision now. It's like less so of like, I don't know if you'll ever, in Vitally log in and day one, see like, here are your renewals.

It's not a problem that we're really focused on solving anymore. You can do it of course, but we're not going to sit down and say that's the first thing you should see.

We're going to suggest some things, but it's going to be up to you and you're going to have to control it to kind of change it yourself.

Isaac: Gimme some examples of like the first things someone's going to see.

Jamie: I mean it varies per customer.

Like an executive can have a dashboard that just showcases the growth of their business, churn per segment, and whatnot as the first thing a VP of customer success can log in and understand what's the work that my team is doing?

How much work is open per CSM and what's the NRR per CSM? They can choose to see that.

Of course the CSM has a different journey though. They don't need to necessarily see, like that kind of high level dashboard.

Maybe they're only focused on like, what are my customer meetings and what are the work I need to get done today?

And so we want the entire business to benefit from our product, because the data that we have is, I would argue, the most valuable data to the business. It is almost everything the CRM has. It is how customers are using your product. It is an understanding of revenue, and how it grows, and when it contracts and why it contracts, and when customers turn and why they churn. That is data that is not just beneficial to customer success, it's beneficial to everyone.

And our mission now is to how to like unlock that data and unlock the workflows on top of that data, so that it can better not just customer success, but better the entirety of the business.

And that requires that more flexible approach that we now take today.

Isaac: I want to know how you operationalize that, like internally with your employees, with your engineers, with your product people, all of that.

I know that you guys shifted over from like QBRs to, where they value reviews, right?

Jamie: Yeah, it was, I gave a talk on this. Yeah, yeah.

Isaac: Yeah. I mean I'd love to hear about that and if you think that's part of what drives it, or your culture, or just where you think the secret sauce is.

Jamie: Well that is an example of something that we came up with as an example of like perhaps there are better ways to leverage the data that we have and and build better processes.

Customer success, it's the newest function in an organization, right? Like it existed after marketing, it existed after sales, it existed after product, it existed after all these things.

And so because of that, it's not gone through a lot of iterations. It was basically invented in the late 2000s, early 2010s, only became ubiquitous in the early 2010s.

That was a high touch white glove, very kind of wine and dine approach to customer success. in the middle 2010s, the late 2010s scaled and tech touch customer success sort of become a thing.

And so one of our, actually our top principles as a company is question convention. Like, just because something is conventional, is the way it's done, doesn't mean it's the way it should be done.

And so MVR was an example of that. Like we are all talking about QBRs, everybody knows what a QBR is, but it today feels like it's a little old fashioned.

It requires the customer to sit in a meeting for an hour and a half, listen to the most polished version of the story that the vendor wants to tell them because the vendor wants, you know, typically QBRs are almost like late minute like decisions to try to get a signal if their customer's going to renew or not.

Like actually I'll schedule a QBR to see if there's any trouble here and that's not the most valuable use of the customer's time.

So the MVR was kind of like, almost a needle in that regard. It's like maybe we don't need to do it this way.

Maybe if we had a better approach to how to like solve the problem of just delivering value to the customer, because that's what the QBR is supposed to do.

It's supposed to communicate value. Can we do that in a more efficient and collaborative way?

Do we really need to get the customer in a Zoom, have everybody sit down for an hour and a half of time?

Or can we just showcase to them value on a recurring basis, on a quicker timeframe?

Let them consume that, comment where they need to, slack some questions over, something like that and free that, you know, hour and a half block that the customer has to do for the QBR up for something else.

And that's applied to our product. Is that like, just because our market has historically taken that opinionated approach of like, how do I show the best renewals dashboard?

How do I implement the best like risk process? Doesn't mean that's the way that we have to do it.

We can learn from what I think the work platforms have done, the Notions, the ClickUps, the Asanas of the world, who kind of went in the low, the no-code, kind of like route of like it's a platform that can flexibly solve a whole bunch of problems.

Can we learn from that and kind of head a little bit more in that direction and less so in the like opinionated sit down, solve all these specific problems ourselves kind of direction.

And so that question commission was kind of core to getting people thinking in the new direction, to the MVR thing and a lot of things that we do.

Christine: That's really interesting to hear in terms of just like, how to iterate on driving more value to customers as you have these sort of longer term relations.

And you know, when I think about the history of customer success, you know, it's really something that like, has evolved hand in hand with B2B SaaS.

You know, like, if you're just selling somebody a packaged CD with software on it, like, you don't have customer success, you just have support.

But now that over the past 15 years, you know, software has been really delivered primarily through the cloud and through these services like, you know, we essentially have developed these long-term relationships with people and you want to kind of partner with them and grow over time because you're constantly able to iterate and ship new features and sort of work together to drive mutual success.

I think it's really interesting to kind of look at what's happening all around us and think a little bit about, kind of where things are going in general.

Like the 2010s were kind of like the era of SaaS, but the 2020s are looking to be the era of AI and certainly beyond that.

How are you thinking about how AI is something that can help drive value for Vitally's customers?

It sounds like you have a, you know, a vision around how, at least in Vitally's case, you sort of started from a place of more specificity on sort of defining workflows and bringing best practices in terms of workflows in this area of customer success to people.

And now sort of what you're seeing is a demand to actually expand from there and to become more flexible. I think there's another piece sort of under this, around like what data lives where and how do you make use of that data?

And then there's this whole kind of other elephant in the room around like, does B2B SaaS even exist in five years or 10 years?

Is it all going to be agents? Are we just going to be talking to computers? Where is it going? So gimme your hot takes on AI.

Jamie: Where to start?

At our core, we do two things. We organize your customer data, so that you can understand it, and then we help you get work done on top of it.

And we can and will apply AI to both areas. Like we can apply it to help you better understand your customers to enrich your data at scale, to showcase to you, like especially in customer success, we deal with a lot of raw text, right?

Meeting transcripts, customer conversations and emails, slack messages. Buried in those are critical insights.

Like if a customer writes in, or like in a meeting says, I need to see feature X and Y, otherwise I might turn to competitor Z.

Like in that are structured in very critical data points that are not going to do any good unless somebody currently manually goes and like tracks those somewhere and like writes it down, right?

But like the world of AI is good at extracting those kind of insights out. And so we can apply AI to help customer assist teams get that data at scale and then of course apply it to the workflows, like make them more productive, make them more efficient, help them actually, you know, write the best version of like a doc, or an email that is being sent to the customer.

And so I think AI's great and will be great for that. I think the biggest challenge in the world of today's AI is...

You got to crawl, walk, run, maybe we're out of the crawl stage right now, but if we're barely there with AI, like we're heading into walk territory and my issue with a lot of what's happening right now is especially like the venture narrative, is like just trying to drive everybody to run.

And by that I mean like, they think AI's going to solve all the problems. I think what that does is, those overlooks human behavior and tendencies.

We've been trained to use software in very particular ways for the last 20 years. We don't just immediately overnight change it because AI's getting shoved down our throats.

And so I think the challenge with AI today is, how do we bake it in the best way possible?

It's like a UI/UX challenge, rather than, and especially for us, like rather than positioning the problem is, let AI do the job for you.

It is suggestions not, you know, actual execution. And so how can we actually suggest and organically bake AI in, into all of these things while not necessarily trying to train the user to do exactly what it's saying, or to trust exactly what it's saying.

Especially in customer success, like with the data that we're getting from the CRM, from product, from, you know, revenue tools, from customer communications, from survey responses, from all this.

Like, it's just so much that is hard to say that like, you know, AI's going to get it all right all the time and we have to do a human led approach with AI for customer success, so that it age what customer success teams are doing and not try to today, at least in today's world, look at it as a full replacement.

I think that narrative is going to continue for many years. I think that though the narrative of AI replacing and changing everything in the near term is not quite as near term as we all think, but I might be wrong, who knows.

Christine: It seems like one thing that you're getting at is that, the user interface with SaaS software today tends to be like very optimised and sort of straightforward for specific workflows and it points people in certain directions.

Whereas at least if you're just sort of talking to a text interface or a voice interface, it's like very, very flexible.

How do you think we might like constrain that over time to like retain some of the benefits of sort of pointing people to specific workflows, but still have some of that flexibility that comes from sort of new paradigms.

Jamie: It's on us as the product people. I don't know, we're trying to figure it out ourselves.

Like we're literally working on this problem, like right now of how do we have the AI sort of like suggest things and help complete that next step without overly relying on the next step.

It's especially critical in customer success. I think, like you've seen support tools be revolutionized by AI already and it makes a ton of sense because, and I'm going to say something that's maybe a little bit loaded, but like people expect kind of a bad support experience.

Like the bar for a support experience is lower than the bar for a customer success experience.

And so if you go to, you know, a website to try to talk to AI, we're already, when we've been trained over the past, even like decade plus to talk to a bot for a bit before we actually get to a human.

But a lot of us also try to shortcut that like give me a human, customer success that borrow is much higher.

Like, you know, the first impression in the customer experience is everything. And so that's the thing I would want all of us in customer success to ask ourselves is like, what do we hear from customers?

Do we hear customers say that they want, you know, customer success replaced by AI?

I'm not hearing that, they specifically want someone that they know understands them as guiding them along.

And so like a good example is, you know, AI writing you an email, I've seen it already where it actually degrades the experience because the user trust the AI, they want to get the job done, they want to move on to the next thing.

And so they don't, you know, they don't edit what the AI did correctly and so they end up sending the customer something to boilerplate, or something that has duplicate content and that can cascade in customer success.

Where in support that can be shrugged off and you can be like, ah, stupid AI bot. And customers is like, I paid for a human to, you know, help me here and this is clearly not a human. What are you doing?

And that's an interesting challenge for us in customer success is how to organically bake in AI without removing that human element from the equation.

Christine: Yeah, I guess the entire point of customer success is to have a human driven tailored approach to customers that is very adaptable and I think it's a really good insight that you have to kind of look at like, what is the bar for expectations in any given area when you sort of think about like, will AI make this better?

And in some places that's going to be a yes and in some places that's going to be definitely not today, but things are changing fast.

Jamie: Yeah, they are.

Christine: We'll see where we're at in a year.

Jamie: Yeah,

Christine: Cool. With that, I think it's a good time for us to jump into our pick section.

So this is a fun thing we like to do where everyone shares something that is capturing their attention.

You know, it could be a tool, it can be a media thing, a book, a show, really anything that you're just excited about right now.

So Isaac, you want to kick things off? Do you have a pick for this week?

Isaac: Yeah, I do.

There are these things that I call intellectual primals, which are, you know, maybe articles, or books, or things that you come back to from time to time, that like truly shape the way you think about reality.

And one of my intellectual primals is a, it's an article called, "You Can't Reach the Brain Through the Ears" and it's without a doubt my favorite article I've ever read of all time.

It's a blog post by this guy who does a podcast called "Experimental History."

And man, I just keep coming back to it like every six months or something and rereading it and reminding me that, when you're trying to convince people of something that, you know, you can't just through speech take these very complex ideas, these you know, symphony ideas that are happening in your brain and turn them into this like single track language and put it in their head.

And it's helped me so much think about communication, because you know, you're thinking of something complex and you're saying it and they're hearing something else, they're hearing, you know, just a very reduced resolution version of it.

And it's really just helped me with introducing people to, you know, complex ideas.

You know, for example, like if I'm working with someone internally and trying to get them to, you know, understand what, you know, AI agents can do for them, right?

If they were really working and building those things, what they hear is very different than what I think until I really fully explain it to them.

Would highly recommend Googling and reading that article, "You Can't Reach The Brain Through The Ears," my favorite of all time.

Christine: Cool. I'm Googling it right now. What about you Jamie? Do you got to pick for this week?

Jamie: So again, I've been a CTO of a company, two different companies actually in the past and I actually still try to code a little bit today just to kind of keep up and recently started to try to figure out how to bake AI into my coding habits, exploring Cursor, the AI code editor.

And I've heard phenomenal things. I haven't fully been able to like bake it into what I'm doing yet, which is the challenge with the AI workflows.

How do I like, you know, start to learn a different way of approaching something that I've been doing a particular way for almost 20 years.

But I'm excited to evolve my approach to that and see if I can have greater impact through, yeah, through some AI code editor, I'll give Cursor a shout out, but not an endorsement yet, I haven't used it enough.

Isaac: That's so cool. I'd agree. You know, everyone should be trying it if you're a coder and it's, you're talking about AI writing emails for you and you're like, God, like it's just writing the, you know, AI slop, right? Generic stuff.

And man, Cursor really reads your mind and it feels like it knows what you're trying to do, which is so rare.

Jamie: Yeah, it's like, yeah, if you can solve minutes, or save minutes or hours, I'll accept more variability on like the output, on like maybe this isn't exactly what I want to do.

I guess I got to do it manually. The email examples I guess saving seconds. I don't want the variability.

I want to know that the output is there and it's going to work the way that I want it to.

I'll take the a hundred percent accuracy of me doing it myself over the safe seconds, but the minutes or the hours or even the day is an interesting area where, yeah, I'm excited to see how it can go.

Christine: So I have a pick for this week and it's a book and it is, it's kind of a classic book I would say, in that it's from 2015, which was 10 years ago, but it's called the "Sales Acceleration Formula" by this guy Mark Roberge, who was the guy who built HubSpot's early sales team, or really kind of like the first few years, you know, that gross super fast.

And the reason this book is my pick is because I mentioned actually earlier that I also used to be the CTO and now I'm the CEO, and for the longest of time, you know, I had a business partner whose focus was on sales and marketing and I just sort of like compartmentalize that off of my brain.

It was like, this is like somebody else's main job and I don't need to really like rock it too much.

But being sort of thrown into the space where that's not the case anymore, I really felt like I needed to just like get a deeper understanding of sales and marketing and Mark Roberge has an engineering background and I found it really helpful just in that, you know, he sort of lays them out in a very sort of systematic way and explains the big picture in a way that just like really made sense to me.

And I found it really helpful just to kind of have that sort of foundation and, you know, some of it's a little bit dated, but it's also really cool to see that, you know, for one of these examples in kind of the technology section in the back of here is around email logging, which apparently, I guess they built some internal automation for that with HubSpot back in the day, you know, 10 years ago.

And it's really interesting from my perspective in that like, you know, starting in kind of the 2017 era, a lot of people started using Nylas to build that kind of email logging for CRM type apps.

And really today it's kind of just like, something that you assume you're going to have in any app like this because the example it brings up is that, you know, like sales technology was built for execs or sales leaders, not for the ICs, and really just creates work for the ICs.

And you know, this was 2015 and so kind of the leading edge back in that era was to, you know, build some automation tech that automatically did, you know, the logging of those messages in the CRM.

And I think it's super cool to just to see how like, the table stakes of things has evolved and yeah, it was just funny to read it in the back of this book when I've seen it sort of happen live.

Isaac: Awesome. Jamie Davidson, CEO of Vitally, the best customer success platform out there.

Thank you for coming on the show. Thanks for your time.

And if anyone out there wants to have the best customer success team out there and one that's operating off of the data that's happening from the product usage, and the business, and all that and, you know, getting those renewals and making customers happy, then definitely give Vitally a shot.

Christine: Yeah, we here at Nylas are a happy customer as well, so appreciate the support from you on that point, Jamie.

And thanks for taking the time to come on the show. It's been awesome to have you.

Jamie: I appreciate you all having me. This was a lot of fun.